Was wanting to build a Alternator powered Inverter set up.
Goal- to provide 3000 watts emergency backup power and semi portable power to run power tools etc. This will also be a test mule for propane engine controls. Want to develop a portable power source that gets great fuel efficiency.
Parts-
Engine- 212cc Honda clone rated at 7hp (Will convert to propane, more than likely will knock it down to 5.5 to 6 hp)
ALT. 4800 series LN, 200amp, 14 volt.
Battery- 27d deepcycle 12 volt 100AH. Battery is only in the system to to provide uninterrupted power and to help meet the demand of a surge.
Inverter- none yet need about 3000 watts ( fridge, microwave, TV, LED lights, Laptop)
Drive will be serpentine belt. 4.2 inch utterpower pulley on motor and approx 2.5 inch pulley on the alt. I can't remember the exact size, but it maybe as small as 2.3 or so. Anyway my ratio should end up be 1.7-to 1.80, somewhere in that range.
Question i have the 14 volt 200 amp LN 4800 series J mount ( sorry can't remember exact model, it's in FL now and I'm in GA).
I was wondering If i might be better off to spend the $150 or so and get the 110-555 and another battery go with a 24 volt system, Like Bob has ingenuously and graciously described in his white paper. The extra cost would about double my project budget. ( another battery, New alt, balmar reg)
My original plan was to use a stepper motor and microprocessor to control the engines governor speed. So that it would run slow when load was low, and then speed up when load was higher. I was also hoping that I could open the governor up quick enough to avoid the engine from stalling or straining under load.
The hold up for me going to the 110-555 and 24 volt setup is is the Balmer regulator. I think it's possible that I maybe able to mimic some of it's functions with a microprocessor , kind of build my own. Problem is if I start experimenting with $150 alternators and fry them i could have bought the balmar. Is there a cheaper regulator? would not any external regulator work, or I'm I missing something?Also another deepcycle battery makes the unit heavier and harder to move.
welcome aboard
your mount in question is referred to as a j180 mount
the 2.3 serp pulley is a common truck pulley and is about as small as you can get away with in my opinion, it generally is an 8 groove.
if efficiency is the ultimate goal i would forgo the use of the leece neville, it will be only a bit over 50% at full load, and it is doubtful you will get anywhere near max output with 5.5 hp, in my opinion
its no secret that i love the 555 series alternator
there are alternative regulators to the balmar
sterling makes one, for less than half the price, and i am thinking that they have one in the 100 dollar range that is compatible with what i have done in the white paper.
search ebay for "sterling regulator"
the important thing is the need for remote sense capability, and you will also need a 12volt isolated supply for the field, if you want to do the 24volt thing as described in the white paper.
the need for highly accurate speed control is unnecessary, so that makes building a microprocessor controlled speed system much easier in my opinion.
hope this helps
bob g
Thanks Bob. I was hoping you would help me. :)
I pretty sure my pulley is 2.3, after looking at what is offered. That gives me a max ALT shaft speed of about 6500RPM. Or I could buy a larger 2. 9 pulley and give my self a little more torque at the shaft. Then my maximum shaft speed would only be about 5200 rpm.
What RPM range I'm I looking at for the 555 running 28 volts? I'm guessing 2500 ALT shaft RPM and up.
So would think the 1.43 ratio or 1.8 ratio would be better?
i personally would target about 3500rpm or so, alternator rpm
this will give you some room to play with, and good cooling.
bob g
Thanks. tell me if I have this right. The 3500 rpm would be the min speed?
So with a 1.43 ratio. and varying the speed with load based on amps.
Engine speed Alt Shaft speed
Low load 2500 rpm 3575 rpm
Med load 3000 rpm 4290 rpm
Max load 3600 rpm 5148 rpm
According to my Math, if the Alt is 70% efficient and the inverter is 90%, I should be able to run a continuous 2500 watts 120 AC
So I should need about 110 amps at 28 volts from the 110-55.
