Discussion:
Seeking ideas how to sync an AC generator to 60 HZ average.
(too old to reply)
Don Stanley
2010-11-23 05:11:49 UTC
Permalink
Hi All;
My next project is a remote off grid 60 HZ power unit.
I am thinking of a EMC2 PID to control the RPM.
I am also expecting maybe a 1-2 HZ momentary shift as the
big power loads come on and off line.

I am looking for a method to get a reliable reference that can be used to
average 60 HZ through the power surges and correct a local timer drift
for long term accuracy. A simple WWV corrected timer
(a Wall mart Atomic clock with outputs).
Anyone know of such a device?

Thanks
Don
John Kasunich
2010-11-23 05:17:49 UTC
Permalink
Do you have internet access at the site?

If you are running an PC for EMC, you can use NTP to keep
the PC's clock synced to the rest of the world, and some HAL
bits and pieces to keep the generator synced to the PC
clock. EMC's encoder component could easily count 60Hz.
Post by Don Stanley
Hi All;
My next project is a remote off grid 60 HZ power unit.
I am thinking of a EMC2 PID to control the RPM.
I am also expecting maybe a 1-2 HZ momentary shift as the
big power loads come on and off line.
I am looking for a method to get a reliable reference that can be used to
average 60 HZ through the power surges and correct a local timer drift
for long term accuracy. A simple WWV corrected timer
(a Wall mart Atomic clock with outputs).
Anyone know of such a device?
Thanks
Don
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for
grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
--
John Kasunich
***@fastmail.fm
Andy Pugh
2010-11-23 10:16:32 UTC
Permalink
Post by John Kasunich
Do you have internet access at the site?
If you are running an PC for EMC, you can use NTP to keep
the PC's clock synced to the rest of the world
Doesn't it need to be synched to mains frequency rather than absolute
time? How accurate is NTP (or even GPS time) in this context?
...
<google> http://gpsinformation.net/main/gpstime.htm seems to think
that is it close enough.

However, what is really needed is to keep synch with the local mains
during outages, and that might be better done by a local clock. You
could potentially use siggen with a PID to correct phase, then lock
the PID when mains voltage is lost. PID dosn't have a "lock" pin, so
you would probably need to use a sample_hold.

When the mains comes back you could use the PID to bring the generator
back in to phase before switching back over. (and it probably wouldn't
drift that much anyway)
--
atp
John Kasunich
2010-11-23 14:06:46 UTC
Permalink
Post by Andy Pugh
Doesn't it need to be synched to mains frequency rather than absolute
time?
(snip)
Post by Andy Pugh
However, what is really needed is to keep synch with the local mains
during outages,
(snip)
Post by Andy Pugh
When the mains comes back you could use the PID to bring the generator
back in to phase before switching back over.
My next project is a remote off grid 60 HZ power unit.
I think this has nothing to do with transitions to/from utility
power.

John
--
John Kasunich
***@fastmail.fm
Andy Pugh
2010-11-23 14:23:18 UTC
Permalink
Post by John Kasunich
Post by Don Stanley
My next project is a remote off grid 60 HZ power unit.
I think this has nothing to do with transitions to/from utility
power.
Ah, in that case, why does it matter?
--
atp
sam sokolik
2010-11-23 14:31:52 UTC
Permalink
Sounds like mainly he want his old 60cycle clocks to work correctly. :)

sam
Post by Andy Pugh
Post by John Kasunich
Post by Don Stanley
My next project is a remote off grid 60 HZ power unit.
I think this has nothing to do with transitions to/from utility
power.
Ah, in that case, why does it matter?
Dave
2010-11-23 14:43:21 UTC
Permalink
Post by Andy Pugh
Post by John Kasunich
Post by Don Stanley
My next project is a remote off grid 60 HZ power unit.
I think this has nothing to do with transitions to/from utility
power.
Ah, in that case, why does it matter?
Not sure. Generally when generators take over after a power outage
they don't try and maintain sync with the utility since the utility has
no power on the lines to synch with. When the power comes back up, the
generator speed is synced to the line and the
transfer is made back to the utility.

I remember back in college that we had a lab experiment in my power
class where we did an actual manual power transfer to and from the
utility lines with a generator. Back then they used 3 light bulbs
connected L1 to L1, L2 to L2, L3 to L3 between the generator and the
utility lines. When all three light bulbs went out the generator was
in synch with the line and you could open and close the transfer
switch. Some wise guy wanted to see what would happen if you threw the
switch when things were out of sync.. so he threw the switch and we all
got to hear the bang when the generator was pulled into sync by the
utility. The professor was not amused. I distinctly remember that
prof as he later tried to throw me out of college after I found he was
using out of date materials and I was forced to confront him with it.
Not my best college memory. Fortunately I prevailed, and the department
chair assigned a different prof as my advisor.

Dave
Ian W. Wright
2010-11-23 14:31:08 UTC
Permalink
John,

I think Andy was assuming that this power generator would be
backup for mains outages when, if the switchover were to be
automatic and seamless, it would indeed to essential to get
the generator frequency very close to mains frequency and,
more importantly, the phases in the correct rotational
relationship before switch back to the mains. It took ages
for the supplier's techs to get this right on the 2 x 1100Hp
deisel gennys I had at work....

If this genny is to completely stand-alone off in the woods
somewhere - I don't really see why it needs tying to any
major time standard - what's going to power a PC while the
genny gets up and running anyway....? Normally engine
governors or feedback electronics should keep the frequency
within limits anyway - hell, I've seen our mains frequency
vary from 47 to 54 Hz over the day many times depending on
the load and no one ever complains.......


Ian
_____________________
Ian W. Wright
Sheffield UK
Andy Pugh
2010-11-23 14:37:59 UTC
Permalink
Post by Ian W. Wright
hell, I've seen our mains frequency
vary from 47 to 54 Hz over the day many times depending on
the load and no one ever complains.......
Though supposedly you should still always get 4320000 cycles per day.
--
atp
Ian W. Wright
2010-11-23 14:49:17 UTC
Permalink
Post by Andy Pugh
Post by Ian W. Wright
hell, I've seen our mains frequency
vary from 47 to 54 Hz over the day many times depending on
the load and no one ever complains.......
Though supposedly you should still always get 4320000 cycles per day.
Yes, you're right but I'm not sure how close they ahere to
that nowadays. With the common frequency dips due to load
problems and the tendency for people to convert to quartz
timing instead of mains synchronous, I wonder if they bother
so much or whether they just try to get it right weekly or
some such. Maybe I'll put a mains frequency monitor on here
out of interest - its 13 years since I retired from the
ratrace of chasing power around plant rooms......

Ian
Roland Jollivet
2010-11-23 15:17:48 UTC
Permalink
I heard once (here in SA) that the mains cycles were adjusted close to
midnight, so that the number remained consistent on a daily basis. I don't
know how it's arranged though, since there are so many power stations to
sync.

