TheSamba.com :: View topic

Mon Sep 24, 2018 4:45 am

I bought the same of the pictures above; manufacturer seems to be JPGroup.

I do not think there are other 6V solid state regulators with this aspect.

There are other types, one from cloversystems, one other from http://dynamoregulators.com/ that seems to be well designed, but are not into an original-looking case.

Mon Sep 24, 2018 7:48 am

NOS Bosch is the way to go.

Mon Sep 24, 2018 8:06 am

NOS on electrical parts isn't always the best choice .....

Mon Sep 24, 2018 8:20 am

Just like anything else. I've seen nothing wrong with NOS regulators. Care to comment about that?

Mon Sep 24, 2018 11:21 pm

I bought a NOS Bosch regulator (german) some years ago ... after about 10 months it didn't work anymore.

Mon Sep 24, 2018 11:23 pm

heimlich wrote: Just like anything else. I've seen nothing wrong with NOS regulators. Care to comment about that?

Probabily is not clear how exactly a 6V regulator works.

All NOS regulators, due to the fact that they have stayed for decades without being used, have the "full charge" contacts with a light layer of oxidation so this full charge is never activated.

However the "half charge" state is still working so both a non-expert customer and a non-expert seller are sure that this NOS regulator is OK because the battery is (slowely and weakly!) charged and the whole electrical system is working, but at its minimum capability, turn signals are slow, etc.

Moreover, NOS regulators have been designed when cars where used mainly for long trips.

Now a NOS regulator in a car used in an urban context is simply an error because its design is made to waste almost 10A when engine is idling and generator not charging... a slow agony for a 6volt battery. 10A plus the power for auto-starter, coil, stop lights, etc.

So a NOS regulator can be an option at two condition:

1) to know someone able to check its funcionality and to restore all its working parts

2) to be aware that a car with 6V electro-mechanical regulator should not be continuosly used into urban traffic.

Fri Jan 10, 2020 1:11 pm

I wanted a electronic regulator on my new to me 1960 Beetle with it's 36 Horse engine and small dia 6 volt generator and began to research it.
Found out a lot of interesting data about my 6 volt generator and discovered it has more umpth built in than the stock mechanical voltage regulator was producing.
I have been using a 6 volt optima AGM battery which I have to be really careful with as the 6 volt generator can easily over charge it.

Some of the issues that came up with electronic voltage regulators are that they need power before they can turn on. So I had to add a 6 volt wire from the coil to my design otherwise it was difficult to start it charging without revving the engine and it would always drop out at idle.
It soon became painfully clear that a VoltMeter and a Over Voltage Alarm was needed in the instrument panel. I also came up with a Amp Meter which I left located on the Voltage Regulator along with a Voltage readout so it can be easily adjusted for both Voltage Output and Current Limit.

The trick with the design I wanted turned out to be to pulse the field to achieve the desired Voltage and Current Regulation.

The Regulator Design is pretty simple and inexpensive as parts go:
A Motorola HDL Power MOSFET $1.00 ea
two LM311P Voltage Comparators $3.45 for 10 ea
1 Barrier Diode: STPS4015CW $1.46 ea (Output Schottky Diode)
Other components including the secret to more efficient operation of the generator a (flywheel diode) that I put across the Field and the heat sink for the barrier diode, I got from an old computer power supply for nothing.
It worked first try!
I have done a lot of testing perfected it to a point building 3 prototypes. It is capable of making the VW Beetle 6 volt generator put out 10 Volts charge loaded with the 1000 Amp Agm battery. (Note this would not be a good idea though to do very much!) (I am more concerned about the Optoma AGM battery than the Generator as it has no problem at all doing this. It easily will put out over 55 amps with the headlights on and the voltage set above 7 volts.) I was able to get the voltage change at the regulator output when the headlights are turned on down to .2 volts which is caused mostly by the Voltage drop across the (Optput Schottky Diode) and 100 amp Current Shunt that provides both the ability to have a digital current display and it provides the needed sampling voltage for current regulation.

