Jump to content
shameless

Regulator, rectifier or both?

Recommended Posts

Hi Guys:

Checking the charging system on my 312-8.

found this after chasing wires.

It was attached to the fan shroud with two bolts.

post-5770-0-97228600-1325600704_thumb.jp

I did a bit of reading and got out my meter to measure.

I grounded one of the bolts directly to the battery to be sure of a good ground.

The outside terminal supplied AC voltage in the 30 volt range at mid throttle and more at higher rpm.

The center terminal connected to a violet colored wire.

There was no output voltage.

Did some more reading and figured the unit was N.F.G.

Built this out of 6 amp diodes from The Source (Canadian version of Radio Shack).for $6.

post-5770-0-93734200-1325601227_thumb.jp

Connected it to the original flat plug at the engine shroud.

The violet colored center DC wire dissapears behind the shroud and reappears at the back of the motor and connects to a molex where the wires change colors.

I cut the wire at this point to ensure that if my creation did not work, I could contain the damage.

I took my measurements at the cut point.

It works and changes the AC from the stator to DC.

The DC output voltage is almost as high as the input voltage when checked with no load.

Will I need to regulate this voltage or will the load (100 watt fornt lights) lower the voltage?

Share this post


Link to post
Share on other sites

I was just about to dig deeper into some of saveoldirons posts for the very same reason...

What you have made will work, but does not limit the maximum charge/battery voltage to around 14.4V so unless you have low revs or a big load there is a danger you will over charge the battery.

A correctly working regulator will rectify, but also cut out when the battery is charged - around 14V (most cars charge between 13.8V - 14.4V)

just found one of the useful posts...

mark

Share this post


Link to post
Share on other sites

hey shameless,thats pretty ingenius,so where about in ontario are you from,im near orangeville,were getting quite a few ontario guys here,it wasnt very far across the border to west bend from sarnia

Share this post


Link to post
Share on other sites

hey can whlvr:

Ingenious...nah............ :)

I copied the design from a youtube video.

It works fine, I hope someone can answer my question about regulating the voltage and no load versus load.

I live in Hamilton.

Share this post


Link to post
Share on other sites

it's a certainty if you have a stator over 3A (most k series kohlers are min 15A) then charging a battery unregulated will cause it to overcharge, it will boil the electrolyte, create excess gas and fail - usually catastrophically !!!

The only way to save the battery is to dump the excess power - you can either use lights or a big dummy load, i.e. a 2 ohm resistor and a switch. You would have to use this in conjunction with a voltmeter and switch the load in when the voltage is over 14.5V.

Personally I think charging this way is too risky, I'd rather run the battery with no charge circuit and charge offline...

Btw if anyone has managed to remove the silicon potting compound and supply pictures - I'm more than happy to do a replacement board and find modern alternatives. The board I have is totalled as I just couldnt remove it without breaking it into pieces.

thanks

mark

Share this post


Link to post
Share on other sites

Hey meadowfield:

Thanks for the response.

When I measured such high DC voltage I knew enough to ask rather than hook it up to the battery and carry on.

I built the rectifier cause it cost $6 rather than the $50 plus I saw on line for the proper replacement part.

Obviously the stock unit uses more than diodes like I did.

Does anyone know what else is in that "brick" other that rectifying diodes? :eusa-think: :eusa-think: :eusa-think:

Share this post


Link to post
Share on other sites

there is a 14V zener diode - used to monitor the voltage, and then an SCR (voltage controlled switch) shunts the stator windings.

I am keen to copy the Kohler variant, if I can get a picture of top and bottom side with the goop removed! :banghead:

mark

Share this post


Link to post
Share on other sites

Mark,

The moderator, Casual Observer, opened an old thread that may help you. It is PDF format in the following post. Hopefully all the pics are intact.

It is maybe the 6th entry down.

Share this post


Link to post
Share on other sites

phew !!! thought I was going mad, I knew I have read an extensive article on the teardown of the reg.

thanks !

Share this post


Link to post
Share on other sites

Managed to make my way through teardowns from SOI and wheelhorseman (not heard from him for some time now).

It's pointless trying to copy the older typr R/R as you would have to make the transformer - which I could unwind one and then get some samples made.

So I think firstly I will do my own PCB copying the newer kind and try it...

Heres the circuit redrawn:

rr.jpg

Looks fairly sensible, I should be able to test it by the weekend when my SCR samples arrive.

If it works then I will do a modern postage stamp variant - that will then fit both the old style case and the new miniature ones.