So my last question is can I get 110 amps at ALT shaft 5148 rpm ?
the 555 will deliver 130amps hot, at 28.8vdc at 4800rpm, however that is about as hard
as i would drive it,,
i would have no problem doing so for short durations, maybe 10 minutes on and backed off, that is why i decided to stick to 100amps max at 28.8vdc at the 4800rpm.
under which conditions the unit is thermally stable and can run continuously at that level.
now having said that, i also need to note this
because my alternator mounting is such that i have to spin it backwards, which is counter to the directional fan, it might well be the unit could do the 130amps continuous if i were to turn it around and/or spin it a bit faster.
fwiw
bob g
My only concern with a variable RPM genset such as you propose, is that I think the engine controller needs to be able to talk to the charge controller to limit the application of field to allow the engine time to ramp up RPM after sudden load increases... If it dosn't I could see the charge controller/regulator sensing the drop in DC voltage from a sudden high load, quickly putting the engine behind the power curve at low RPM.
At the very least, a programmable charge controller/regulator set to apply field increases at a rate slow enough that the engine controller can increase the engine RPM to deal with a load increas.
Thanks for the replies. They are much appreciated.
Ronmar, I share your fear , but i believe my algorithm will be able to deal with it.
My plan is to build the complete unit without the rpm controller.
I will then set the engine speed to 2500 rpm and load the inverter until the engine rpm starts to drop. Then taking note of that load in amps, I will multiply it by 80%. I will repeat this measurement at several rpms, more than likely every 200 rpm. This will allow me to build a 2D map based on amps consumed and engine rpm. At each rpm the motor will run at a 80% load. if the load is greater, then it will instantly go to the appropriate rpm for the amp load detected.
This way i hope to always be running at 80% throttle, no matter what rpm. ( unless the load is less at the lowest rpm) I believe this will work because the governor on the engine controls engine speed, not throttle.
This should give me the best fuel economy and also give me a 20% reserve on engine power, plus the 100ah 24 volt battery pack. also remember the engine speed controller will be asking for more engine rpm as soon as it detects an increase in amp draw.
I think it will work, but this is why I wanted to ask. This thread has already vastly improved my design and been a great help.
If doesn't work then I will try to incorporate your idea.
EDITED to clarify my idea.
With the engine controller monitoring the current output, that might work out. With the batteries in there to buffer the loading of the alternator, the current draw should increase before the voltage drops so far that it triggers a massive response from the voltage regulator to increase alternator output/engine load... But without some control/delay of the voltage reg on the alternator, it still might outpace engine capacity...
Sounds like an interesting project...
Quote from: Ronmar on October 16, 2012, 03:37:28 PM
My only concern with a variable RPM genset such as you propose, is that I think the engine controller needs to be able to talk to the charge controller to limit the application of field to allow the engine time to ramp up RPM after sudden load increases... If it dosn't I could see the charge controller/regulator sensing the drop in DC voltage from a sudden high load, quickly putting the engine behind the power curve at low RPM.
At the very least, a programmable charge controller/regulator set to apply field increases at a rate slow enough that the engine controller can increase the engine RPM to deal with a load increas.
Ronmar & Madlabs,
This is the exact approach that i took with my variable speed perky-cat dc charger, which has become an inverter-generator as well. http://www.microcogen.info/index.php?topic=709.0 (http://www.microcogen.info/index.php?topic=709.0) I am using the Balmar MC614, and Precession Governor. I have a 720 amp hr bank at 12 volt, and a Pure Sine 2800watt Magnum inverter. When acting as inverter gen I can draw 2800 PS AC from the inverter and still put 50 amps charge into the bats at 14.5 volts.
Myine is three speed #1. start up at 1000 rpm no power to the field, after warm up #2. Governor ramps up to 1800 rpm, turns the field on, and shorts the bat temp sensor(small engine mode on the Balmar) The Balmar ramps up the field after a time delay the governor ramps to #3. 2500 rpm opens the small engine mode switch and the gen is at full output.
Then after bulk charge a Voltage/current sense switch, when 14.8 and less then 80 amps current, shorts the SME mode, and drops the governor back to 1800.
Any time a large load is seen a time delay relay detects the voltage drop when a large load is applied to the inverter, and shorts the SME mode 3 sec time delay again and sends a signal to ramp the governor up to 2500 rpm again.
Lloyd
You have a good idea, I just not sure how I can incorporate it in my design. I will do more research. :)
Quote from: Ronmar on October 16, 2012, 07:04:54 PM
With the engine controller monitoring the current output, that might work out. With the batteries in there to buffer the loading of the alternator, the current draw should increase before the voltage drops so far that it triggers a massive response from the voltage regulator to increase alternator output/engine load... But without some control/delay of the voltage reg on the alternator, it still might outpace engine capacity...
