Jump to content
JUSS10

custom rectifier/regulator for my 654

Recommended Posts

JUSS10

So I’ve decided that my 654 is in need of a rewire job. My grandfather had been using it for the last year and I have noticed that besides wiring it has some charging issues. This has the H60 engine in it and uses the half wave rectifier with no regulator in it. I’ve found that with two diodes it appears to over charge the battery and if I run it on just one, the battery doesn’t seem to charge well enough. My thought was why not build a new custom rectifier regulator with a fixed voltage output of 13.75 volts to the battery. Since this is only a 3 amp charging system I was able to use an off the shelf bridge rectifier from radio shack (rated for 4 amps) and then ordered a LM338 regulator which is adjustable and capable of 5 amps. Add a few resistors and a fuse and we are in business.

 

Below is my schematic.

 

The main reason I am posting this is to ask if anyone thinks this could do harm to the charging system in the motor. Is there any other protection I should add other than the 3 amp fuse I plan to use between the regulator and the battery? Not sure how many electrical gurus we have out there. I’m hoping SOI will chime in on this.

 

Justin

 

RECTIFIERREGULATOR_zps57b066e4.jpg

  • Like 1

Share this post


Link to post
Share on other sites
 
meadowfield

you will firstly need to add a reasonable amount of capacitance on the output of the bridge rectifier. The voltage rating must be adequate to cover full revs with the battery disconnected. So if the ac voltage is 30V then suggest 50V rated capacitors. At 3A charge current the ripple current through the capacitors will be high, so they will need to be low ESR, long lifetime parts.

 

At this stage you now have a smooth (ish) DC voltage into the regulator. LM338 is fine for this (I have a few chargers with this part fitted). I would add a reverse biased 1A diode (1N4001) across in and out on the regulator for some additional protection. Plus a couple of 100nf multi layer ceramics on the input and output to reduce HF noise.

 

I'd say for 3A then you stand a chance of it working.

 

By now you start to see why most RR units use an SCR as it's a much simpler circuit. I've thought about this in detail and had many discussions with SOI and Wheeledhorseman offline.... :D

 

mark

Share this post


Link to post
Share on other sites
 
JUSS10

what do you mean by SCR? Is it essential to smooth out the voltage going into the regulator? i had most of the parts sitting around here at work.What value caps would you put behind the rectifier? 

Share this post


Link to post
Share on other sites
 
JUSS10

I should also ask, is there a better way to do this? 

Share this post


Link to post
Share on other sites
 
meadowfield

It's not a good idea to put rectified AC through a regulator - It will not regulate properly. You have to put some filtering/smoothing ahead of the LM338.

 

Guessing you will need minimum 2x2200uF 50V parts with low esr (<0.030 ohms), but that's purely a finger in the air !

 

 

re: better way, the SCR looks clunky and old fashioned - but I've rebuilt a few RR units and re engineered them and the SCR is a very elegant method. The slight downside is it is half wave, but that doesn't make a massive difference to charge current. The simple bit is, that it's basically a diode that can be disabled very easily when the battery reaches 13.8v (ish).

 

Here's the newer Kohler RR circuit

 

rr.jpg 

 

mark

Share this post


Link to post
Share on other sites
 
JUSS10

interesting, I don't quite understand the circuit as i see one of the AC lines going to the battery positive but I'll take your word for it. I'll order up the components and give this one a try!

 

Thanks!

 

Justin

Share this post


Link to post
Share on other sites
 
meadowfield

Before you start, what engine and stator? This circuit replaces the kohler reg, for a 15 amp stator. By virtue of the two diodes your circuit sounds like a techy v/h70. I have a v70 in my lawn ranger.....

Wheeledhorseman has done an scr circuit to replace that simple diode circuit. Let soi and Andy chime in first....

Mark

Share this post


Link to post
Share on other sites
 
JUSS10

The engine is a techy H60, not sure on the stator. It has two AC lines coming from it. Thats really all i know.

Share this post


Link to post
Share on other sites
 
wheeledhorseman

The stator has two coils - the center connection between them is earthed making it a simple two phase alternator.

This way Tecumseh acieved full wave rectification using just two diodes. As you know, there's no regulation and in the manual, Tecumseh acknowledge that overcharging of the battery can occur - their solution, remove one of the diodes.