Then look to add the additional circuitry to put a charge light and low battery light. Both would be very useful in my opinion - I did this on my sons briggs powered westwood.

w11_7.JPG

mark

  • Like 1

Share this post


Link to post
Share on other sites

Nice work....I have to tell you I have no idea how these electrical components work. I am an electronics dunce. I hope you do figure it out though. The indicators for charging and low battery are a good idea.

Share this post


Link to post
Share on other sites

I'm actually tempted on my C-105 where the wiring loom has bit the dust to make a custom panel with the fuses, reg/rect all self contained to simplify things.

I do rather like the idea of modern charge/batt lights...

Share this post


Link to post
Share on other sites

Hi Mark

I'm back online again, just needed a break from 'horsing around' to get lots of other jobs that were building up done.

Glad the input I made has been of use and if you have any questions about the circuit feel free to ask.

Here's a few things that I can't recall if I covered in the original SOI thread on the old site.

The heatsink doesn't need to be the size of the beast that sits behind the ammeter on a C series, it's a leftover from the original shunt regulators where excess power was disposed of as heat to regulate voltage. A heatsink is needed for the SCR but nothing like the original size. ( I tried it without but it got rather hot, I seem to recall burning my thumb on it!)

The original circuit doesn't include any adjustment but bearing in mind that individual electronic components aren't made to a 100% accuracy what the designers did was allow access to one of the resistors after the unit was assembled by mounting it on the 'wrong' side of the circuit board.

If the assembled unit regulated voltage fell outside of the accepted limits then the resistor could be changed before the board was sealed with rubbery gloop. The output voltage is set by resistors R3 and R4 in the above circuit.

Even with a reasonable knowledge of electronics this type of circuit is pretty difficult to fathom out because it is not a linear regulator but deals with pulses. SOI gave a pretty good description of the principle but it's the actual values of the components that can be difficult to follow.

The regulator can only be tested with it connected to the alternator and a battery - the pulses it sends to the battery peak at more than 14 volts but the battery acts as a huge reservoir so you can only test its function by running up the engine with a voltmeter connected to a good and well charged battery. I say well charged otherwise you'd have to wait ages for it to reach regulation point. 14 volts is ideal but mine does about 14.4 which is still okay and I didn't bother making any adjustments.

The repaired board has worked fine ever since.

Andy

Share this post


Link to post
Share on other sites

Just looked up the original thread from the old site - the value of R3 was 1K7 not 1K8 as shown in the circuit above.

I mentioned in the original thread that in reality it is made up of two resistors, one of which tweaks the output.

On mine there was a 1K5 in series with a 220R hence 1K7 give or take.

It's the 220R that was mounted on the wrong side of the board and could be altered for a slightly higher or lower value (e.g. 270R or 150R) if needed to adjust the regulator.

Using a 1K8 resistor could result in the voltage output being either too high or too low unless you get lucky.

Share this post


Link to post
Share on other sites

Andy,

thanks for the replies, due to the nature of the SCR and other tolerances the circuit has to be tweaked. I just copied your values - I was just pulling the circuit together. I would probably place a 1K2 and add a 470R pot in series for the real one. Though I do have a cunning plan which maintains the same rectification, but saves 4 components and make the regulator 1% accurate...

mark

Share this post


Link to post
Share on other sites

Great stuff! - I'll follow this thread and your development work with much interest so do keep us posted on progress.

I spent quite a bit of time trying to get my head round designing a circuit to build for the D-200 which came with just a bridge rectifier! I decided against in the end and got a genuine RR from a D series being parted in the States.

I got bogged down by the way that, once fired into conduction, an SCR latches on for the rest of the half cycle, which in the end led me to think that if there was an easier (lower component) and more accurate way it would have been used already.

Re the 'Kohler' circuit - even a high quality cermet pot will be subject to probs of dampness if not in a sealed unit and possibly also vibration. Although padding out a fixed resistor is 'clumsy' it has potentially greater long term stability / reliability which is why, I think, it's rare to find adjustment pots in this type of application. (Just a thought for what it's worth)

Anyway its great to see you've taken up the challenge and as I said, keep us posted.

Next an ECU for the Kohler?

Andy

Share this post


Link to post
Share on other sites

Andy,

I thought long and hard over why they only half wave rectify and use an scr. The answer is easy...

40 years ago a 10 amp diode was hard to come by, using 4 in a bridge would have been expensive madness. Also then the transistor reg element would have been chunky and probably expensive too !