Regards
Roland
Post by Ian W. Wright
Post by Andy Pugh
Post by Ian W. Wright
hell, I've seen our mains frequency
vary from 47 to 54 Hz over the day many times depending on
the load and no one ever complains.......
Though supposedly you should still always get 4320000 cycles per day.
Yes, you're right but I'm not sure how close they ahere to
that nowadays. With the common frequency dips due to load
problems and the tendency for people to convert to quartz
timing instead of mains synchronous, I wonder if they bother
so much or whether they just try to get it right weekly or
some such. Maybe I'll put a mains frequency monitor on here
out of interest - its 13 years since I retired from the
ratrace of chasing power around plant rooms......
Ian
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Leslie Newell
2010-11-23 10:36:41 UTC
Permalink
Generally they adjust the frequency throughout the day. As the load on
the system increases the frequency drops. To increase the frequency they
bring more generators on line. If the frequency gets too high they
reduce the generating capacity. The really tricky bit is matching
generating capacity to the load well enough to maintain a reasonably
accurate 50/60Hz

Les
Post by Roland Jollivet
I heard once (here in SA) that the mains cycles were adjusted close to
midnight, so that the number remained consistent on a daily basis. I don't
know how it's arranged though, since there are so many power stations to
sync.
Regards
Roland
Jon Elson
2010-11-23 18:24:47 UTC
Permalink
Post by Leslie Newell
Generally they adjust the frequency throughout the day. As the load on
the system increases the frequency drops. To increase the frequency they
bring more generators on line. If the frequency gets too high they
reduce the generating capacity. The really tricky bit is matching
generating capacity to the load well enough to maintain a reasonably
accurate 50/60Hz
The power stations are self-synched to the overall grid. Opening the
throttle on the
turbines acts to increase frequency, but due to the enormous stiffness
of all the alternators,
many of the throttles have to be opened wider and then the frequency
starts to rise very
slowly. So, it is a VERY tricky and slightly unstable network. You can
measure the
effect of each individual alternator by measuring the phase angle
between current and
voltage, although that gets complicated by the excitation level of the
alternator. Reducing field
excitation causes the alternator to produce current that lags the
voltage, like an induction motor,
and increasing the excitation causes current to lead the voltage.

Experts have tried to model the whole grid, and found it to be very
nearly mathematically intractable.
Resistance and leakage inductance in transformers, etc. are the only
thing keeping the grid (barely)
stable.

Jon
Edward Bernard
2010-11-23 23:11:00 UTC
Permalink
----- Original Message ----
From: Jon Elson <***@pico-systems.com>
To: Enhanced Machine Controller (EMC) <emc-***@lists.sourceforge.net>
Sent: Tue, November 23, 2010 12:24:47 PM
Subject: Re: [Emc-users] Seeking ideas how to sync an AC generator to 60 HZ
average.



Experts have tried to model the whole grid, and found it to be very
nearly mathematically intractable.
Resistance and leakage inductance in transformers, etc. are the only
thing keeping the grid (barely)
stable.

Jon

Does this mean the "smart grid" we have heard so much about is just a pipe
dream?
Jon Elson
2010-11-24 02:23:40 UTC
Permalink
Post by Edward Bernard
Does this mean the "smart grid" we have heard so much about is just a pipe
dream?
Smart grid has absolutely nothing to do with stability of large
electrical grid systems.
Smart grid is mostly related to having individual appliances able to be
turned off at peak
load times, and giving customers a credit for doing that.

There is some worry, I think, about all the non-linear loads that draw
constant power over
a range of voltages. In cases of extreme shortage of generating
capacity, they used to do
"brownouts" where they would reduce voltage, and incandescent and
heating loads would
be reduced. Compact fluorescents, motors on VFDs, computers, etc. all
will draw constant
power during a brownout, and they look a lot less resistive also, so may
be less absorptive
to network instability (less damping).

Jon
dave
2010-11-23 16:25:00 UTC
Permalink
50 some years ago a couple of engineers came to Coulee from the Denver
office (USBR) and installed a mag amp in feed forward mode to control
the penstock gate on one of the small units. This was before the third
powerhouse so the units were all 108 Mw. This one unit was used to swing
the entire Northwest Power Pool ( OR, WA, ID, UT, and most of MT and
WY ). I don't have any quantitative figures but the frequency regulation
got 10X better.

Dave
Post by Roland Jollivet
I heard once (here in SA) that the mains cycles were adjusted close to
midnight, so that the number remained consistent on a daily basis. I don't
know how it's arranged though, since there are so many power stations to
sync.
Regards
Roland
Post by Ian W. Wright
Post by Andy Pugh
Post by Ian W. Wright
hell, I've seen our mains frequency
vary from 47 to 54 Hz over the day many times depending on
the load and no one ever complains.......
Though supposedly you should still always get 4320000 cycles per day.
Yes, you're right but I'm not sure how close they ahere to
that nowadays. With the common frequency dips due to load
problems and the tendency for people to convert to quartz
timing instead of mains synchronous, I wonder if they bother
so much or whether they just try to get it right weekly or
some such. Maybe I'll put a mains frequency monitor on here
out of interest - its 13 years since I retired from the
ratrace of chasing power around plant rooms......
Ian
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Dave
2010-11-23 16:52:16 UTC
Permalink
Interesting.

Apparently there is a large DC inverter station out in the northwest??
that is used to allow different synchronization of the grids in the
eastern US with the western US or across some other regional divide.

Sort of like a giant AC drive taken to an entirely different level.. ;-)

I remember finding it via a google lookup one time.

Dave
Post by dave
50 some years ago a couple of engineers came to Coulee from the Denver
office (USBR) and installed a mag amp in feed forward mode to control
the penstock gate on one of the small units. This was before the third
powerhouse so the units were all 108 Mw. This one unit was used to swing
the entire Northwest Power Pool ( OR, WA, ID, UT, and most of MT and
WY ). I don't have any quantitative figures but the frequency regulation
got 10X better.
Dave
Post by Roland Jollivet
I heard once (here in SA) that the mains cycles were adjusted close to
midnight, so that the number remained consistent on a daily basis. I don't
know how it's arranged though, since there are so many power stations to
sync.
Regards
Roland
Post by Ian W. Wright
Post by Andy Pugh
Post by Ian W. Wright
hell, I've seen our mains frequency
vary from 47 to 54 Hz over the day many times depending on
the load and no one ever complains.......
Though supposedly you should still always get 4320000 cycles per day.
Yes, you're right but I'm not sure how close they ahere to
that nowadays. With the common frequency dips due to load
problems and the tendency for people to convert to quartz
timing instead of mains synchronous, I wonder if they bother
so much or whether they just try to get it right weekly or
some such. Maybe I'll put a mains frequency monitor on here
out of interest - its 13 years since I retired from the
ratrace of chasing power around plant rooms......
Ian
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App& Earn a Chance To Win $500!
Tap into the largest installed PC base& get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App& Earn a Chance To Win $500!
Tap into the largest installed PC base& get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App& Earn a Chance To Win $500!
Tap into the largest installed PC base& get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Andy Pugh
2010-11-23 18:10:42 UTC
Permalink
Post by Dave
Apparently there is a large DC inverter station out in the northwest??
that is used to allow different synchronization of  the grids in the
eastern US with the western US or across some other regional divide.
I think that might be to prevent the US grid being a rather efficient
antenna and radiating half the energy into space.