Sounds like an interesting project...
So I have been thinking. Tell me how this would work.
I have my system as described above. Except I've decided that I need a tach input so i know exactly how fast the engine is spinning. This will also keep my stepper motor in calibration.
My little genverter is humming along at 2500 rpm engine speed, when suddenly the fridge desides to kick on.
My microproccesser detects it because it's monitoring the amps drawn by the inverter.
it then
1. Tells the engine to increase RPM to the speed corresponding to the amps draw as mapped in my above post.
2. Cuts the connection to the 12v excitation source. The inverter is now drawing 100% from the 100AH 24 volt battery pack.
3. Waits until engine reaches desired rpm.
4 reconnects excitation source.
This what I fear now, is the surge caused by the compressor starting will be almost over by the time the motor gets going and the ALT starts producing again.
So the amp draw will drop. And the motor will have reve'd up for no reason, wasting fuel.
Just trying to think of a smart way to do this. i also have the ability to PWM, so i could reduce the cycle on the excitation circuit. But I fear it may confuse the Voltage reg if I do that.
Now I just think I need to build the voltage reg in with my motor speed controller.
I can figure out to program it to do what I want, just need help figuring out what is the ideal solution.
I think completely cutting out the field excitation might be a problem. I think modulating the excitation with PWM via your controller would be the ideal. Then you could set high and low excitation limits to suit the alternator and the available torque. I am guessing that is how the newer programmable charge controllers are dealing with controlling the field. You could then control the rate at which it is applied in concert with your RPM increases. You would then have full control of engine speed and load.
In answer to your problem scenario, you could program in some delay so short transient loads are handeled by the battery with the engine making up the battery energy used over a longer period after the transient has passed, then the controller responds with an engine and field increase after the longer duration load increase is determined.
you might consider a current transformer to monitor the load line
the micro could be also monitor it with a buffer circuit adc converter or whatever
and if it sees an increase in load that is longer than a typical motor start (say over 10 seconds) then the coding could direct an increase in engine speed first and then an increase in field current last, so that the engine is never overwhelmed or called upon to increase either speed or field current for a short term motor start type load.
you should probably integrate a CT in the scheme anyway, they are so good at monitoring the load, and actually you could implement them on individual circuits
each of which might have very different power level requirements, such as a well pump, refrigeration, or perhaps a furnace blower.
so of which are not only must service type loads but are also don't interrupt loads once they are started. things like furnace blowers must remain powered once called upon by the temp switch to start, otherwise if they shutdown prematurely the furnace goes into overheat and the overheat safety switch has to take over the job of shutting down the furnace, not something you want to rely on in a good design.
i like CT's and their use in control scheme with microcontrollers for this sort of thing.
they greatly simplify problems and allow much more flexibility in my opinion.
actually they can be used as a logic level device, rather than an analog to digital device. if you use them on each circuit, and you know what is on that circuit, you really don't need to have digital level info to work with, simply knowing if the load is on or off is good enough. a micro CT coupled with a discrete component or two, a transistor and you got a simple input to the micro, low and there is no load, high and there is a load on the line.
fwiw
bob g
Quote from: Ronmar on October 17, 2012, 09:40:36 PM
I think completely cutting out the field excitation might be a problem. I think modulating the excitation with PWM via your controller would be the ideal. Then you could set high and low excitation limits to suit the alternator and the available torque. I am guessing that is how the newer programmable charge controllers are dealing with controlling the field. You could then control the rate at which it is applied in concert with your RPM increases. You would then have full control of engine speed and load.
In answer to your problem scenario, you could program in some delay so short transient loads are handeled by the battery with the engine making up the battery energy used over a longer period after the transient has passed, then the controller responds with an engine and field increase after the longer duration load increase is determined.
Ronmar,
That begs a question. The Balmar max charge has the ability to regulate the field of 2 alternators with separate fields and a 10 amp per field PWM . Why not use 1 for the alt and 2 for the throttle solenoid? As the field increases the rpm also increases, use time delay for short term transients that the battery can handle.
Lloyd
I have to say I've been very happy with the Sterling regulator. I'm using the PDAR with the 555, and I'm very satisfied with the results. A lot cheaper than the Balmar.
