Alternative Power sources - 12V Generator Problems

Author

12V Generator Problems

Chris F.

2005-11-12, 12:21 pm

I'm trying to build a 12V generator using plans from the following page:
http://www.homepower.com/files/mark8.pdf
However it seems that the schematic for the electronic field regulator is
flawed. I cannot get it to function at all, and I suspect that some errors
have been made in the schematic. First of all, the output of the LM723
should be pin 10, not pin 9 as the schematic states. Second, isn't the LM555
supposed to be a timer chip and not a regulator?? Perhaps it was supposed to
be an op-amp or something.
Has anyone else tried to build this thing, and can perhaps give me some
advice?
Any help would be much appreciated.

Martin Riddle

2005-11-12, 6:21 pm

The LM555 is used a s a PWM controller. It uses the LM723 pin 9 (whis is a Zener connected to pin 10) to drive the internal LM555
opamp. Vz appears to be the anode of a 6.2v zener connected to Vout.

Check your wiring. What in the voltage at pin 9? Do you see pulses at pin 3 of the 555?

Cheers

"Chris F." <zappymanREMOVETHIS@hotmail.com> wrote in message news:P0odf.123060$Ph4.3774345@ursa-nb00s0.nbnet.nb.ca...
> I'm trying to build a 12V generator using plans from the following page:
> http://www.homepower.com/files/mark8.pdf
> However it seems that the schematic for the electronic field regulator is
> flawed. I cannot get it to function at all, and I suspect that some errors
> have been made in the schematic. First of all, the output of the LM723
> should be pin 10, not pin 9 as the schematic states. Second, isn't the LM555
> supposed to be a timer chip and not a regulator?? Perhaps it was supposed to
> be an op-amp or something.
> Has anyone else tried to build this thing, and can perhaps give me some
> advice?
> Any help would be much appreciated.
>
>

Gordon reeder

2005-11-12, 7:21 pm

"Chris F." <zappymanREMOVETHIS@hotmail.com> wrote in
news:P0odf.123060$Ph4.3774345@ursa-nb00s0.nbnet.nb.ca:

> I'm trying to build a 12V generator using plans from the following
> page:
> http://www.homepower.com/files/mark8.pdf
> However it seems that the schematic for the electronic field
> regulator is
> flawed. I cannot get it to function at all, and I suspect that some
> errors have been made in the schematic. First of all, the output of
> the LM723 should be pin 10, not pin 9 as the schematic states. Second,
> isn't the LM555 supposed to be a timer chip and not a regulator??
> Perhaps it was supposed to be an op-amp or something.
> Has anyone else tried to build this thing, and can perhaps give me
> some
> advice?
> Any help would be much appreciated.
>
>
>

I do not have datasheets at hand for the two parts in question.
But note that the pinning of the device will vary according to
its packaging. Today it is not unusual to find the same device
available in several package styles and each package has a different
pinout. It's rather annoying that the schimatic doesn't denote
package style nor does it have reference desigantors for the parts.

That said...

The 555 is indead a timer. It is used here as an astable multivibrator.
It's uotput is (as beat I can tell) a constant frequency. But the
duty cycle (ratio of high and low times) is variable. The 555 by
it's self determines the constant current charge charge rate. The
LM723 monitors system voltage and when it gets too high it reduces
the charge current by pulling the control input to the 555. By
pulling the control input it reduses the pulse width duty cycle
and if need be, completly turn it off.

To bench test this circuit, connect a battery to the input terminals
and a automobile tail light bulb to the output. If you have a meter,
that would be good too. disconnect the wire from the output of the
LM273 that goes to the LM555. This allows the 555 to run without
interfearence from the LM273. Once you get the 555 part of the
circuit working, reconnect the LM273 and work on debugging that
part of the circuit.

Chris F.