60Hz x speed of light = 3000 miles.
--
atp
Igor Chudov
2010-11-23 18:14:43 UTC
Permalink
Post by Andy Pugh
Post by Dave
Apparently there is a large DC inverter station out in the northwest??
that is used to allow different synchronization of  the grids in the
eastern US with the western US or across some other regional divide.
I think that might be to prevent the US grid being a rather efficient
antenna and radiating half the energy into space.
60Hz x speed of light = 3000 miles.
But, but, the wires usually go in pairs (or threes for three phase),
would that not cancel out the antenna effects?

i
Przemek Klosowski
2010-11-23 22:06:10 UTC
Permalink
Post by Igor Chudov
Post by Andy Pugh
Post by Dave
Apparently there is a large DC inverter station out in the northwest??
that is used to allow different synchronization of  the grids
I think that might be to prevent the US grid being a rather efficient
antenna and radiating half the energy into space.
60Hz x speed of light = 3000 miles.
But, but, the wires usually go in pairs (or threes for three phase),
would that not cancel out the antenna effects?
Common configuration of HV lines returns through earth, to save the
cost of wire.
http://en.wikipedia.org/wiki/High-voltage_direct_current
dave
2010-11-24 16:00:46 UTC
Permalink
Typical losses in high voltage transmission lines are on the order of 6
to 7% distributed between corona and IR losses depending on the
voltage.

D
Post by Andy Pugh
Post by Dave
Apparently there is a large DC inverter station out in the northwest??
that is used to allow different synchronization of the grids in the
eastern US with the western US or across some other regional divide.
I think that might be to prevent the US grid being a rather efficient
antenna and radiating half the energy into space.
60Hz x speed of light = 3000 miles.
Jon Elson
2010-11-23 18:37:09 UTC
Permalink
Post by Dave
Interesting.
Apparently there is a large DC inverter station out in the northwest??
that is used to allow different synchronization of the grids in the
eastern US with the western US or across some other regional divide.
Sort of like a giant AC drive taken to an entirely different level.. ;-)
My understanding is that most of the time the Eastern and Western grids
are not tied,
but that during shortages, they can interconnect that way. Many years
ago, maybe in the
1960's or so, they tried a direct AC interconnect across the Hoover Dam,
which bridges the
two grids. The system became unstable, as they had two big pools with a
HUGE length
of transmission lines between them, thus a large inductor between what
acted somewhat like
two big capacitors. They developed a large amount of reactive power
flowing back and forth
between the two systems at a rate of a couple Hz, or maybe it was below
one Hz.

Anyway, they later installed a huge cycloconverter that could couple the
two systems without
regard to phase angle, and send power whichever way it was needed.


Jon
dave
2010-11-24 15:54:35 UTC
Permalink
Google "HVDC in the pacific northwest" will get you a couple of links.
There are two links (npi) from near The Dalles dam on the Columbia to S
CA.

Also there is a site that lays out DC interties as firewalls between
major grids therefore minimizing intergrid interferences leading to
instability.

I expected to find something on the line that runs across WA and crosses
Hyw 18 near Covington (major substation) but have come up blank. The
geometry is that of a HVDC line ... three conductors in a triangle maybe
a foot or so on a side. Nice pictures in a pdf on HVDC in Alberta.

Dave
Post by Dave
Interesting.
Apparently there is a large DC inverter station out in the northwest??
that is used to allow different synchronization of the grids in the
eastern US with the western US or across some other regional divide.
Sort of like a giant AC drive taken to an entirely different level.. ;-)
I remember finding it via a google lookup one time.
Dave
Post by dave
50 some years ago a couple of engineers came to Coulee from the Denver
office (USBR) and installed a mag amp in feed forward mode to control
the penstock gate on one of the small units. This was before the third
powerhouse so the units were all 108 Mw. This one unit was used to swing
the entire Northwest Power Pool ( OR, WA, ID, UT, and most of MT and
WY ). I don't have any quantitative figures but the frequency regulation
got 10X better.
Dave
Post by Roland Jollivet
I heard once (here in SA) that the mains cycles were adjusted close to
midnight, so that the number remained consistent on a daily basis. I don't
know how it's arranged though, since there are so many power stations to
sync.
Regards
Roland
Post by Ian W. Wright
Post by Andy Pugh
Post by Ian W. Wright
hell, I've seen our mains frequency
vary from 47 to 54 Hz over the day many times depending on
the load and no one ever complains.......
Though supposedly you should still always get 4320000 cycles per day.
Yes, you're right but I'm not sure how close they ahere to
that nowadays. With the common frequency dips due to load
problems and the tendency for people to convert to quartz
timing instead of mains synchronous, I wonder if they bother
so much or whether they just try to get it right weekly or
some such. Maybe I'll put a mains frequency monitor on here
out of interest - its 13 years since I retired from the
ratrace of chasing power around plant rooms......
Ian
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App& Earn a Chance To Win $500!
Tap into the largest installed PC base& get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App& Earn a Chance To Win $500!
Tap into the largest installed PC base& get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App& Earn a Chance To Win $500!
Tap into the largest installed PC base& get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Jon Elson
2010-11-23 18:31:08 UTC
Permalink
Post by Ian W. Wright
Yes, you're right but I'm not sure how close they ahere to
that nowadays. With the common frequency dips due to load
problems and the tendency for people to convert to quartz
timing instead of mains synchronous, I wonder if they bother
so much or whether they just try to get it right weekly or
some such. Maybe I'll put a mains frequency monitor on here
out of interest - its 13 years since I retired from the
ratrace of chasing power around plant rooms......
We have several line-synched electronic clocks here, and they all keep
excellent
time. So, it seems the utility here (and thus, the entire midwest-US
electric grid)
maintains very good long-term synch. In fact, these line-sync clocks
are more
accurate than the clocks in some of my computers that do not have NTP
running
to correct their drift.

Jon
Jack Coats
2010-11-23 18:42:30 UTC
Permalink
At one time I heard there were legal reasons that caused the 'grid' to keep
frequency long
term stable. This is why AC clocks are basically 'never off', the frequency
does vary, but
within 24 hours they still must have generated the same number of 'cycles'.
This allows
them to 'slow' the cycles or 'speed up' a bit, but still counting cycles
over a long period is
what makes it appears so frequency stable. ... just my thoughts.
Jon Elson
2010-11-23 18:27:22 UTC
Permalink
Post by Ian W. Wright
John,
I think Andy was assuming that this power generator would be
backup for mains outages when, if the switchover were to be
automatic and seamless,
If you want seamless transfer to emergency power, then the alternator needs
to be online all the time. For big data centers, they have electronic
UPS's and
flywheel energy storage, but that is not really practical for small users.