As to the current sensor, have you seen the Pololu 30A sensor ( http://www.hacktronics.com/Sensors/Current-Sensor-30-Amp/flypage.tpl.html (http://www.hacktronics.com/Sensors/Current-Sensor-30-Amp/flypage.tpl.html)? It may be too small for your application, but they work very well with Arduino's and other pico-boards.
Pete
I have see the pololu 30amp sensor. I'm using an arduinio clone that is made to be inbedded, The ardunio is a great device, easy to program and work with.
Quote from: scottpeterd on October 18, 2012, 01:55:34 AM
I have to say I've been very happy with the Sterling regulator. I'm using the PDAR with the 555, and I'm very satisfied with the results. A lot cheaper than the Balmar.
As to the current sensor, have you seen the Pololu 30A sensor ( http://www.hacktronics.com/Sensors/Current-Sensor-30-Amp/flypage.tpl.html (http://www.hacktronics.com/Sensors/Current-Sensor-30-Amp/flypage.tpl.html)? It may be too small for your application, but they work very well with Arduino's and other pico-boards.
Pete
If you are using a uC to control the RPM, why not just make your own field controller? Then it can behave however you like. That's what I do, mine can read volts, output amps and battery temp.. I am working on a new one that will also keep track of fuel volume consumed. I will eventually add rpm control as well, but that is more complex than field control and adds a new dimension to the field controller.
Jonathan
Quote from: Lloyd on October 18, 2012, 01:21:46 AM
Ronmar,
That begs a question. The Balmar max charge has the ability to regulate the field of 2 alternators with separate fields and a 10 amp per field PWM . Why not use 1 for the alt and 2 for the throttle solenoid? As the field increases the rpm also increases, use time delay for short term transients that the battery can handle.
Lloyd
Well I would guess that the controller you are talking about probably applies those 2 PWM channels based on feedback from 2 separate alternators. IE, if it ups the duty cycle on alternator 1, it is expecting to increase or maintain the voltage output of alternator 1. That might be usefull, but for engine control, I think you would need some voltage feedback from the engine that the controller could respond to and manage RPM...
I thought about it, but I'm not sure exactly how they limit voltage and current, kinda new to power generation. I get that they are PWM the excitation current, not sure how you can control voltage and current independently.
Can you help ?:)
Quote from: Mad_Labs on October 18, 2012, 09:05:23 AM
If you are using a uC to control the RPM, why not just make your own field controller? Then it can behave however you like. That's what I do, mine can read volts, output amps and battery temp.. I am working on a new one that will also keep track of fuel volume consumed. I will eventually add rpm control as well, but that is more complex than field control and adds a new dimension to the field controller.
Jonathan
Quote from: Ronmar on October 18, 2012, 03:13:32 PM
Quote from: Lloyd on October 18, 2012, 01:21:46 AM
Ronmar,
That begs a question. The Balmar max charge has the ability to regulate the field of 2 alternators with separate fields and a 10 amp per field PWM . Why not use 1 for the alt and 2 for the throttle solenoid? As the field increases the rpm also increases, use time delay for short term transients that the battery can handle.
Lloyd
Well I would guess that the controller you are talking about probably applies those 2 PWM channels based on feedback from 2 separate alternators. IE, if it ups the duty cycle on alternator 1, it is expecting to increase or maintain the voltage output of alternator 1. That might be usefull, but for engine control, I think you would need some voltage feedback from the engine that the controller could respond to and manage RPM...
The Balmar unit use batter sensey to determine the voltage on the field. It a three stage charge bulk, asorb, float, in bulk the field is running between 75 and 85% on each channel. The idea is the one regulator controls 2 alternators on one engine. That being the case each field is powered the same so if one is connected to the alt and the other to a PWM acuator, as the battery sense request to increase the field on alt one, it would also increase the the field on channel 2 which if connected to the PWM throttle actuator then it would increase the speed.
The Balmer is rated at 10 amps per channel, and most PWM actuators are rated at 8 amps. I was just having a thought...don't know if it would work.
Lloyd
That is an interesting idea Lloyd. For that matter, PW is PW and you could use a single channel charge controller. A sensor looking at the field drive of an alternator could sense the duty cycle of the PWM driving the alternator and could drive a throttle servo that actuates the engine governed speed control lever. A PWM servo may even be an off the shelf item, but there would be no delay between throttle and field control, so the linkage to the servo would have to be setup so the engine power curve is always ahead of the alternator load. But you might be able to make it span a larger RPM range more easilly like the honda EU sets do, from high idle to full rpm based on load... You might have difficulty configuring it for short transient loads, unless you can program a delay into the controller... Interesting idea...