The simplicity of the scr circuit is just brilliant. Providing the battery voltage is below the set threshold, the scr switches on and always switches on until the positive cycle is finished whereby even if the gate were still high the diode is reverse biased and it switches off.

The only losses in the circuit at 15A is the diode drop of the scr. Anything more complex could actually be less efficient!!

I design power supplies for consumer equipment so I can come up with more effective ways sometimes... The biggest improvement I can make is to replace the diode/transistor with a shunt regulator. These are 2% accurate and that way there would be no output voltage trimming at all.

Before I waffle on anymore and boast of a triumph I need to build and test the circuit... I need my workshop back first, but that's another story!

Mark

Share this post


Link to post
Share on other sites

an update...

I was forced to remove the rectifier from my GT14 at the weekend - it was looking rather sorry for itself !

IMG-20120211-00161-1.jpg

one of the terminals had clearly cracked at some point and arced until the board was burnt away. A repair had been fashioned which had worked as it was charging, but clearly needed remedial action.

So following chucks old thread (SOI) I tried to pry the PCB out after cutter the leads on the SCR...

the result was I cracked the PCB in exactly the same places :)

reg.jpg

So with the circuit fresh in my hand I did a cad drawing so I can replace the PCB and effect a reasonable repair !

reg2.jpg

It's pretty well an exact clone of the original using the original parts in the same locations so will look perfect.

I don't like the transformer design so will also build a pcb reflecting the newer design. Andy - any chance you can photograph the topside of your PCB for me? or is it now fully assembled?

Failing that I'll do my own interpretation with a cutout to allow a more standard TO-220 device to be screwed to the heatsink.

mark

Share this post


Link to post
Share on other sites

Consider adding a ground run between J4 and J5. The original design calls for connection thru the heat sink assembly. This may lessen the chance of one loose / corroded ground screw on the circuit board causing regulator malfunction.

Share this post


Link to post
Share on other sites

I will add the ground link - I didn't have time to move it around before etching...

heres the result - but for the terminals and the rectifier a prefectly good clone of the original

IMG_1393.jpg

will keep you posted on progress of it working !

Share this post


Link to post
Share on other sites

Good work so far.

Share this post


Link to post
Share on other sites

Hi Mark

Re my RR - it's in the tractor at the moment but shouldn't be too difficult to remove and open up for a photo - I'll see what I can do.

Andy

Share this post


Link to post
Share on other sites

The transformer has always been the final mystery. My experience with gate signals thru transformers has been strictly for use in "fail safe" circuits where gate signal would be disconnected from the DC control circuit in event of a circuit fault. I can understand the absence of a capacitor coupler due to failure from environmental effects.

AC theory has never been my strongest area of expertise. I have not progressed in the reverse engineering of the board to the point of understanding if the transformer acted soley as a gate drive signal or added a phase shift in the gate signal to allow triggering of the SCR on the TRAILING edge of the AC charge pulse. Without my differential module on the scope working, I may never know the answer to that question.

With the seemingly endless power hungry additions being added to tractors, I dropped study of the SCR (with significant "on resistance values") in favor of the modern MOSFET technology with milliohm on resistances. I'm sure the inductive reactance and DC resistive value of the stator will be the limiting factors in the MOSFET designs.here. PWM charger operation and self contained desulfation circuitry can be included with additional of very little real estate or $$$ added. That's the direction in which I will be investing my ever dwindling design time.

Congratulations on your PCB construction. Folks who wish to maintain originality of the charge circuit will certainly have a viable option. I have original spare transformers available if the need ever arises.

Glad you folks are around taking an interest in this phase of the hobby and to bounce these ideas around.

Share this post


Link to post
Share on other sites

I'm just playing to satisfy my curiousity... I prefer not to use the transformer and drive the gate of the SCR directly using a shunt regulator - it's very low component count !

The larger problems with mosfets is protecting them from spurious and reverse spikes. I would assume use of a full wave bridge also, but then a 20A bridge takes up a lot of space. I can see why even after all these years a reg/rectifier is still pretty well the same construction and technology has not won !

I can design what would be more efficient, but more costly and would have many more parts... maybe a job for another day :)

Share this post


Link to post
Share on other sites

Let me know what you are considering for conformal coating. The suppliers I investigated were all really high $$$ unless you purchased gallon quantities. Thanks.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


  • Recently Browsing   0 members

    No registered users viewing this page.

×