Jon
Jack Coats
2010-11-23 18:55:19 UTC
Permalink
I helped build a 'small data center', we put in 3 1MW diesel generators to
feed two 1MW APC UPSes, that ran
the data center. The AC ran off of the diesel generators and grid power.
We ran full time off the APC UPSes,
and if the commercial power failed, we automatically brought up all 3
generators. Assuming one would fail,
the others could carry the full load. If one did not fail, the one with the
longest run time in the logs would be
shout down. The UPSes were big enough to run the data center at full load
for about 20 minutes, but it should
be no more than 60 seconds before we were on generator power if commercial
failed, or was out for more than a few
moments.

In the old days I worked on IBM mainframes, and a small mainframe (4341 if I
remember right) had a generator
set to help isolate the power. Internal to that machine the generator set
generated 400Hz, but we fed it from our
commercial building power into the generator. It was not mechanically
efficient, but it kept the machine going.
The HVAC was on regular building power. ,,, Even at larger datacenters where
I worked, the idea was enough
building UPS to keep the computer and networking going. The UPS and
building HVAC and a few essential
services were powered by the generator sets if commercial power died, and we
tried to have power fed from two
or more separate sub-stations linked to different parts of the 'grid'. One
side on the same 'grid feed' as a local
hospital or two if possible ... their power seems to get fixed first :) ...
but at the ones where I was, no major
fly wheel storage or things like that.
Ian W. Wright
2010-11-23 22:33:19 UTC
Permalink
Sounds just like the data centre I ran the services for -
except that we used ICL mainframes ....yuk!!

Ian
Post by Jack Coats
I helped build a 'small data center', we put in 3 1MW diesel generators to
feed two 1MW APC UPSes, that ran
the data center. The AC ran off of the diesel generators and grid power.
We ran full time off the APC UPSes,
and if the commercial power failed, we automatically brought up all 3
generators. Assuming one would fail,
the others could carry the full load. If one did not fail, the one with the
longest run time in the logs would be
shout down. The UPSes were big enough to run the data center at full load
for about 20 minutes, but it should
be no more than 60 seconds before we were on generator power if commercial
failed, or was out for more than a few
moments.
In the old days I worked on IBM mainframes, and a small mainframe (4341 if I
remember right) had a generator
set to help isolate the power. Internal to that machine the generator set
generated 400Hz, but we fed it from our
commercial building power into the generator. It was not mechanically
efficient, but it kept the machine going.
The HVAC was on regular building power. ,,, Even at larger datacenters where
I worked, the idea was enough
building UPS to keep the computer and networking going. The UPS and
building HVAC and a few essential
services were powered by the generator sets if commercial power died, and we
tried to have power fed from two
or more separate sub-stations linked to different parts of the 'grid'. One
side on the same 'grid feed' as a local
hospital or two if possible ... their power seems to get fixed first :) ...
but at the ones where I was, no major
fly wheel storage or things like that.
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App& Earn a Chance To Win $500!
Tap into the largest installed PC base& get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 10.0.1153 / Virus Database: 424/3274 - Release Date: 11/23/10
Jon Elson
2010-11-24 02:16:23 UTC
Permalink
Post by Jack Coats
In the old days I worked on IBM mainframes, and a small mainframe (4341 if I
remember right) had a generator
set to help isolate the power.
A crazy friend of mine bought a pair of 370-145 "mainframes" that were
retired at our
work. We tore one apart for parts, and actually tried to get the other
one working. They had a
17 KVA motor generator set in the back, that converted 208 3 phase 60 Hz
power to 115 V
3 phase 415 Hz power. I guess that means a 2-pole induction motor
turning a 14-pole alternator
at about 3560 RPM. Unfortunately, his small house had only a 60 A 240 V
service, and we
were never able to get the motor up to delta connection without tripping
his main breaker.

So, important digital loads on the machine ran off the MG set, but the
cooling fans, floppy drive
motor and front panel lights ran off the 60 Hz power. The power
supplies were quite small for
linear supplies. They had 3-phase transformer-rectifiers, and then a
thing they called an electronic
capacitor. During the line peaks they shunted current through a big
inductor, then turned this
off during the dips between cycles. This allowed them to use a TINY
capacitor for a pretty hefty
supply, like 5 V at 300 A. The 370/145 was an insanely primitive
minicomputer running an
amazingly "vertical" microcode emulation of the 370 instruction set, and
the performance showed
it. They ran VM/370 and a bunch of MVS and TSO systems under it. It
generally took them 45
minutes to an hour to bring all these OSes back up after a crash. TOTAL
insanity.


Jon
Jon Elson
2010-11-23 18:14:10 UTC
Permalink
Post by Andy Pugh
Doesn't it need to be synched to mains frequency rather than absolute
time? How accurate is NTP (or even GPS time) in this context?
...
The GPS system has atomic clocks accurate to some insane level like one
second in 10,000
years. They have to periodically adjust the clocks for relativistic
effects. The satellites
transmit time to receivers on earth, and the time code system can
resolve the broadcast time
to a nanosecond - literally. Then, after solving the equations for path
delay from several satellites,
the position can be fixed, and so the real time on an arbitrary point on
earth can be determined
to great precision.

NTP is supposed to be good to well within a second, assuming any decent
net connection.

On the other hand, unless this generator is extremely reliable and run
24/7 constantly for months,
trying to sync electric clocks to it seems a poor choice. Maybe find a
good computer with a good
real time clock on it and have it "broadcast" time over a serial line to
remote time displays.
That's what I do here, and Linux even automatically corrects the time
for daylight savings time
twice a year. (I also have temperature and humidity sensors, so the
displays show time, date,
temp and humidity at convenient locations.)

Jon
Przemek Klosowski
2010-11-24 00:12:03 UTC
Permalink
Post by Jon Elson
The GPS system has atomic clocks accurate to some insane level like one
second in 10,000 years.  They have to periodically adjust the clocks for relativistic
effects.
NIST runs the master clocks for the system from Boulder, and keeps
improving them. The most recent generation uses single Aluminum ion,
and is accurate to 1s in over 3 billion years. This is so sensitive
that when the latest clock was being tested it showed a discrepancy
which at first got everyone worried, but then turned out to be the
result of a 17 cm elevation difference between it and the previous
clock used for reference. The clock instrument scientists wrote a
nice Science paper on general relativistic effects of Earth
gravitational field:
http://www.sciencemag.org/content/329/5999/1630.abstract
http://tf.boulder.nist.gov/general/pdf/2447.pdf
Jon Elson
2010-11-24 02:27:48 UTC
Permalink
Post by Przemek Klosowski
Post by Jon Elson
The GPS system has atomic clocks accurate to some insane level like one
second in 10,000 years. They have to periodically adjust the clocks for relativistic
effects.
NIST runs the master clocks for the system from Boulder, and keeps
improving them. The most recent generation uses single Aluminum ion,
and is accurate to 1s in over 3 billion years.
Right, and the clocks on board the GPS satellites are periodically reset
from the NIST master
clocks. The time standards carried on board the satellites are not this
accurate, but are still
Rubidium atomic clocks, and WAYYY more accurate than anything us mere
mortals can
afford. At least, that is the way the system used to be set up, there
might have been updates
since.

