The member photo gallery is now integrated and live!!  All user albums and pictures have been ported from old gallery.


To register send an e-mail to admin@bmwr65.org and provide your location and desired user name.

Author Topic: Diode board replacement  (Read 5161 times)

Offline dogshome

  • Lives in Foothills of Mt. Olympus
  • **
  • Posts: 402
Re: Diode board replacement
« Reply #15 on: January 23, 2021, 01:14:01 PM »
Major FAIL with the regulator. I now know there are NPN and PNP types. Airheads use PNP type that drives +ve to the coil, Fiats and several others use NPN that expect the coil to be connected to +ve and then sink current down to ground. Live and learn. Or is that LIVE! and learn?

There does not appear to be any info as to which is which except to find the alternator it fits to and scrutinise how the brushes are connected. In true Haynes/automotive style (make it really complicated) this isn't immediately obvious.

To be fair, the original one is solidly designed and efficient. Mine is clean and in good condition. A few more ergs might occur with a MOSFET driver and temperature compensation would be nice, but your average Ebay special or 20 year young alternator bolt-on doesn't do that.

13.8V at 2,000 and 12.9V at 1,000V with the lights off is good enough. Apart from a few specialist marine circuits (where a lot of expensive fried batteries stranded in a force 8 on the North Sea might be a problem) I've not found anything to crib.

P.S. That black tape is great stuff. I bet you could stick wet nappies with it :-)
« Last Edit: January 23, 2021, 01:15:55 PM by dogshome »
肉(r?u)包(bāo)子(zi)打(dǎ)狗(gǒu) (meat+bun(2nd and 3rd)+hit+dog)
* Literally: To hit a dog with a meat-bun.:-O

Offline mrclubike

  • Mt. Olympus Resident
  • ****
  • Posts: 1437
  • Jungheinrich Master Tech
Re: Diode board replacement
« Reply #16 on: January 23, 2021, 08:23:06 PM »
Yes their is 2 types of charging circuits
The airhead comes from the factory with whats called a "B" circuit
Regulator controls the B+ going to the field
 The GM Delco I use is an "A" circuit
Controls the B- going to the field coil 

If you go to the online store I posted the link to many of the REG's description will tell you if its A or B
It is easy to isolate the ground on the brush holder if you would like to try it again
I have been wondering if  the A circuit reg in general comes up quicker or if it just the higher set point that the GM reg has that causes it to perform better   
1982 R65 running tubeless Snowflakes
2004 R1150R

Offline dogshome

  • Lives in Foothills of Mt. Olympus
  • **
  • Posts: 402
Re: Diode board replacement
« Reply #17 on: January 24, 2021, 02:07:56 AM »
Thank you 😎

Typical auto, why call it a and b when it's npn and pnp?

Npn do work better than their pnp equivalents, but I don't think you would notice on something relatively sluggish like an alternator field. How hard and how fast the circuit is designed to turn on might be though. if it is on the high gain and minimum lag side, you might notice faster response and a little instability or hysteresis. Lower gain and or larger lag might mean better stability, even some slow drifting.

I think the lag introduced by the capacitors in these circuit is is a bit arbitrary. If it doesn't oscillate, then it will be close enough for jazz. Without a scope, lots of revving up and testing I can't see me messing. Someone knows.......
肉(r?u)包(bāo)子(zi)打(dǎ)狗(gǒu) (meat+bun(2nd and 3rd)+hit+dog)
* Literally: To hit a dog with a meat-bun.:-O

Offline dogshome

  • Lives in Foothills of Mt. Olympus
  • **
  • Posts: 402
Re: Diode board replacement
« Reply #18 on: January 24, 2021, 09:35:25 AM »
The caps vs resistor sizes in the BMW regulator suggests it's not slugged very much i.e. fast response. MOSFETS theoretically give better dissipation when being switched, but not at the sub 500Hz we are looking at. So I did a quick test on the Cargo NPN one I have here:

Only running a lamp load of ~300mA not 3A expected. VCE is 30mV. It could be a BJT of FET in there. The BD244 in the BMW regulator is typically about the same at this current. Rising to around 200mV at 3A (bang in it's saturated design range). So I don't think a MOSFET with it's effectively 0 RDS (and 0 volt drop) is going to make any difference.