Fri Jan 10, 2020 1:16 pm

That's great Dan! Looking forward to seeing a pic of your packaging.

Fri Jan 10, 2020 1:35 pm

Well I am sure some folks would consider that another drawback of my homemade Electronic 6 Volt Voltage Regulator. I was going to package it into a old VW mechanical Regulator housing but Have decided not to do that just because it's such a wild looking thing hanging on the 36 horse. The only real problem I see with putting into a Mechanical Regulator Housing is the heat dissipation needed may not be adequate.
Really though it's just the (Output Schottky Diode) that needs the heat sink.

Fri Jan 10, 2020 5:23 pm

I designed a regulator for a 1926 FORD MODEL T WOODY WAGON.
The diode was mounted to an aluminum block machined with the matching diameter of the generator case. With thermal compound it used the generator case as a heatsink.

You could also mount the diode on a heat sink in a remote location.

Parallel two or more FET's could replace the diode. A few milliohms would keep the heat sink requirements low.

Fri Jan 10, 2020 6:30 pm

Notice the shunts, 10 turns around a pushrod of #16 copper wire. about 4 m ohms (4X10 to the -3) Ohms total that way good for about 60 amps. The copper gets hot! Kind of needs to be not inside the housing. Higher current shunts could solve that and they make little tiny surface mount ones. (Here: is how it is on my prototype 2), Probably just could do the same thing with the mechanical reg cover? On the mechanical reg base there is the original output for B+ that could be used for D+. And as you say, The barrier diode and shunts could be external.

Fri Jan 10, 2020 7:58 pm

Well, I do have a 180 watt regulator on my 160 watt (thin slot) generator. It doesn't seem to care.

If you look at the schematic of a mechanical regulator you will see both voltage and current windings on the relays. The interesting thing is that magnetic field strength of a coil is based on "amp turns." So a few turns taking full output current can create as much magnetic strength, needed to switch a relay, as many turns of fine windings. Roughly, the many fine turns measures voltage (how you get current through the high resistance) and the heavy few turns measures current (with almost no loss of voltage.)

You can use or old 6 volt Bug as a city commuter car without any problems, unless you start hanging more electrical devices onto the small electrical system. You cannot add a big stereo or large driving lights on a 6 volt Bug because there is just not the power available. A solid state regulator will get it done if you are just over the edge, but you really cannot sustain a load over 180 watts on a Bug with a 6 volt generator. That is only around 30 amps at 6 volts, the same amount of power as a 12 volts at 15 amps. A common 55 amp Bug alternator puts out over 3.5 times as much power!

Thanks to 2Pack here is VW's explanation of the voltage regulator.

Sat Jan 11, 2020 9:59 am

At the end of the eighties I built this solid state 6 volt regulator.
Still works in my VW Bug 1200 / 6 V.
Sorry only in German, but not that hard to understand.

Regards.