Jon
Ian W. Wright
2010-11-24 08:50:03 UTC
Permalink
I think the satellites have caesium clocks now but that is
only from memory. The reason they need to be adjusted fairly
frequently is due to the satellites propensity to fall back
to earth - i.e. their orbits decay over time making the
satellites closer to earth and hence the signal travel time
less. They 'correct' the clock on the satellite for so many
weeks / months and then boost the satellite back up into its
original orbit and start again. They do it this way because
of the high cost of boosting the satellite back up and the
limited fuel on board.

Ian
______________________
Ian W. Wright
Sheffield UK
Post by Jon Elson
Post by Przemek Klosowski
Post by Jon Elson
The GPS system has atomic clocks accurate to some insane level like one
second in 10,000 years. They have to periodically adjust the clocks for relativistic
effects.
NIST runs the master clocks for the system from Boulder, and keeps
improving them. The most recent generation uses single Aluminum ion,
and is accurate to 1s in over 3 billion years.
Right, and the clocks on board the GPS satellites are periodically reset
from the NIST master
clocks. The time standards carried on board the satellites are not this
accurate, but are still
Rubidium atomic clocks, and WAYYY more accurate than anything us mere
mortals can
afford. At least, that is the way the system used to be set up, there
might have been updates
since.
Jon
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App& Earn a Chance To Win $500!
Tap into the largest installed PC base& get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 10.0.1153 / Virus Database: 424/3274 - Release Date: 11/23/10
Don Stanley
2010-11-24 05:00:11 UTC
Permalink
Dave
2010-11-23 05:37:12 UTC
Permalink
A cheap USB GPS can be used to extract the time off the satellite signal.

Easier to interface to than a Walmart Atomic clock.

I have a couple of those "Atomic" clocks and they sometimes get confused.

Dave
Post by Don Stanley
Hi All;
My next project is a remote off grid 60 HZ power unit.
I am thinking of a EMC2 PID to control the RPM.
I am also expecting maybe a 1-2 HZ momentary shift as the
big power loads come on and off line.
I am looking for a method to get a reliable reference that can be used to
average 60 HZ through the power surges and correct a local timer drift
for long term accuracy. A simple WWV corrected timer
(a Wall mart Atomic clock with outputs).
Anyone know of such a device?
Thanks
Don
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App& Earn a Chance To Win $500!
Tap into the largest installed PC base& get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Jack Coats
2010-11-23 18:37:35 UTC
Permalink
Keeping exactly 60Hz isn't that critical. The real timing is being done by
crystals
and the clock can be set using a little GPS or even easier NTP software if
you have
reasonable internet access.

Are you considering staying in sync if commercial power fails?

For computer loads UPSes are great, but their frequency does drift a little.
For
industrial applications (like EMC2 :) even if it is 'home industry' ) exact
timing
doesn't matter, unless you are planning on 'grid tie' your shop to the grid.
Then
you should look into the 'grid tie' hardware that the solar and wind folks
use.
They have been doing it for a long time now (10+ years) and have it down.

Personally I would like to have a 'off the grid' shop. Solar for lights and
light work,
kick in a generator when 'real power' is needed, and as an optional battery
recharger.

I guess it is just the little bit of 'green geek' in me that wants that.
But since I don't
have a 'shop' I can dream. Practicality says that building it and tie it to
the grid and
just pay the bills makes the most economic sense in the short to mid-term (2
to 10 years)
to me.
dave
2010-11-24 16:14:49 UTC
Permalink
Hi Jack,

Have you computed the real costs of running a diesel powered genset for
local power. It comes out to be not very attractive.
Just the cost of the fuel alone is more per KWH than the grid.

Just my tuppence.

Dave
Post by Jack Coats
Keeping exactly 60Hz isn't that critical. The real timing is being done by
crystals
and the clock can be set using a little GPS or even easier NTP software if
you have
reasonable internet access.
Are you considering staying in sync if commercial power fails?
For computer loads UPSes are great, but their frequency does drift a little.
For
industrial applications (like EMC2 :) even if it is 'home industry' ) exact
timing
doesn't matter, unless you are planning on 'grid tie' your shop to the grid.
Then
you should look into the 'grid tie' hardware that the solar and wind folks
use.
They have been doing it for a long time now (10+ years) and have it down.
Personally I would like to have a 'off the grid' shop. Solar for lights and
light work,
kick in a generator when 'real power' is needed, and as an optional battery
recharger.
I guess it is just the little bit of 'green geek' in me that wants that.
But since I don't
have a 'shop' I can dream. Practicality says that building it and tie it to
the grid and
just pay the bills makes the most economic sense in the short to mid-term (2
to 10 years)
to me.
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Jon Elson
2010-11-24 17:29:23 UTC
Permalink
Post by dave
Hi Jack,
Have you computed the real costs of running a diesel powered genset for
local power. It comes out to be not very attractive.
Just the cost of the fuel alone is more per KWH than the grid.
Yes, but if you are 20 miles from the grid, and the only person out
there who wants
to hook up, the power company will usually charge you a HUGE fee, well
over $10K
to bring out the lines. If you are really lucky, they might be planning
on bringing power
out your way, so you should always ask, but the answer may not be pleasing.

I agree, running a Diesel generator 24/7 is totally insane, and the
off-grid home power people
have all sorts of solutions for this. For low-power appliances like
digital alarm clocks,
they have 12 V DC versions with crystal oscillators for RV use, and LED
lighting would
be the best thing to get, and run off 12 V power, too. You could run
the rest of the place
off batteries and an inverter, and fire up the generator once a day to
charge the batteries.
Get solar panels to charge the batteries for days when the machine shop
is not being used.
Get a Beagle Board and car-type LCD screen for you general purpose
computer, the Beagle
only draws 3 W and the 12 V LCDs take maybe 8 W when the backlight is
on. The Beagle
can't run EMC2 just yet, but eventually there will be a real time
package for it.