2005-11-12, 8:21 pm

That makes a bit more sense. I didn't realize the 555 output was pulsed,
maybe that's why I got so confused. I have a scope I will use to do some
more precise troubleshooting.
One more question; the PNP transistor I'm using (the big one on a
heatsink) seems a bit strange. When testing with an ohmetter (actually the
diode check function of a DMM) it appears that there is a damper diode from
collector to emitter. I checked half a dozen identical transistors and got
the same results. The transistor shown in the plans doesn't suggest it has
any built-in diodes. Might my transistor be the wrong type for this? (it
crosses to an NTE 104)
Thanks again.
"Gordon reeder" <gonzo@alltomyself.com> wrote in message
news:Xns970C96DB4A4B4greederxprtnet@213.155.197.138...
> "Chris F." <zappymanREMOVETHIS@hotmail.com> wrote in
> news:P0odf.123060$Ph4.3774345@ursa-nb00s0.nbnet.nb.ca:
>
>
> I do not have datasheets at hand for the two parts in question.
> But note that the pinning of the device will vary according to
> its packaging. Today it is not unusual to find the same device
> available in several package styles and each package has a different
> pinout. It's rather annoying that the schimatic doesn't denote
> package style nor does it have reference desigantors for the parts.
>
> That said...
>
> The 555 is indead a timer. It is used here as an astable multivibrator.
> It's uotput is (as beat I can tell) a constant frequency. But the
> duty cycle (ratio of high and low times) is variable. The 555 by
> it's self determines the constant current charge charge rate. The
> LM723 monitors system voltage and when it gets too high it reduces
> the charge current by pulling the control input to the 555. By
> pulling the control input it reduses the pulse width duty cycle
> and if need be, completly turn it off.
>
> To bench test this circuit, connect a battery to the input terminals
> and a automobile tail light bulb to the output. If you have a meter,
> that would be good too. disconnect the wire from the output of the
> LM273 that goes to the LM555. This allows the 555 to run without
> interfearence from the LM273. Once you get the 555 part of the
> circuit working, reconnect the LM273 and work on debugging that
> part of the circuit.

Martin Riddle

2005-11-12, 9:21 pm

The NTE104 should be fine, There is no CE diode in that part. And it would not affect the circuit.
Make sure your are using a 2N2222A for Q1, or a transistor with an Ic rating of >=100ma for pulse currents of >5 Amps in the MJE2955
(beta is ~100).

You'll need a Mica Insulator for the 295.

Pin 4 of the NE555 is the reset, so the 1k pot on the LM723 must be for the High voltage limit. He could have used a opamp or
comparator.
The 50k pot on the NE555 is the Charging current.

It looks like under low voltage the NE555 output is a square wave with a duty cycle dependant on the 50k pot..
Once the voltage gets to the High limit the NE555 is forced into reset.

The Author either didnt know how the circuit worked or the publisher left out the details on how the circuit works.

Cheers

"Chris F." <zappymanREMOVETHIS@hotmail.com> wrote in message news:_9vdf.123252$Ph4.3780295@ursa-nb00s0.nbnet.nb.ca...
> That makes a bit more sense. I didn't realize the 555 output was pulsed,
> maybe that's why I got so confused. I have a scope I will use to do some
> more precise troubleshooting.
> One more question; the PNP transistor I'm using (the big one on a
> heatsink) seems a bit strange. When testing with an ohmetter (actually the
> diode check function of a DMM) it appears that there is a damper diode from
> collector to emitter. I checked half a dozen identical transistors and got
> the same results. The transistor shown in the plans doesn't suggest it has
> any built-in diodes. Might my transistor be the wrong type for this? (it
> crosses to an NTE 104)
> Thanks again.
> "Gordon reeder" <gonzo@alltomyself.com> wrote in message
> news:Xns970C96DB4A4B4greederxprtnet@213.155.197.138...
>
>

Chris F.

2005-11-13, 11:21 am

I tried redoing the 555 stage, and the results were something like this;
with a small 12V auto light hooked on the output, I could vary the output
current only somewhat by adjusting the 50k pot. Is this supposed to be
adjustable all the way to 0, or just partway instead?
I think I'm on the right track anyway. Or at least I hope.

"Martin Riddle" <martinriddle@hotmail.com> wrote in message
news:rbwdf.2828$C62.2262@trndny04...
> The NTE104 should be fine, There is no CE diode in that part. And it would
not affect the circuit.
> Make sure your are using a 2N2222A for Q1, or a transistor with an Ic
rating of >=100ma for pulse currents of >5 Amps in the MJE2955
> (beta is ~100).
>
> You'll need a Mica Insulator for the 295.
>
> Pin 4 of the NE555 is the reset, so the 1k pot on the LM723 must be for
the High voltage limit. He could have used a opamp or
> comparator.
> The 50k pot on the NE555 is the Charging current.
>
> It looks like under low voltage the NE555 output is a square wave with a
duty cycle dependant on the 50k pot..
> Once the voltage gets to the High limit the NE555 is forced into reset.
>
> The Author either didnt know how the circuit worked or the publisher left
out the details on how the circuit works.
>
> Cheers
>
>
> "Chris F." <zappymanREMOVETHIS@hotmail.com> wrote in message
news:_9vdf.123252$Ph4.3780295@ursa-nb00s0.nbnet.nb.ca...
pulsed,[color=darkred]
the[color=darkred]
from[color=darkred]
got[color=darkred]
has[color=darkred]
Second,[color=darkred]
multivibrator.[color=darkred]
>
>

Chris F.