If you build the Kohler circuit you'll only be able to use one of the coils (one of the two wires that come out from behind the flywheel) The result won't be a lot better than the one diode you have now that doesn't overcharge. In fact looking at the Kohler ciircuit, the fact that there's a centre tap earth means, I think, you cant use it as it stands on a Tecumseh alternator.

 

 

My aim is to improve on this when I get my A-800 project up and running again soon. The circuit I sketched down but haven't built yet looks like this.

 

tecrr.jpg

 Sorry about the quality it was drawn on the back of a bit of card in the workshop. Without getting too technical the diodes are replaced by SCRs. A reference voltage for both halves of the circuit is provided by a zener diode. The SCRs cease to turn on and conduct when the reference voltage is reached so a 14volt zener diode would limit the charge to 14 volts. Although 14 volt zeners are available I was thinking of using say a 6.8 and 7.5 zener in series (14.3 volts) as the tractor isn't going to be run for long periods so it's important that the battery drain from starting it is replaced fairly quickly.

 

 I haven't got round to building a prototype board yet but  I've run this circuit on an emulator with 330 ohm resistors and it seems to work fine (in theory at least). The SCRs will probably be BT151 because they're easy to obtain here, the diodes will be whatever I have to hand as long as they can handle say 100 volts as a reasonable safety margin.

 

Best not steal Mark's thunder, but he can talk you through replacing the zener to make it either adjustable or just accurate to the voltage of your choice rather than the zener voltage.

 

I'd be interested in Chuck's view if he latches on to this thread.

 

Andy

Edited by wheeledhorseman

Share this post


Link to post
Share on other sites
 
JUSS10

ok, I think i understand how these SCRs work now. Thanks for the circuit Andy. I may order up the parts and give your schematic a try. When i looked up the BT151 on digikey.com two different ones came up, slightly different specs, does that matter? The two diodes in each half, do those need to be anything specific or just diodes that have the 100V rating?

 

Thanks!~

 

Justin

Share this post


Link to post
Share on other sites
 
wheeledhorseman

The lowest spec BT151 is the '500' i.e. 500 volt which is rather more than adequate but the higher voltage ones will do equally well.

I'll be using 1N400x series diodes because I've got lots of them lying around. The 1N4001 is rated at 50 volts which would probably be ok but I'll dig out 1N4002s to be sure (100v) though could well be 1N4003 or even 1N4004 it just depends whats in the drawer.

The SCRs will need heatsinks - I did a quick calculation and I recon they could be generating up to 5 watts - worst case.

Unfortunately the mounting tab of a BT151 is connected to the anode which means they can't both be mounted on the same heatsink unless insulating kits are used. I'll probably use two separate heatsinks on mine. It's all a bit experimental as the power disipation may not be as bad in practice so it will be a matter of seeing how small the heat sink can be before the SCRs get too hot.

I'd use 1 watt resistors I think but I didn't finish all of the calculations.

 

As I said, I haven't got round to building a prototype yet and the circuit has only been tested on a circuit emulator.

 

Andy

Edited by wheeledhorseman

Share this post


Link to post
Share on other sites
 
Save Old Iron

Hey Justin, et al

 

Great post. You have received very sound advice so far.

 

Yes, an input filter capacitor is needed for three terminal voltage regulators. A 4700uf  35v might be a good place to start.

 

One of the restrictions on using a three terminal regulator is the need to have the voltage on the input of the regulator at least 3 to 4 volts HIGHER than the needed output voltage. To see if your design is feasible with the the stator you have, borrow a 4700uf  (or larger) electrolytic cap rated 35 volts or higher and place it across the output of the current rectifier pc board. Positive (+) capacitor lead goes to the B+ terminal of the rectifier and the negative (-) capacitor lead can go to any chassis ground.

 

Use your multimeter on DC volts to measure the voltage present on the capacitor. If the voltage is higher than 17 volts, the 3 terminal regulators will most likely work. I do believe there are some newer regulator chips that can work with a difference of 1 volt or under from the input to the output. I will have to research that statement a little further but I believe low-differential regs are available. In 5 amps ? - not sure.

 

If you wish to put a small breadboard together for a feasibility test, push the B+ output of the current rectifier board onto the test board, The 4700uf cap will be across the + and - from the rectifier board, then to the LM reg and output of the LM reg can be filtered by a smaller (10 - 100uf) electrolytic cap - then out to the battery.

 

Give it a shot Justin, this is how you learn. Trust me, my pile of failures is way bigger than my pile of successful designs.