VW, you basically control amperage with voltage and I guess vis versa. When charging a battery, the greater the alternator voltage in relation to the battery voltage, the greater the current flow into the battery. So a typical auto alternator control looks at the battery voltage and adjusts the field to deliver a particular output voltage at the battery. If the battery is discharged, it's voltage will be low and the controller will boost output to deliver the current necessary to maintain it's regulated voltage. As the battery becomes charged, this becomes easier and easier for the alternator to accomplish so the field decreases as state of charge increases, untill the alternator is delivering just a trickle of current to maintain it's commanded voltage and float the battery at full charge. Switch on the headlights the alternator sees the voltage drop from the applied load and boosts field and current output to maintain voltage. The extra current this delivers powers the load while still floating the battery at full charge... The electrical system dosn't monitor current, but it delivers varying ammounts of it to maintain a set voltage.
A controller that senses current could vary the voltage output of the alternator to deliver a particular current.
A programmable controller basically has several voltages it can switch between depending upon the need. A high voltage to bulk charge a really dead battery and get the charge process started. A slightly lower voltage for absorption charging which delivers current for the middle/majority of a charge cycle at a more efficient and less harmfull level for the battery. Once the battery approaches full charge, then it switches to a lower regulated voltage for floating the cells at full charge. This is known as 3 stage charging. There are other methods as well. A fancy controller may also have a very high voltage for cell equalization and mixing the electrolyte...
Beamer,
So, that alt has a field coil. The harder you drive the field coil, the more the alt puts out. I drive the field coil with the uC's PWM and a NPN Darlignton transistor. So the uC reads the amps and the volts and in my case the battery temp. In my case, the uC drives the field coil until A: I reach 14.4 volts or B: the current from the alt is more than 150 amps. Once it reaches 14.4V it lowers the PWM on the field coil, maintaining 14.4 volts, until my current drops below 20 amps. Then it shuts itself off. So, if you can control RPM (much trickier than alt output) you can easily make your own charge controller.
HTH,
Jonathan
I realized I didn't quite answer your question, how do control BOTH volts and amps.
Actually, you can't control them seperately, they are ALWAYS tied together. Let's take a battery at 50% charge for an example. You start the engine and engage the alt. The controller/alt will try to raise the voltage to 14.4V. But it can't because the battery is so discharged. So the controller runs the alt as hard as it can, or if a smart controller, as hard as is safe for the alt. The alt continues to dump as much as it can and the battery voltage starts to rise. When it reaches 14.4V, the controller starts to back off the PWM on the field coil and the amps start to drop. When the amps drop low enough (how low depends on the size of the battery bank) the charge is complete. In my case with my controller, it then politely shuts the engine off. :)
So, amps and volts are always inextricably tied together and adjusting the field coil PWM duty controls both at the same time.
HTH and doesn't make it more confusing. :)
Jonathan
Interesting project, Jonathan. You might be at liberty to get even better fuel efficiency by tailoring your controller software to match your power usage- such as no point in starting the engine for a tiny load and battery fully charged. Perhaps a basic state of charge calculation would help do that, and isn't too tough if you know amps in, amps out, and adjust for Peukert effect and your batteries relative charge efficiency at various state of charge. I do this for my 120V battery bank but it's all fixed point, table look up, specific for my Walmart/Johnson Controls 110AH marine batteries.
Wet lead acid batteries are more like 50-60% efficient in charge acceptance for the last 10% of charge, though of course for long life you'd like to keep them fully charged most of the time. This is based on pretty reliable Sandia Labs data which I posted here before. It's a design conflict that has to be worked out depending on your application and usage pattern. Topping off the batteries at least once a day might be a good compromise. AGM's should have less of this effect, and Lithium batteries likely the best regarding this effect. Both AGM and Lithium batteries cost compared to wet lead acid rules them out for many practical applications.
BobG has posted some efficiency strategy ideas for off grid generator charging of wet lead batteries which I thought were promising, too.