What I did see with this Cargo unit it that it doesn't turn off until 14.6V ~ 14.7V. Getting marginally higher as it warmed up (eek!). So I think that is why your system is performing 'better' - Not sure I want it that high even with an AGM battery. Cold weather and trips to work probably OK, a run to the coast in summer, maybe not.


[Yes my meter is ancient, it is still very accurate and safe. Yes it is a Maplin special. I bought a new one for work recently, simply to avoid questions about it's suitability on LV busbars and drives. Both have proper HRC fuses, over voltage and overcurrent protection and are shrouded internally. Hasn't killed me in 40 years of use....]
肉(r?u)包(bāo)子(zi)打(dǎ)狗(gǒu) (meat+bun(2nd and 3rd)+hit+dog)
* Literally: To hit a dog with a meat-bun.:-O

Offline dogshome

  • Lives in Foothills of Mt. Olympus
  • **
  • Posts: 402
Re: Diode board replacement
« Reply #19 on: January 24, 2021, 10:22:19 AM »
MOSFET, N Channel (NPN style).

PT001
VDSS=150V ID=30A RDS (ON)=0.049Ω
Surface Mount package
Model                 Brand      package        Polarity    VDSS(V)   ID(A)   RDS(ON)(Ω)   VGS(th)(V) 
PT001                 MORE    SOT-252        N CHANNEL   150        30          0.049            ?

Couple of transistors, flyback diode and a zener. Just tiny. Not a great RDS hence volt drop I measured. Very cheap and similar to a decent BJT in terms of performance here. Too small to reverse engineer with a P channel MOSFET!
« Last Edit: January 24, 2021, 10:29:54 AM by dogshome »
肉(r?u)包(bāo)子(zi)打(dǎ)狗(gǒu) (meat+bun(2nd and 3rd)+hit+dog)
* Literally: To hit a dog with a meat-bun.:-O

Offline mrclubike

  • Mt. Olympus Resident
  • ****
  • Posts: 1437
  • Jungheinrich Master Tech
Re: Diode board replacement
« Reply #20 on: January 24, 2021, 08:27:29 PM »
The "A" and "B" terminology predates transistors
Not sure if the terms PNP and NPN were used before the advent of transistors   
A and B was used to describe old generator systems also and was very important to know what you had when diagnosing them
   

 The Optima battery likes the 14.7 volts that the Delco reg tops out at
Got 6 years out of the first one and it did not fail catastrophically. It just got to weak to get good spark and cranking speed

I do know if you put the Delco reg under the front cover with the diode board.   It gets hot and drops to 13.8volts out put.
My diode board mounting plate was made to have it there but I didn't like the 13.8 charging voltage it caused
It was pretty neat because I had only 1 wire for the charging circuit coming out of the front cover  :'( 
1982 R65 running tubeless Snowflakes
2004 R1150R

Offline Tony Smith

  • Mt. Olympus Resident
  • ****
  • Posts: 2331
  • Graduate, Wallace and Gromit School of Engineering
Re: Diode board replacement
« Reply #21 on: January 24, 2021, 09:29:25 PM »
Whilst I am content to leave the airhead charging system pretty much stock, events in VW world have forced me to investigate alternatives.

The alternator:
VW aircooled Type 4 engines (1700, 1800 & 2000cc) used a hybridised version of the Bosch "universal" 55amp alternator.
The hybridisation consists of:
1 Unique outer housing
2 No fan on shaft
3 A rear cover with orifice to allow blast cooling air bled off from main fain housing
4 Unique diode board to fit in the minimum diameter housing
5 Unique rotor (however oter bosch universal rotors can be made to fit by parting off a small amount the rear end of the rotor shaft

The problem
Rotors NLA (but see above)
Housings NLA (not really a poblem as every VW shop has lots of dead ones
Diode board NLA (which is the major problem).

Commentary
VW / Bosch did briefly market a 70amp alternator to assist those vehicles equipped with a petrol air heater and therefore potentially parked up with heater and fans running. These 70amp alternators are near mythical, Imhave held a dead one in my hands but they have an even longer list of NLA parts than the 55amp.