 

Never be afraid of failures.

Share this post


Link to post
Share on other sites
 
Save Old Iron

looks like you will need about 2.2v volts diff from input to output - you will need 16.7 volts on the input cap to the regulator to maintain operation of the regulator under 3 amps of current. That difference increases to about 2.6 volts at 5 amp output from the regulator.

post-1689-0-77305600-1366279349_thumb.gi

Share this post


Link to post
Share on other sites
 
 
 
wheeledhorseman

I'd go with the first (part no. 140417) for the SCR

Feel a bit guilty that I'm not doing the prototype, particularly as some tweaking may be needed, but my tractor wont be up and running for a week or two yet.

A 1.3 watt zener diode should be fine.

I'm pretty sure 1 watt resistors will do but if you want to be sure go for 3 watt.

 

Andy

Share this post


Link to post
Share on other sites
 
Save Old Iron

Justin,

 

another very simple way of solving the overcharging with both HH diodes connected is to to put a third diode in series with the B+ output of the rectifier board . Diodes have a nearly constant .6 to .7 voltage drop across the diode thru the current range you would encounter in a tractor charging system.

 

Simply inserting a 1N5401 diode in series between the battery and the HH rectifier board will knock down a 15 volt overcharge to 14.3 volts. Two diodes in series would drop 1.2 to 1.4 volts off the charge voltage. A 15 volt overcharge would be knocked down to around 13.8 volts. Not nearly as sexy as a custom designed regulator, but very effective in resolving overcharge conditions.

 

 

briggsdualoutputaltreduced_zps350b47fa.g

 

 

You would benefit by getting a faster charge using both diodes on the HH rectifier board but you would limit the output voltage seen by the battery to a more tolerable level. A win on both accounts.

 

I certainly do not want to stifle your effort on the regulator board concept but given the modest charge currents we are dealing with, simple 1N diodes can be used. I thought about doing the exact same experiment with my 3 amp charge system on my 212-5. I ended up placing two 1N5401's in series with the stator output and the maximum charge voltage dropped from 16 down to 14.6.

 

Just a thought.

  • Like 1

Share this post


Link to post
Share on other sites
 
JUSS10

Thanks SOI, I added some of those to an order I am making for the other electrical parts. May give that a try first. We will have to see, i already re wired the tractor with the thought of a regulator/rectifier hidden behind the gas tank so I may have to move the stock rectifier up there or just make a new one on a board with some diodes. can i use the 1N5401 in place of the rectifier diodes that it originally used then place however many behind to get my desired voltage?

Share this post


Link to post
Share on other sites
 
Save Old Iron

Justin, the original 1N1084 were rated for 400 volts but only 1/2 amp - yes, 0.5amp.

 

I'm not sure why the design used a 400 rating but the diodes would not see a reverse voltage higher then the battery voltage plus the peak value of the negative sine wave coming from the stator windings. In other words, when the diode is blocking half the AC wave from the stator, it blocks that peak voltage PLUS the voltage from the battery.

 

Let's guesstimate the peak voltage from the stator is 17 RMS AC which translates into about 50 volts peak which the diode must block from the stator alone. Add to that peak value the DC value of the battery (15 ish). The total voltage the diode must block is 50 + 15 or 65 volts total.

 

If you calculate a safety margin by doubling the 65 volt requirement, a 200 PIV or (P)eak (I)nverse (V)oltage diode should be fine. Correction. I guess I'm showing my age saying PIV, modern term appears to be Vdc or maximum DC blocking voltage.

 

A 3 amp rating should be more than sufficient.

 

So we need a 200 PIV 3 amp diode.  That would be the 1N5402

 

To your question if additional diodes can be mounted to the HH board. Yes, you could solder a few diodes in series and place them into the holders on the existing pc board. The arrangement may be a little messy, but electrically it is no problem. I might start with 3 diodes in series in each leg of the HH board. The 2 additional diodes would knock off about 1.3 volts of charge voltage. You could easily monitor the battery voltage after a long session with the tractor. Target about 14.5 volts on the battery while the engine is running at full speed. If you run the engine at lower than full speed, 2 diodes may be more appropriate. If you do a bit of work with the tractor in colder weather, one diode in each leg would be good. Remember, colder weather requires a higher charge voltage to maintain good battery health and promote longer battery life. Higher charge voltage which could overcharge a battery in the summer is just about right in the winter.

  • Like 1

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

×
×
  • Create New...