Jon
Kirk Wallace
2010-11-24 18:20:23 UTC
Permalink
Post by dave
Hi Jack,
Have you computed the real costs of running a diesel powered genset
for > local power. It comes out to be not very attractive.
Post by dave
Just the cost of the fuel alone is more per KWH than the grid.
Yes, but if you are 20 miles from the grid, and the only person out
there who wants to hook up, the power company will usually charge you
a HUGE fee, well over $10K to bring out the lines. If you are really
... snip

I have been giving this some thought. If the grid is available, the only
way off-grid power makes since is if it's almost free. One of my
thoughts is to run my generator head with a steam engine which could
burn, wood, waste oil, propane, whatever. If necessary, enough batteries
would be needed to fill the gap when the mains get interrupted. There
would be enough batteries to to keep the power up until the
hydro-generator kicks in. The hydro source would be a tank on a tower,
or a tank at the top of my hill with the turbine at the bottom. The tank
would be pumped up when I have excess power. Hopefully there would be
enough hydro to allow enough time to get the steam engine up to speed.
Or maybe have the hydro turbine do double duty as a water and steam
turbine. Anyway, this is some of what I think about when I do my morning
thinking.
--
Kirk Wallace
http://www.wallacecompany.com/machine_shop/
http://www.wallacecompany.com/E45/index.html
California, USA
Roland Jollivet
2010-11-24 20:29:38 UTC
Permalink
This is an endless topic. Here's a nice mechanism. Very suitable for the
telemetry/automation orientated.

http://video.google.com/videoplay?docid=-2084492353533644927#

Regards
Roland
Post by Kirk Wallace
Post by dave
Hi Jack,
Have you computed the real costs of running a diesel powered genset
for > local power. It comes out to be not very attractive.
Post by dave
Just the cost of the fuel alone is more per KWH than the grid.
Yes, but if you are 20 miles from the grid, and the only person out
there who wants to hook up, the power company will usually charge you
a HUGE fee, well over $10K to bring out the lines. If you are really
... snip
I have been giving this some thought. If the grid is available, the only
way off-grid power makes since is if it's almost free. One of my
thoughts is to run my generator head with a steam engine which could
burn, wood, waste oil, propane, whatever. If necessary, enough batteries
would be needed to fill the gap when the mains get interrupted. There
would be enough batteries to to keep the power up until the
hydro-generator kicks in. The hydro source would be a tank on a tower,
or a tank at the top of my hill with the turbine at the bottom. The tank
would be pumped up when I have excess power. Hopefully there would be
enough hydro to allow enough time to get the steam engine up to speed.
Or maybe have the hydro turbine do double duty as a water and steam
turbine. Anyway, this is some of what I think about when I do my morning
thinking.
--
Kirk Wallace
http://www.wallacecompany.com/machine_shop/
http://www.wallacecompany.com/E45/index.html
California, USA
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Erik Christiansen
2010-11-30 12:02:45 UTC
Permalink
Post by Kirk Wallace
I have been giving this some thought. If the grid is available, the only
way off-grid power makes since is if it's almost free. One of my
thoughts is to run my generator head with a steam engine which could
burn, wood, waste oil, propane, whatever. If necessary, enough batteries
would be needed to fill the gap when the mains get interrupted. There
would be enough batteries to to keep the power up until the
hydro-generator kicks in. The hydro source would be a tank on a tower,
or a tank at the top of my hill with the turbine at the bottom. The tank
would be pumped up when I have excess power. Hopefully there would be
enough hydro to allow enough time to get the steam engine up to speed.
Or maybe have the hydro turbine do double duty as a water and steam
turbine. Anyway, this is some of what I think about when I do my morning
thinking.
With 2 sq km of forest on the farm, my original reason for buying the
lathe and mill was to build a steam engine, or convert an old horizontal
IC engine with fat flywheels. (There's one in a corner of the shed.)

It's only progressed as far as calculations of power output vs RPM,
average steam pressure, cylinder dimensions, etc. The other calcs are
for the boiler. Due to stiff regulations, having one made isn't cheap.
So ... flash steam (monotube boiler) is very tempting. (They need a
constant load, because there's no hot water to provide a steam reserve.
I figure that batteries could substitute.)

We don't have much clean water out there. (18,000 L total, for domestic
use.) That mandates a condenser. Dam water could be used on the outside
of that. All that's lacking is the round tuit.

Gasification is an alternative, but cleaning the tar and crap from the
gas generator output necessitates filters, which need regular cleaning.
That's a grubby and non-trivial job, from the writings of those who've
done it.

Ideal would be a stirling engine with integral alternator:

http://www.stirling.dk/ # Nifty, but still in one-off production,
# after about 5 years. Smallest is 9 kw.
# Larger units use woodchips.

It's a while since I went through:

http://www.stirlingengines.org.uk/manufact/post.html

Maybe someone's within cooee of thinking hard about going commercial
RSN.

Erik

PS: If my simple script for calculating steam engine power vs various
% cut-off, etc., would amuse, then just yodel off-list. It looks
like this in use:

Input: Pressure(PSI) Stroke(cm) Bore(cm) RPM

200 10 8 400 <- User input
------------------
Cutoff HP kW
------------------
25% 3.7 2.8
33% 4.2 3.1
37.5% 4.6 3.4
50% 5.2 3.9
62.5% 5.7 4.2
67% 5.8 4.3
75% 6.0 4.5
87.5% 6.1 4.6
100% 6.2 4.6
------------------

There's also two pages of maths & musings for engine & boiler to
hand, from last time I nudged closer toward something workable.

PPS: Water and steam turbines look very different, unfortunately.
A single Pelton wheel could do for the former, but alternating
stator and rotor wheels of increasing diameter are needed in the
latter, to utilise the expanding steam. (AIUI, anyway. ;-)

PPPS: Hope there's some ideas in that lot.
Erik Christiansen
2010-11-30 12:37:10 UTC
Permalink
Post by Jon Elson
Yes, but if you are 20 miles from the grid, and the only person out
there who wants to hook up, the power company will usually charge you
a HUGE fee, well over $10K to bring out the lines. If you are really
lucky, they might be planning on bringing power out your way, so you
should always ask, but the answer may not be pleasing.
When we last asked, about 20 years ago, it was about $60k to connect out
on the farm, and about 20 acres of forest would have to be bulldozed for
the line. Prices aren't likely to have fallen much, and now we have to
plant 10 trees for every one we 'doze.
Post by Jon Elson
I agree, running a Diesel generator 24/7 is totally insane, and the
off-grid home power people have all sorts of solutions for this.
Yeah. There's plenty to do outside, so who needs power until dark?
With the help of a full-sized gas refrigerator, that works for us.
In the evening we arc up the genny.
Post by Jon Elson
For low-power appliances like digital alarm clocks, they have 12 V DC
versions with crystal oscillators for RV use, and LED lighting would
be the best thing to get, and run off 12 V power, too. You could run
the rest of the place off batteries and an inverter, and fire up the
generator once a day to charge the batteries. Get solar panels to
charge the batteries for days when the machine shop is not being used.
Get a Beagle Board and car-type LCD screen for you general purpose
computer, the Beagle only draws 3 W and the 12 V LCDs take maybe 8 W
when the backlight is on. The Beagle can't run EMC2 just yet, but
eventually there will be a real time package for it.
If we still lived out there the whole time, I'd do much of that. (And
get new deep cycle batteries for the 24v [1] inverter.) But we lost
between 600 and 700 large trees in the storms in 2006. That's several
thousand tonnes of hardwood, which will rot away in 50 years or so.
Burning fossil fuel in the petrol generator isn't as appealing as a
steam engine fooshing away quietly, with some boiler management
electronics, and an automatic stoker. (It's just that boilers are a bit
dangerous, unless you go for a monotube.)