There are severe space constraints and it is not a practical proposition to fit an alternate alternator, the housing was cunningly designed to fit the available space. A larger diameter could only be accomodated by firstly removing the exhaust system, but the length of any replacement alternator cannot be much more than 10mm longer than stock in order to fit in the available space and take a feed of cooling air. The blast cooling air is essential as the alternator sits only a few millimetres above the exhaust heater boxes.

I have tried abgout a dozen alternatives, the only ones that fit the space, eg Kubota and older Honda Civic, have less output than stock and would requeire custom tine ware to take the cooling air blast at thhe rear of the housing and pass it forward.  The Delco alternator from Robinson 44 and 22 helicopters could (I think) be made to fit with a bit of machining on the housing, they have the advantage of being set up to receive blast air cooling from the rear, they are also available in 70/90 amp.  The problem with these is that they are a controlled item and the only way I will ever get one without paying $1,200 outright purchase for a new one, or $700 for a time-ex one is to rob one from a wreck before the aviation wreckers get their hands on it. In short - not going to happen.

Solutions
The easiest solution given that the original diodes are simply pressed in, wouldbe to obtain new diodes and replace the failed ones. Easier said than done, not making much progress on that one. Any ideas?

The second easiest solution would be to find a source for replacement diode boards. There is a firm in the Netherlands that claims to have them - Euro 29 for the board is reasonable, but their postage impost of Euro 99 means I will never find out if they are for real.  Another firm in Russia claims to have them but they have a minimum 10 buy policy.

Lastly I should mention that you can buy quiite cheaply whole replacement alternators from Poland, these come with no manufacturer details on them whatsoever, they also do not have a stella reputation for reliability. I have fitted one to our Kombi after firstly replacing the bearings with ones that the manufacturer was sufficiently proud of to put their name on. It runs with a 14.2 volt output irregardless of what regulator I fit to it which is a concern. Thinking back to when I had it apart to fit bearings I *think* it may have a regulator module as part of the brush assembly. Needless to say I am not keen to head off on a long trip at the mercy of that alternator.

So, I've reached what I think is the best solution - fit an exernal diode module. What do you think?


One last comment. I the course of hunting a replacement regulator late last year I discovered that the excellent Bosch (Australia) RE-55 and RE-57 regulators are no longer available. I bought an adjustable one from Motobins which I have not used as in the meantime I found these devices from China. Seems to work very well although of course it remains to be seen if they are reliable. The cost was a whole $11 each!

1978 R100RS| 1981 R100RS (JPS) | 1984 R65 | 1992 KLE500 | 2002 R1150GSA |

Offline dogshome

  • Lives in Foothills of Mt. Olympus
  • **
  • Posts: 402
Re: Diode board replacement
« Reply #22 on: January 24, 2021, 09:44:05 PM »
My first mini had a dynamo. I seem to think one end of the field was always earthed, not positive tied. Alternators only came about due to the silicon diode. Power transistors were available not long after and npn would have been the choice with the a circuit. B circuit maybe coming from the convention of switching positive using a mechanical regulator on a grounded dynamo field, carried over to alternators?

Apologies if I appear argumentative, this stuff interests me, which is why I really enjoyed working on steel mill drives. Magamps and mercury arc rectifiers....
« Last Edit: January 24, 2021, 10:05:32 PM by dogshome »
肉(r?u)包(bāo)子(zi)打(dǎ)狗(gǒu) (meat+bun(2nd and 3rd)+hit+dog)
* Literally: To hit a dog with a meat-bun.:-O

Offline dogshome

  • Lives in Foothills of Mt. Olympus
  • **
  • Posts: 402
Re: Diode board replacement
« Reply #23 on: January 24, 2021, 09:58:42 PM »
Re VW: An external diode board should work fine. Saabs kill alternator diodes with regularity too. Probably a design fault rather than heat on these though.

I saw you regulator on the other thread. I can confidently ask now, is it a or b circuit?
« Last Edit: January 24, 2021, 10:06:31 PM by dogshome »
肉(r?u)包(bāo)子(zi)打(dǎ)狗(gǒu) (meat+bun(2nd and 3rd)+hit+dog)
* Literally: To hit a dog with a meat-bun.:-O

Offline Tony Smith

  • Mt. Olympus Resident
  • ****
  • Posts: 2331
  • Graduate, Wallace and Gromit School of Engineering
Re: Diode board replacement
« Reply #24 on: January 25, 2021, 04:31:46 PM »
Negative ground is all I know.
1978 R100RS| 1981 R100RS (JPS) | 1984 R65 | 1992 KLE500 | 2002 R1150GSA |

Offline dogshome

  • Lives in Foothills of Mt. Olympus
  • **
  • Posts: 402
Re: Diode board replacement
« Reply #25 on: January 25, 2021, 05:35:28 PM »
I think this means what you said   :2vrolijk_08:

"F" is the battery +.And the other terminal is ground( motor and battery must be connected to the ground together when using).