2005-11-13, 3:21 pm

My mistake, there was a wiring error. Seems to work great, now to hook it
up to the alternator and see how it works.
I'm trying to add an extra feature to this design; the ability to switch
between two modes of current depending on the task. In normal mode, it will
limit the current to 10 amps, to suit my 100Ah battery. When a heavy load
(45+ amps) is running, it will switch to high current and use the full 55
amps from the alternator. I plan to do this by having a DPDT relay switch
between two 50k pots, one adjusted for 10 amps and the other for max
current. A current sensing circuit will engage the relay when necessary (I
plan to use the inverters own current measuring circuit to trip the relay -
hard to explain)
Hopefully this will work out in the end, and I'll have a reliable backup
power source for those many winter blackouts.
Thanks again for the advice.
"Chris F." <zappymanREMOVETHIS@hotmail.com> wrote in message
news:nDIdf.123516$Ph4.3792068@ursa-nb00s0.nbnet.nb.ca...
> I tried redoing the 555 stage, and the results were something like this;
> with a small 12V auto light hooked on the output, I could vary the output
> current only somewhat by adjusting the 50k pot. Is this supposed to be
> adjustable all the way to 0, or just partway instead?
> I think I'm on the right track anyway. Or at least I hope.
>
> "Martin Riddle" <martinriddle@hotmail.com> wrote in message
> news:rbwdf.2828$C62.2262@trndny04...
would[color=darkred]
> not affect the circuit.
> rating of >=100ma for pulse currents of >5 Amps in the MJE2955
> the High voltage limit. He could have used a opamp or
> duty cycle dependant on the 50k pot..
left[color=darkred]
> out the details on how the circuit works.
> news:_9vdf.123252$Ph4.3780295@ursa-nb00s0.nbnet.nb.ca...
> pulsed,
some[color=darkred]
> the
> from
> got
> has
(it[color=darkred]
following[color=darkred]
some[color=darkred]
of[color=darkred]
> Second,
me[color=darkred]
> multivibrator.
meter,[color=darkred]
>
>

Ulysses

2005-11-13, 6:21 pm

"Chris F." <zappymanREMOVETHIS@hotmail.com> wrote in message
news:fXLdf.123628$Ph4.3794686@ursa-nb00s0.nbnet.nb.ca...
> My mistake, there was a wiring error. Seems to work great, now to hook
it
> up to the alternator and see how it works.
> I'm trying to add an extra feature to this design; the ability to switch
> between two modes of current depending on the task. In normal mode, it
will
> limit the current to 10 amps, to suit my 100Ah battery. When a heavy load
> (45+ amps) is running, it will switch to high current and use the full 55
> amps from the alternator. I plan to do this by having a DPDT relay switch
> between two 50k pots, one adjusted for 10 amps and the other for max
> current. A current sensing circuit will engage the relay when necessary (I
> plan to use the inverters own current measuring circuit to trip the
relay -
> hard to explain)
> Hopefully this will work out in the end, and I'll have a reliable backup
> power source for those many winter blackouts.
> Thanks again for the advice.
> "Chris F." <zappymanREMOVETHIS@hotmail.com> wrote in message
> news:nDIdf.123516$Ph4.3792068@ursa-nb00s0.nbnet.nb.ca...
this;[color=darkred]
output[color=darkred]
> would
for[color=darkred]
a[color=darkred]
reset.[color=darkred]
> left
> some
(actually[color=darkred]
diode[color=darkred]
and[color=darkred]
it[color=darkred]
> (it
> following
> some
> of
regulator??[color=darkred]
give[color=darkred]
> me
different[color=darkred]
parts.[color=darkred]
terminals[color=darkred]
> meter,
the[color=darkred]

Now that you have it all figured out can you explain to me the diodes on the
schematic with arrows coming out of them? I've never seen that before.
They are not indicated as D1 or D2. They must be variable diodes.[color=darkred]
>
>

Martin Riddle

2005-11-13, 6:21 pm

"Ulysses" <therealulysses@yahoo.com> wrote in message news:11nfbgrm323qv64@corp.supernews.com...
>
> "Chris F." <zappymanREMOVETHIS@hotmail.com> wrote in message
> news:fXLdf.123628$Ph4.3794686@ursa-nb00s0.nbnet.nb.ca...
> it
> will
> relay -
> this;
> output
> for
> a
> reset.
> (actually
> diode
> and
> it
> regulator??
> give
> different
> parts.
> terminals
> the
>
> Now that you have it all figured out can you explain to me the diodes on the
> schematic with arrows coming out of them? I've never seen that before.
> They are not indicated as D1 or D2. They must be variable diodes.
>
>
Those are LED's.

