Honda Generator Automatic Voltage Regulator

Thread Starter

dmechanic

Joined Jan 14, 2009
4
Hello,
First post here. I've aquired a small Honda portable generator (vintage 1977). The model is EG 1500 (1.5kw output). This generator is equipped with an automatic voltage regulator (avr) that is physically damaged. This particular avr uses the input of 120vac from the stator and apparently rectifies it for use as an output to the brushes to control rotor field current and therefore voltage output. Frequency appears to be regulated by engine rpm and does not appear to be controlled by the avr. The avr assembly is filled with what I believe to be fiberglass resin (or similar) material to seal all the components. This stuff is very difficult to remove and I don't think I can do it without destroying all the internal components. Honda no longer services this regulator and the universal ones are quite expensive. I'd like to recreate a suitable circuit to regulate the generator as a project for my own learning. At a minimum, I would need to rectify 120vac and use the the corresponding dc to drive the rotor field current to maintain 120vac under varying loads. Rotor resistance is 54 ohms. There is also an external cap of 200uF 250WV attached to the original avr. There are 4 wires from the stator for input to the avr (althought I'm not sure all would be required) and the coresponding 2 dc output wires from the avr to the rotor brushes. Any suggestions would be appreciated.
 

jj_alukkas

Joined Jan 8, 2009
753
Ohh Boy.. you want to build a 1.5Kw voltage regulator? Components will be quite expensive. Did you try for spares from any one of the new 1.5Kw models? Maybe that might work. And it would be a lot cheaper for you. check about it before you jump in.
 

vks_foe

Joined Jan 11, 2009
61
hi,
I'm doing a similar project.. i have to design an AVR for a single phase alternator(220v, <10 KVA).The existing AVRs have a feature of underfrequency shut down... i.e. below a specified frequency, which can be adjusted by a potentiometer...
Kindly suggest me a frequecy to voltage converter IC which is low in cost..
 

nagaloo

Joined Jan 27, 2009
29
Ohh Boy.. you want to build a 1.5Kw voltage regulator? Components will be quite expensive. Did you try for spares from any one of the new 1.5Kw models? Maybe that might work. And it would be a lot cheaper for you. check about it before you jump in.
So typically what is in these AVR's that is so expensive that they charge between 200 and 400 dollars for them?

I have 4 Coleman Powermate 5000 watt gens in my shop for repair. There is about 5 or 6 film caps about 10 resistors 2 transistors, some doides of different types, a divice I am told is a switch rated at 400v 10 amp it looks like a scr or voltage regulator (T-220 case) and 2 or 3 electrolitic capacitors.

Just a bunch of junk on a circuit board. Powermate went out of business and my customers are left with a useless machine after only 1 or 2 uses.

I suspose I could draw up the board and try to ID the components but it must be a simple circuit there isnt much to this PCB. Could anyone help trouble shoot it? All 4 have over heated the "switch" device. Or if there is some other way to skin this cat.

If you have any sugestions I can detail this some more.

Stew
 

SgtWookie

Joined Jul 17, 2007
22,230
Stew,
Without a schematic, we won't be much help. It does look like Powermate filed for bankruptcy in March of 2008.

But first, you could simply try to find identifying part numbers on the TO-220 packaged devices. Generating a schematic might not be too difficult seeing as your board has just a couple dozen parts.
 

nagaloo

Joined Jan 27, 2009
29
I will do my best to make a diagram of the board and its components. It may possibly work for different gens. It has exciter ac 2 yellow wires that feed a pair of diodes with a common output which after feeding a filtering capacitor end up on one of the leads of the "switch" and a center tap that feeds one of the brushes to the rotating fields. The middle lead on the Switch" feeds the other brush. It has 2 other wires that feed the output voltage from the main stator to the control circuit. I will also post the "switch" info, the guy at the electronics supplier had trouble finding it on his listings. From what I see there are 2 circuits on the board one is for making dc for the rotating fields and the other is to turn the "switch" on and off in relation to the main stator output. I suspect the control circuit has a problem and thats why the "switch" looks fried.
Stew
 