Erik

[1] A 48v sinewave inverter would be better, but expensive.
(Even at 24v, the DC draw is hefty when you pull a few amps of 240v)
Don Stanley
2010-12-01 01:28:59 UTC
Permalink
On Tue, Nov 30, 2010 at 7:37 AM, Erik Christiansen
Post by Erik Christiansen
Post by Jon Elson
Yes, but if you are 20 miles from the grid, and the only person out
there who wants to hook up, the power company will usually charge you
a HUGE fee, well over $10K to bring out the lines. If you are really
lucky, they might be planning on bringing power out your way, so you
should always ask, but the answer may not be pleasing.
When we last asked, about 20 years ago, it was about $60k to connect out
on the farm, and about 20 acres of forest would have to be bulldozed for
the line. Prices aren't likely to have fallen much, and now we have to
plant 10 trees for every one we 'doze.
Post by Jon Elson
I agree, running a Diesel generator 24/7 is totally insane, and the
off-grid home power people have all sorts of solutions for this.
Yeah. There's plenty to do outside, so who needs power until dark?
With the help of a full-sized gas refrigerator, that works for us.
In the evening we arc up the genny.
Post by Jon Elson
For low-power appliances like digital alarm clocks, they have 12 V DC
versions with crystal oscillators for RV use, and LED lighting would
be the best thing to get, and run off 12 V power, too. You could run
the rest of the place off batteries and an inverter, and fire up the
generator once a day to charge the batteries. Get solar panels to
charge the batteries for days when the machine shop is not being used.
Get a Beagle Board and car-type LCD screen for you general purpose
computer, the Beagle only draws 3 W and the 12 V LCDs take maybe 8 W
when the backlight is on. The Beagle can't run EMC2 just yet, but
eventually there will be a real time package for it.
If we still lived out there the whole time, I'd do much of that. (And
get new deep cycle batteries for the 24v [1] inverter.) But we lost
between 600 and 700 large trees in the storms in 2006. That's several
thousand tonnes of hardwood, which will rot away in 50 years or so.
Burning fossil fuel in the petrol generator isn't as appealing as a
steam engine fooshing away quietly, with some boiler management
electronics, and an automatic stoker. (It's just that boilers are a bit
dangerous, unless you go for a monotube.)
Erik
[1] A 48v sinewave inverter would be better, but expensive.
(Even at 24v, the DC draw is hefty when you pull a few amps of 240v)
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Hi to All who responded or were interested:
It will totally amazing at the number of structures that will be off the
grid
over the next decade.

More to come next year.

Thanks
Don
yann jautard
2010-12-01 08:26:07 UTC
Permalink
Post by Erik Christiansen
If we still lived out there the whole time, I'd do much of that. (And
get new deep cycle batteries for the 24v [1] inverter.) But we lost
between 600 and 700 large trees in the storms in 2006. That's several
thousand tonnes of hardwood, which will rot away in 50 years or so.
Burning fossil fuel in the petrol generator isn't as appealing as a
steam engine fooshing away quietly, with some boiler management
electronics, and an automatic stoker. (It's just that boilers are a bit
dangerous, unless you go for a monotube.)
I work in a company that makes electrification system for houses that
are away from grid.
In my opinion, photovoltaics panels + deep cycles batteries are the
better way to insure a reliable power source.
But with such a system, if you don't want it to cost you too much,
you'll get limited energy ressource. Enough for the light and the
basical home usage. Don't even expect to warm your water with it, for
instance.

So if you need sometimes large power amount, having a diesel generator
coupled to you inverter is a ood idea. If you use a smart enough
inverter, like one you can get from Xantrex, it can be programmed to
automatically start generator when the required AC current go over a
limit. This will preserve batteries.
You can also do not use photovoltaics panels at all, and only use the
generator to charge the batterie every 2 or 3 days, but in y opinion
this is not a good solution. PV panel never give up, mecanical machine
such a generator engine does. So having just enough pv panels to insure
the basical energy production is a good idea.

Regarding the generator, if full automatisation is not required, and
considering you have a fairly amount of wood available, you can look for
making a wood gas generator and run a old gasoline generator with it.
I think this is really simpler thant steam engine, and really safer.
With a lower efficiency anyway, but if you have lot of wood this will
not be a problem.

With some clever thinking, I think it is also possible to automatise the
wood gas generator. For example starting the fire with a electrical hot
air generator, like they do in pellets stoves. The you may have to make
some automated system to feed wood in the burner, and you should be done.

http://en.wikipedia.org/wiki/Wood_gas_generator

http://fr.wikipedia.org/wiki/Gazog%C3%A8ne
Jack Coats
2010-12-01 14:16:40 UTC
Permalink
I like the idea of using the wood you want to burn as heat, there are
many places
around here that use the external boilers, small building built remote
from houses
or other heated structures by 20 to 100' or so, that then pipe the
heated water to
the buildings. I have seen them advertised in Mother Earth News.

Then use PV for battery charging. If you put up LED lights (expensive
currently,
but use lots less electricity per lumen) especially the DC ones, you would be in
good shape. Running dual electric lines around your house, one for DC only
(use extra sized wire for DC to keep the resistance losses down), and a separate
set for AC that could all go back to an inverter or small motor
generator set for
when it is needed.

If I had it to do over myself, and expense was no object :) ... I
would do 48VDC,
charge with solar, have an 'automatic generator' setup like Xantrex
that Yann talked about.
Probably set up the generator to kick in if the batteries got to low or the AC
needed was over what an inverter could use. But first, it would need to go into
a new house/workshop that was thermally efficient (something I don't have now).
If I had the wood Erik has, the outside boiler or making some 'rocket
stoves' (efficent
heating type stoves, not really for cooking) would be great to heat with.
Dreams are good.
<> ... Jack
Whatever you do, work at it with all your heart... Colossians 3:23
Stuart Stevenson
2010-11-24 17:47:13 UTC
Permalink
Gentlemen,
Having researched the subject of isolating from the grid to try to convert
my shop I will interject a comment or two.
The KWH cost of utility power is amazingly cheap compared to generating it
locally.
The KWH cost is not the only cost. At least once a year we experience a
power glitch of the form necessary to harm sensitive electronics (and at
times other things connected to the power lines).
We have ground devices to attempt to shield our machine controls. These
shield devices do not always work. I do not know how much the added cost
truly is but the aggravation is tremendous. When you see the lights flicker
bright or dim you cringe. If a machine stops you do not know if it will
restart. Many times one of the machines will not restart. The remedy for
restart is usually a board replacement. The board is not cheap. The downtime
is not cheap. How do you quantify it?
How do you know if the next board you replace was not damaged by the same
event that killed the previous board? We have replaced all four boards in
each of three machines in a year. This replacement came after a particularly
damaging power fluctuation.
I would definitely like to be isolated from the grid.
thanks
Stuart
--
dos centavos
Jim Fleig
2010-11-24 18:11:41 UTC
Permalink
Hello All,

Happy Thanksgiving one day early!

To protect sensitive electronics in machine tool controls (and many
other places) put an UPS (Uninterruptible Power Supply) between the
power source and the sensitive electronics. Most DC power supplies
are converting 110 VAC to whatever DC voltages are needed. The UPS
unit is easy to put into the circuit and for most machine tool
controls will cost about $150.00. A higher joule rating provides
better protection. Problem power, toasted UPS unit, $150.00
replacement expense purchasing the item at the local big box computer
store and the system is up and running again. Relatively inexpensive
and quick recovery.