Yoda translation from mandarin.
« Last Edit: January 25, 2021, 05:36:59 PM by dogshome »
肉(r?u)包(bāo)子(zi)打(dǎ)狗(gǒu) (meat+bun(2nd and 3rd)+hit+dog)
* Literally: To hit a dog with a meat-bun.:-O

Offline dogshome

  • Lives in Foothills of Mt. Olympus
  • **
  • Posts: 402
Re: Diode board replacement
« Reply #26 on: February 05, 2021, 02:24:14 PM »
OK this regulator is definitely a BJT type PNP and b circuit for standard airheads.

The volt drop is quite significant and the switching range is baggy, almost like a linear amp around trip voltage. In comparison, the Cargo MOSFET type switched very quickly and did not get warm. This one will get warm and might dither and get warmer still. 1.3+ V is significant when only dealing with 12V, so although very useable with a slightly higher than standard voltage (fine for AGM batteries) I think a home-brew is in order. Of course it depends on what voltage the field coil is designed to saturate.
« Last Edit: February 05, 2021, 02:43:44 PM by dogshome »
肉(r?u)包(bāo)子(zi)打(dǎ)狗(gǒu) (meat+bun(2nd and 3rd)+hit+dog)
* Literally: To hit a dog with a meat-bun.:-O

Offline dogshome

  • Lives in Foothills of Mt. Olympus
  • **
  • Posts: 402
Re: Diode board replacement
« Reply #27 on: February 07, 2021, 08:15:49 AM »
I spotted this one before and it appeals to me as a simple / good design. Mosfets superseded, but plenty of alternatives and I have all the other parts in various biscuit tins apart from those :-) The 30 year old caps I found I might pass on and maybe the equally vintage tantalum one... Transistors don't have a sell by date do they? Not like my mums box of 15 year old Assam tea I found? Voyager is still ticking, so I guess not.

http://chemelec.com/Projects/Alternator/Reg-SCH-P-Mosfet.png
肉(r?u)包(bāo)子(zi)打(dǎ)狗(gǒu) (meat+bun(2nd and 3rd)+hit+dog)
* Literally: To hit a dog with a meat-bun.:-O

Offline dogshome

  • Lives in Foothills of Mt. Olympus
  • **
  • Posts: 402
Re: Diode board replacement
« Reply #28 on: February 11, 2021, 01:45:11 PM »
A cunning plan....  :naughty:
肉(r?u)包(bāo)子(zi)打(dǎ)狗(gǒu) (meat+bun(2nd and 3rd)+hit+dog)
* Literally: To hit a dog with a meat-bun.:-O

Offline mrclubike

  • Mt. Olympus Resident
  • ****
  • Posts: 1437
  • Jungheinrich Master Tech
Re: Diode board replacement
« Reply #29 on: February 11, 2021, 08:39:44 PM »
I have been trying to install a inductive pick up in my beancan to get the full benefits of the HEI ignition module
The winter this year has been too wet and icy so this is keeping me occupied
I am using a pick up coil from a Mitsubishi forklift
Mounting the pick up on the original breaker plate was the easy part
Getting a custom made reluctor to have near perfect timing between the 2 cylinders is proving to be difficult without a proper mill or lathe
All i have is a drill press with a compound slide table on it  :furious3:
I have the 2 cylinders firing within 2 degs but I would like to get it it closer

The old  Honda CB owners do it with GM Cavalier coils and that is what I am using
http://gpzweb.s3-website-us-east-1.amazonaws.com/Ignition/CBheiModCavCoil.html
But they have it easy because they already have an inductive pick up all they have to do is mount all the parts 
« Last Edit: February 11, 2021, 08:51:12 PM by mrclubike »
1982 R65 running tubeless Snowflakes
2004 R1150R