Thread Starter

dmechanic

Joined Jan 14, 2009
4
This is exactly whats going on with the Honda that I originally posted about. Unfortunately, the resin they used to seal all the components prevents me from seeing any of the components for schematic purposes. Like you, I agree that this component does not seem overly complex and that was the reason for my initial post. The generator I have is also set up the same way with a exciter winding that is rectified and then used for the controlling dc rotor current. It also has two wires from the main stator winding for voltage monitoring. So, 4 wires in (2 from exciter winding, 2 from main stator), 2 wires out (dc for rotor current). My suspicion is that the dcv output for rotor current control would be in the range of 12-85v with max 3A current. Unlike a previous post suggesting this is a big deal, like you I don't believe this is nearly as expensive as some here seem to believe. I also concur that the price of these things is rediculous and for those of us with either discontinued parts or bankrupt manufacturers, we have no choice. I hope you can see enough of your regulator to generate a schematic. Thanks for your post.
 

nagaloo

Joined Jan 27, 2009
29
It turns out I should be able to get a good pic or print of my PCB. Turns out 3 of what I thought were small caps are maybe varisitors. I got all the junk off the board. I think I will unsolder everything and put it in the scanner. Hopefully someone here will be able to improve it or suggest what might be wrong with the circuits. I just found out today that the local Canadian Tire has a bunch of these gens and would like a solution too. Maybe I will get a chance over the weekend. I will take some pics too.
Stew
 

nagaloo

Joined Jan 27, 2009
29
Here are the components on the board:
R1 Dale CPF2 10k 1% TI 0504 I asume this is a 10k
R2 30k 5%
R3 56k 5%
R4 Rd Yel Wh Br Br Could be the reverse Small low watt
R5 Rd Rd Br Br Br Could be the reverse Small low watt
R6 pot 202 23 with a w no bottoms
R7 not used
D5 Glass no #’s
D6 MCC 3 1N4006
D7 MCC 3 1N4006
D11 MCC 3 60S10
D9 MCC 3 60S10
DB2 MCC DB104 Small bridge rect
C1 2.2uf 250v
C2 Had this on it 100B with a bar over 10 and 103 with 2 dots under 10
C3 470uf 200v
Q1 MPS A06 502
Q2 2N6 515 510
Q3 BU931T CC04L W MRC 449 This had over heated
S1 S2 S3 All the same CNR 10D391K These may be varisitors


I didnt get to test any of this stuff yet. I will need to find part #'s to match if these are not "normal" #'s
R4 & 5 I dont know how to decode these ?

I cant get the "attachments" to work for uploading my files? Any help there?
 

SgtWookie

Joined Jul 17, 2007
22,230
re: can't get attachments to work
File format is important. Images are preferred in .png format, as they are not "lossy" like .jpg format. There is a size limit of 100k for .png files.
If the files are too large, they won't upload.
.ZIP format files can also be attached; I think they're limited to 1 megabyte.
 

nagaloo

Joined Jan 27, 2009
29
Front.jpg

Back.jpg

I got it to work If anyone needs a good look at those 5 color band resistors I can email them to you. I am working on hand drawing the PCB with the print over it.
 

nagaloo

Joined Jan 27, 2009
29
re: can't get attachments to work
File format is important. Images are preferred in .png format, as they are not "lossy" like .jpg format. There is a size limit of 100k for .png files.
If the files are too large, they won't upload.
.ZIP format files can also be attached; I think they're limited to 1 megabyte.
Let me know how my jpg pics look to you before I put up a few more
Stew
 

SgtWookie

Joined Jul 17, 2007
22,230
R4 Rd Yel Wh Br Br Could be the reverse Small low watt
The above is 2.49k Ohms, 1% tolerance; if the body is light blue then it is a metal film resistor.
R5 Rd Rd Br Br Br Could be the reverse Small low watt
The above is 2.21k Ohms, 1% tolerance.
R6 pot 202 23 with a w no bottoms
Likely a 2k pot.
D5 Glass no #’s
Hard to say; could be a Zener or a standard diode like a 1N4148/1N914.
D6 MCC 3 1N4006
Standard 1A 800PIV rectifier diode.
D7 MCC 3 1N4006
D11 MCC 3 60S10
D9 MCC 3 60S10
D11 and D9 are unknown at the moment. Sure you have the numbers right?
DB2 MCC DB104 Small bridge rect
1A 400V dip bridge rectifier.
C1 2.2uf 250v
C2 Had this on it 100B with a bar over 10 and 103 with 2 dots under 10[/QUOTE]
Just a guess; 10nF 100V 10% tolerance.
C3 470uf 200v
Q1 MPS A06 502
Probably MPSA06; NPN transistor, Vce=80,Ic=500mA,hFE >= 100 @Ic =10mA,100mA
Q2 2N6 515 510
2N6515, NPN high voltage (250v) transistor
Q3 BU931T CC04L W MRC 449 This had over heated
BU931T is a high-voltage, high power NPN Darlington transistor.
S1 S2 S3 All the same CNR 10D391K These may be varisitors
No impact there.
I didnt get to test any of this stuff yet. I will need to find part #'s to match if these are not "normal" #'s
R4 & 5 I dont know how to decode these ?
Here's an online calculator for 4-, 5- and 6- band resistors:
http://samengstrom.com/nxl/10116/5_band_resistor_color_code_page.en.html
Click the color, and it'll advance to the next band.
 