Have a good day,

Jim Fleig
CNC Services

585 975-9618
Post by Stuart Stevenson
Gentlemen,
Having researched the subject of isolating from the grid to try to convert
my shop I will interject a comment or two.
The KWH cost of utility power is amazingly cheap compared to
generating it
locally.
The KWH cost is not the only cost. At least once a year we
experience a
power glitch of the form necessary to harm sensitive electronics (and at
times other things connected to the power lines).
We have ground devices to attempt to shield our machine controls. These
shield devices do not always work. I do not know how much the added cost
truly is but the aggravation is tremendous. When you see the lights flicker
bright or dim you cringe. If a machine stops you do not know if it will
restart. Many times one of the machines will not restart. The remedy for
restart is usually a board replacement. The board is not cheap. The downtime
is not cheap. How do you quantify it?
How do you know if the next board you replace was not damaged by the same
event that killed the previous board? We have replaced all four boards in
each of three machines in a year. This replacement came after a particularly
damaging power fluctuation.
I would definitely like to be isolated from the grid.
thanks
Stuart
--
dos centavos
---
---
---
---------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Przemek Klosowski
2010-11-24 18:54:04 UTC
Permalink
Post by Jim Fleig
Hello All,
Happy Thanksgiving one day early!
To protect sensitive electronics in machine tool controls (and many
other places) put an UPS (Uninterruptible Power Supply) between the
power source and the sensitive electronics.  Most DC power supplies
are converting 110 VAC to whatever DC voltages are needed.  The UPS
unit is easy to put into the circuit and for most machine tool
controls will cost about $150.00.
Absolutely, but the price goes up tremendously with power. I am now
pricing an 80kW system and it seems to be 70k$ (about half of it is
the batteries for 30 min uptime so some of the price could be cut).
Come to think about it, the rule of thumb seems to be 1k$ per 1kW

I suppose a CNC system could use a simple power conditioner for the
power feed to the motors and the UPS for the electronics/controller,
but that does not guarantee against power surge getting into the power
stage and flowing back into the controls.
Peter Blodow
2010-11-24 19:25:41 UTC
Permalink
Hello gentlemen,
I have another solution for the power supply problem: move to Europe! In
our village (270 some people) we have two transformer stations on a
grid-connected 20 kV three-phase line. The power lines go to (or come
from) three neighboring villages and eventually end up at the region's
110 kV-transformer station about 20 km away.

Our house has a 400V, 50 A, 3-phase supply as all the houses do, this is
average. All the cabling is under the roads under ground, in our street
a PVC cable 3x75 mm sq. + 50 mm sq.for the middle conductor (implicit
ground). This is standard für all single houses, a corresponding bit
more for larger ones. I have experienced one or two power failures since
we live here (32 years). In a larger city like Munich, for instance, 50
km from us, power failures are completely unknown. I have experienced
one when I was a little boy. Sometimes we have short power drops when
lighning hit the power line, but only for a few sine periods, so most
PC's and other appliances don't even recognize it. America should
improve their energy supply!

By the way: I have learned electricity, house wiring and TV repair at
high school in Michigan some 45 years ago and know what the difference ist!

Peter Blodow
Post by Przemek Klosowski
Post by Jim Fleig
Hello All,
Happy Thanksgiving one day early!
To protect sensitive electronics in machine tool controls (and many
other places) put an UPS (Uninterruptible Power Supply) between the
power source and the sensitive electronics. Most DC power supplies
are converting 110 VAC to whatever DC voltages are needed. The UPS
unit is easy to put into the circuit and for most machine tool
controls will cost about $150.00.
Absolutely, but the price goes up tremendously with power. I am now
pricing an 80kW system and it seems to be 70k$ (about half of it is
the batteries for 30 min uptime so some of the price could be cut).
Come to think about it, the rule of thumb seems to be 1k$ per 1kW
I suppose a CNC system could use a simple power conditioner for the
power feed to the motors and the UPS for the electronics/controller,
but that does not guarantee against power surge getting into the power
stage and flowing back into the controls.
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Przemek Klosowski
2010-11-24 18:54:04 UTC
Permalink
Post by Jim Fleig
Hello All,
Happy Thanksgiving one day early!
To protect sensitive electronics in machine tool controls (and many
other places) put an UPS (Uninterruptible Power Supply) between the
power source and the sensitive electronics.  Most DC power supplies
are converting 110 VAC to whatever DC voltages are needed.  The UPS
unit is easy to put into the circuit and for most machine tool
controls will cost about $150.00.
Absolutely, but the price goes up tremendously with power. I am now
pricing an 80kW system and it seems to be 70k$ (about half of it is
the batteries for 30 min uptime so some of the price could be cut).
Come to think about it, the rule of thumb seems to be 1k$ per 1kW

I suppose a CNC system could use a simple power conditioner for the
power feed to the motors and the UPS for the electronics/controller,
but that does not guarantee against power surge getting into the power
stage and flowing back into the controls.
unknown
1970-01-01 00:00:00 UTC
Permalink
Thanks John and All;
It appears NTP solves the clock drift problem in Linux systems, completely.
As soon as I can ferret out the HAL bits and pieces needed to keep the
generator synced to the PC clock, this problem is solved.
Any clues of the bits and pieces needed would give me a great jump start.

Sincere Thanks
Don
Post by John Kasunich
Do you have internet access at the site?
If you are running an PC for EMC, you can use NTP to keep
the PC's clock synced to the rest of the world, and some HAL
bits and pieces to keep the generator synced to the PC
clock. EMC's encoder component could easily count 60Hz.
Post by Don Stanley
Hi All;
My next project is a remote off grid 60 HZ power unit.
I am thinking of a EMC2 PID to control the RPM.
I am also expecting maybe a 1-2 HZ momentary shift as the
big power loads come on and off line.
I am looking for a method to get a reliable reference that can be used to
average 60 HZ through the power surges and correct a local timer drift
for long term accuracy. A simple WWV corrected timer
(a Wall mart Atomic clock with outputs).
Anyone know of such a device?
Thanks
Don
------------------------------------------------------------------------------
Post by Don Stanley
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for
grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
--
John Kasunich
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
KMC@gmail.com
2010-11-24 19:47:15 UTC
Permalink
I have researched this subject over the last 10+ years.
And I know a couple people that have tried limited phases of this experiment
as well.
My conclusion is, if do not have enough hobbies and too much free time,
adding 'home shop power station' to the list might be fun. But there is no
savings to be had over grid power. Even if you use the waste heat from the
generator in winter, plan on being much less efficient in your work or shop
activities, as your generating system will demand time and attention. I have
a waste oil heater in the shop, been here for 10 years. I finally got tired
of all the time it wastes (waste oil is NOT free heat) and am installing a
gas furnace in it's place.

Back to your regularly scheduled programming.

dk
Loading...