Chris F.

2005-11-15, 11:21 am

One final question about my alternator; it's a GM Delco type, but the wiring
is labeled a bit differently than most. Instead of a R and F next to the
rotor and field windings, it says "1" and "2". Any idea which is which?

"Chris F." <zappymanREMOVETHIS@hotmail.com> wrote in message
news:P0odf.123060$Ph4.3774345@ursa-nb00s0.nbnet.nb.ca...
> I'm trying to build a 12V generator using plans from the following page:
> http://www.homepower.com/files/mark8.pdf
> However it seems that the schematic for the electronic field regulator
is
> flawed. I cannot get it to function at all, and I suspect that some errors
> have been made in the schematic. First of all, the output of the LM723
> should be pin 10, not pin 9 as the schematic states. Second, isn't the
LM555
> supposed to be a timer chip and not a regulator?? Perhaps it was supposed
to
> be an op-amp or something.
> Has anyone else tried to build this thing, and can perhaps give me some
> advice?
> Any help would be much appreciated.
>
>

wmbjk

2005-11-15, 1:21 pm

On Tue, 15 Nov 2005 14:51:40 GMT, "Chris F."
<zappymanREMOVETHIS@hotmail.com> wrote:

>One final question about my alternator; it's a GM Delco type, but the wiring
>is labeled a bit differently than most. Instead of a R and F next to the
>rotor and field windings, it says "1" and "2". Any idea which is which?

http://www.newindo.com/pdf/1G286.pdf nice repair manual for 21si,
typical Delco construction

#1 goes to one side of idiot light, other side to positive.
#2 goes to positive.

But I would think that if you're using an external regulator and want
to use the original terminals (to avoid opening the case), then you'd
probably short the negative brush lead to the case to bypass the
internal regulator. There's a tab meant to be shorted with a
screwdriver for testing. Look for a D-shaped hole in the back of the
case. Perhaps you could break out a bit of the case, and fish a wire
with a spade connector through the hole and onto that tab .

FWIW, mine is a 27si (larger case but otherwise typical old Delco)
originally 12V,100A). I have it on a 24V system, alternator speed
about 4800, fixed output about 75A for manually controlled bulk
charging. I removed the diode trio and resistor, and bypassed the
regulator by routing new wires directly from the brush terminals.
External rheostat control. I found that the rectifier was the weak
link, probably mostly due to inadequate cooling. I threw away the
stock one (which opens up the case for improved airflow), and use an
external rectifier that's cooled by the same (electrically powered)
fans that cool the generator enclosure.

Wayne

Ulysses

2005-11-15, 1:21 pm

"Martin Riddle" <martinriddle@hotmail.com> wrote in message
news:SbOdf.2231$vS4.1302@trnddc01...
>
> "Ulysses" <therealulysses@yahoo.com> wrote in message
news:11nfbgrm323qv64@corp.supernews.com...
hook[color=darkred]
switch[color=darkred]
load[color=darkred]
55[color=darkred]
switch[color=darkred]
necessary (I[color=darkred]
backup[color=darkred]
be[color=darkred]
it[color=darkred]
Ic[color=darkred]
be[color=darkred]
with[color=darkred]
publisher[color=darkred]
was[color=darkred]
do[color=darkred]
on a[color=darkred]
transistors[color=darkred]
suggest[color=darkred]
this?[color=darkred]
field[color=darkred]
that[color=darkred]
output[color=darkred]
states.[color=darkred]
question.[color=darkred]
device[color=darkred]
denote[color=darkred]
the[color=darkred]
by[color=darkred]
The[color=darkred]
reduces[color=darkred]
By[color=darkred]
cycle[color=darkred]
of[color=darkred]
without[color=darkred]
the[color=darkred]
that[color=darkred]
the[color=darkred]
> Those are LED's.
>
>
>

Ah. This is what happens when you get old. First the hair starts to fall
out, then the knees go, then the, um, what's that called. Oh yea, memory
fades...