nagaloo

Joined Jan 27, 2009
29
No there was a heat sink it didn’t have any heat conducting paste on it. It might have dissipated heat better if it was up in the air. But this board is screwed to the inside of the endbell of the generator so there is lots of air flow. I am posting a copy of the PCB with the copper traces marked over the component print. If something doesn’t look right I will have another look. Getting down to my questions if there is enough info. Does this board have a problem or is the transistor just too small? Is there a good site to check how to test some of this stuff ? I have basic skills but no electronics training as such. I am a motor winder/machinist, I have an Industrial Electrical Ticket so I have a understanding of some of it. Is there a good online place to lay out circuits and test them? Something like Electronic Work Bench or a free down load of something like that but simple ?
Stew
In my pic the conections are marked from the gen. 2 blacks are hooked to the main winding output either 120v or 240v the 2 yellow and the blue are the exciter supply with the blue the center tap or common. I will try to check tomorrow on voltages and currents from the gen at normal operating output by using external excitation.
 

Attachments

Last edited:

nagaloo

Joined Jan 27, 2009
29
To the best of my ability I drew out the PCB If something doesn't make sense point it out and I will have a look at the board. The transistors C E B may not be marked in the proper order but they are the same as on the PCB. I did it that way to minimize the lines that had to be crossed over in my drawing. Didn’t get time to test anything yet. As soon as I do I will post if and what was bad or damaged.

Also I tried to delete the pic of the PCB I traced over in marker. I need to repost it with a few repairs. I made a few traces that dont exist.

Stewart
 

Attachments

ifixit

Joined Nov 20, 2008
652
Hi

I have a colman gen as well, still works okay, but I'm glad to follow along with this thread to learn details of the generators design.

Attached is an LTspice schematic of the sketch you did which is more readable for understanding the circuit function.

Q1 is likely under rated for heat dissipation and/or needs more heatsink.

It would be nice to know what the excitor voltage is when the gen is running and what the field winding resistance is also. Measure resistance with gen off.

Perhapes Q1 can be sub'd with a better choice.

Regards, Ifixit
 

Attachments

nagaloo

Joined Jan 27, 2009
29
Attached is an LTspice schematic of the sketch you did which is more readable for understanding the circuit function.
Well it must have taken about an hour to draw it up, but I wasn't very pleased with it. Surprizing where things are actually connected once you trace the PCB out.

I will get some numbers to you later today. I wondered if thats what might be going on with it. There is lots of heat as the board is turning brown. Where did you get the schematic? LTspice? Is that a program?
 

nagaloo

Joined Jan 27, 2009
29
Well I only got so far on testing and checking output on this generator.
Rotating field resistance 28 ohms Read through the brushes.
The following readings were taken from the 120 v output
No excitation 2.2 v
At 20 vdc external excitation 99.7 v
Fed through 2 diodes and a var. resistor using the generators excitation winding 120 vac with 30-31 vdc reading on the brushes. at 1.1 amps
Maxed out with no resistance 142 vac at 82 vdc at the brushes.

Then I got ready to test with a load and bang the connecting rod broke and smashed a hole the size of my fist in the engine case. This machine was destined to die no matter what. It had only 50 hours tops of use for 2 power outages. About 1 1/2 years old.

Tomorrow I move on to machine #2
LOL
Stewart
 

ifixit

Joined Nov 20, 2008
652
What bad luck. That's good data you collected, which gives a good idea of what kind of transistor to use, however, is gen #2 the same make and size? Same control board?

Regards, Ifixit
 
Top