Best Wishes,
Bruce
Thnaks, that helps. :)
Quote from: Mad_Labs on October 19, 2012, 07:36:56 AM
I realized I didn't quite answer your question, how do control BOTH volts and amps.
Actually, you can't control them seperately, they are ALWAYS tied together. Let's take a battery at 50% charge for an example. You start the engine and engage the alt. The controller/alt will try to raise the voltage to 14.4V. But it can't because the battery is so discharged. So the controller runs the alt as hard as it can, or if a smart controller, as hard as is safe for the alt. The alt continues to dump as much as it can and the battery voltage starts to rise. When it reaches 14.4V, the controller starts to back off the PWM on the field coil and the amps start to drop. When the amps drop low enough (how low depends on the size of the battery bank) the charge is complete. In my case with my controller, it then politely shuts the engine off. :)
So, amps and volts are always inextricably tied together and adjusting the field coil PWM duty controls both at the same time.
HTH and doesn't make it more confusing. :)
Jonathan
Thanks to all :)
Quote from: Ronmar on October 18, 2012, 10:32:16 PM
That is an interesting idea Lloyd. For that matter, PW is PW and you could use a single channel charge controller. A sensor looking at the field drive of an alternator could sense the duty cycle of the PWM driving the alternator and could drive a throttle servo that actuates the engine governed speed control lever. A PWM servo may even be an off the shelf item, but there would be no delay between throttle and field control, so the linkage to the servo would have to be setup so the engine power curve is always ahead of the alternator load. But you might be able to make it span a larger RPM range more easilly like the honda EU sets do, from high idle to full rpm based on load... You might have difficulty configuring it for short transient loads, unless you can program a delay into the controller... Interesting idea...
VW, you basically control amperage with voltage and I guess vis versa. When charging a battery, the greater the alternator voltage in relation to the battery voltage, the greater the current flow into the battery. So a typical auto alternator control looks at the battery voltage and adjusts the field to deliver a particular output voltage at the battery. If the battery is discharged, it's voltage will be low and the controller will boost output to deliver the current necessary to maintain it's regulated voltage. As the battery becomes charged, this becomes easier and easier for the alternator to accomplish so the field decreases as state of charge increases, untill the alternator is delivering just a trickle of current to maintain it's commanded voltage and float the battery at full charge. Switch on the headlights the alternator sees the voltage drop from the applied load and boosts field and current output to maintain voltage. The extra current this delivers powers the load while still floating the battery at full charge... The electrical system dosn't monitor current, but it delivers varying ammounts of it to maintain a set voltage.
A controller that senses current could vary the voltage output of the alternator to deliver a particular current.
A programmable controller basically has several voltages it can switch between depending upon the need. A high voltage to bulk charge a really dead battery and get the charge process started. A slightly lower voltage for absorption charging which delivers current for the middle/majority of a charge cycle at a more efficient and less harmfull level for the battery. Once the battery approaches full charge, then it switches to a lower regulated voltage for floating the cells at full charge. This is known as 3 stage charging. There are other methods as well. A fancy controller may also have a very high voltage for cell equalization and mixing the electrolyte...
it took me a long time to figure out how to control both amps and volts with a controller
balmar does it in a fashion as do others
from what i have learned from them, this is what i have in mind for a custom built controller
i would sense both current and voltage, current going from the alternator to the battery
and by extension the load, and voltage at the battery.
then i would set the microcontroller to upon startup do the following
1. give me a delay before energizing the field, 60 seconds or so, programmable of course so it could be anything
2. ramp up slowly the field current, while monitoring the current leaving the alternator to the battery/load, ramp it up to whatever i program the amps to max out at
this might be limited by size of battery bank, or the max i want to load the alternator or both. while also monitoring the voltage as a second control handle, making sure the voltage stays below my maximum setpoint for whatever part of the charging i have in place (bulk, absorption or float)
3. have the micro periodically recalculate the voltage/amps etc to reset max amperage and voltage, based on what inputs are reading and also what my code lookup table parameters might be.
4. complete bulk, move to absorption, and finally float
5. have the capability to switch to a manual equalization, or a programmed equalization.
i think we can do both volts and amps, although it is difficult to do charging a battery, it is easier to do when the system is also providing power to a load.
bob g
reays in a can.
if anybody is interested.
http://www.kussmaul.com/control%20relays%20index.html (http://www.kussmaul.com/control%20relays%20index.html)
lloyd