daestrom

2005-11-15, 5:21 pm

"wmbjk" <wmbjkREMOVE@citlink.net> wrote in message
news:lj3kn19q4fpdssuf43cfgk3sq3l78ibvod@4ax.com...
> On Tue, 15 Nov 2005 14:51:40 GMT, "Chris F."
> <zappymanREMOVETHIS@hotmail.com> wrote:
>
>
> http://www.newindo.com/pdf/1G286.pdf nice repair manual for 21si,
> typical Delco construction
>
> #1 goes to one side of idiot light, other side to positive.
> #2 goes to positive.
>
> But I would think that if you're using an external regulator and want
> to use the original terminals (to avoid opening the case), then you'd
> probably short the negative brush lead to the case to bypass the
> internal regulator. There's a tab meant to be shorted with a
> screwdriver for testing. Look for a D-shaped hole in the back of the
> case. Perhaps you could break out a bit of the case, and fish a wire
> with a spade connector through the hole and onto that tab .
>

If he does that (short neg to ground to bypass the internal regulator), he
has to be sure to disconnect the smaller diode bridge from the positive
brush and hook up his home-made regulator to the positive brush. Otherwise
he just has the field directly across the output and it will not be
controllable.

Like what you did here...
> FWIW, mine is a 27si (larger case but otherwise typical old Delco)
> originally 12V,100A). I have it on a 24V system, alternator speed
> about 4800, fixed output about 75A for manually controlled bulk
> charging. I removed the diode trio and resistor, and bypassed the
> regulator by routing new wires directly from the brush terminals.

daestrom

wmbjk

2005-11-15, 8:21 pm

On Tue, 15 Nov 2005 20:42:15 GMT, "daestrom"
<daestrom@NO_SPAM_HEREtwcny.rr.com> wrote:
[color=darkred]
>If he does that (short neg to ground to bypass the internal regulator), he
>has to be sure to disconnect the smaller diode bridge from the positive
>brush and hook up his home-made regulator to the positive brush. Otherwise
>he just has the field directly across the output and it will not be
>controllable.
>
>Like what you did here...

Yep, brain fade on my part, second time recently.... only last week I
accidentally ran my rig without switching the cooling fan on. It
worked just fine for about 90 seconds before the diodes blew. I was
using a surplus potted diode bridge that was mounted on a piece of
aluminum. I always wondered how marginal that might be. :-) Today I
modified a heat sink out of a HD alternator. It had been the negative
side of a 12 diode bridge (I think the other half may have been the
alternator case itself), with 6 paralleled (70A) diodes. I cut it in
half and swapped out three of the diodes. Mounted it at the end of an
arm using the finest 1/4" plywood. ;-)
http://www.citlink.net/~wmbjk/image...natorbridge.jpg I just
tested it, I can hold my finger right at a diode for several seconds,
which I think can be considered cool running. Best of all, now I have
easily replaceable diodes, a feature which hopefully won't be needed
again. I think I might wire up the fan power fail-safe with the field
switch as well.

Have you seen these? http://www.hydrogenappliances.com/powerpmas.html
http://www.hydrogenappliances.com/pmacurves.html

Wayne

Tony Wesley

2005-11-17, 1:21 am

wmbjk wrote:
> Yep, brain fade on my part, second time recently.... only last week I
> accidentally ran my rig without switching the cooling fan on. [...]

I'm predicting that someone will use your post to claim that you don't
design things, your numbers don't add up, that you have to rebuild them
when they don't work right the first time, and you don't have two days
autonomy.

Hopefully, I'm wrong.

wmbjk

2005-11-17, 11:21 am

On 16 Nov 2005 21:15:43 -0800, "Tony Wesley" <tonywesley@gmail.com>
wrote:

>wmbjk wrote:
>
>I'm predicting that someone will use your post to claim that you don't
>design things, your numbers don't add up, that you have to rebuild them
>when they don't work right the first time, and you don't have two days
>autonomy.
>
>Hopefully, I'm wrong.

Heck, I'm looking forward to it! Here are some links for him:
http://tinyurl.com/9hzol ($5 digital camera which oft times sells for
99 cents) and http://tinyurl.com/755x8 (free photo hosting). Now he
can include in his criticisms some pictures of *his* "designs". I'd
like to see that old plug-in rheostat and "other control" (whatever
that is). I'm thinking that compared to his bodges, my generator
control panel might as well be a NASA production. ;-)

Then again, let's not forget that you're probably one of the 50% of
group members who are all actually a single "supertroll". :-)

Wayne

LinkBot