Sun Jan 12, 2020 11:29 am

Great to see those schematics and yes probably should not exceed the 30-40 amps continuously.
I gained a new respect for the Mechanical Regulators when I tried to make an electronic one start charging without a Relay to ground.
It always amazes me just how precisely electromagnets were used in the Mechanical voltage regulators.
The Design I came up with mimics the mechanical regulator in a couple of ways in that it does both Voltage Control and Current Regulation.
A lot of advances have been made in transistors since the BJT. Now Power MOSFET's are capable of Hundreds of Amps and can be biased to conduct down to just a few milli ohms without hardy any current flow needed to gate them on and off.
The lower voltage power Mosfets do have a linear bias range usually between 1 and 2 volts on the gate where they can be used as a linear device. However they heat up there so this range is typically avoided in power control circuits in favor of timed or controlled switching on and off to control current.
The more voltage on the gate of a MOSFET at least the (common N channel Power Mosfet) that is best suited to operate between the (Field Winding and Ground of a Negative Ground 6 Volt Dynamo): The more voltage to the Gate the lower it's resistance to current flowing +Drain to -Source. (The 6Volt Regulator connects the Source to ground and the Drain to DF) ( Because of this lack of current flow at startup I could not figure out a way to start the generator charging without help from the switched 6 volt circuit on the ignition coil. This is one reason to stick with the Mechanical Voltage Regulators on these old 6 volt generators.
However:
Precise voltage and current regulation, lower field coil temperature, Adjustable Voltage and Current and improved (Power Factors are achieved with on/off switching especially when a flywheel diode is used across the field).
Another drawback to my design is it can fail full on if there is an open circuit condition in the Voltage Adjust circuit. For this reason I installed an Over Voltage Alarm in my beetle as well as a voltage display on the dash.
Thanks for your interest?

My Schematic:

Mon Jan 13, 2020 1:24 am

I developed 5-6 different schemas of solid state 6V regulators so I think I can define me as a little "expert" :)

The BJT schematic posted by Ovally has a big issue: it simply does not run at low rpms, the field excitation is on only when generator reaches enough rpms, plus the stability should be very low for the presence of a zener diode. And the external TO3 transistor is very prone to bad contact due to vibrations and thermal schocks.

The MOSFET schematic posted by Danvw suffers from the fact that the circuit has no connection with the B+ terminal, so it is on only when generator is ready, no "start charge" action can be performed.

My solution uses:
- Low vg mosfet (IRL...) that are ALWAYS ON because their VG is connected via a resistor to the B+ terminal; current sink is minimal. This is the philosophy of the BOSCH 12V electronic regulator

- operational amplifiers are LM x93.. that have "open collector" outputss, so you can parallel the voltage and current regulation

- the Mosfet and the shottky diodes (in TO220 case) are screwed on the floor of a dismantled original regulator, so the whole chassis, and the generator on which the regulator is mounted, work to dissipate heat

- the current regulation (that I removed) is via a Hall-effect chip that outputs 2,5mV/A and easy to check with an operational amplifier, without the use of strange resistor

- the output is very stable, up to 7,2V, because the voltage check is performed AFTER the diode, directly on the B+ terminal, and diode/wiring voltage drops are compensated by voltage regulation

- the whole regulator fits perfectly in the old regulator case, and the connections are only the ones of an electromechanics regulator, so the installation and wiring are the same of an old regulator, making the substitution very easy. This is, for me, a useful facility: I always have a perfectly working spare regulator that I can install when the electronic regulator fails, this helped me every time my selfmade regulator failed or burned.

Finally, I mounted on my dashboard:
- a LED between D+ and DF terminals; this gives my a continuos control on how the regulator is working
- a digital volt/ampere instrument that feeds voltage and current directly from the battery (~10euros)
Those indications are always useful, also when using mechanical regulators, they permit to appreciate how field voltage is managed by the particular regulator is used, and look at difference and strange behaviours.

Mon Jan 13, 2020 1:17 pm

herbie1200 wrote: "My solution uses: - Low vg mosfet (IRL...) that are ALWAYS ON because their VG is connected via a resistor to the B+ terminal; current sink is minimal. This is the philosophy of the BOSCH 12V electronic regulator"

"- the current regulation (that I removed) is via a Hall-effect chip that outputs 2,5mV/A and easy to check with an operational amplifier, without the use of strange resistor"

"- the output is very stable, up to 7,2V, because the voltage check is performed AFTER the diode, directly on the B+ terminal, and diode/wiring voltage drops are compensated by voltage regulation"

The resistor to B+ sounds simple and a great Idea I toyed with it however never did it as I was trying to make a direct replacement to the mechanical regulator. But yes that looks promising.
As far as the current goes that hall-effect is interesting. I wonder how stable and accurate it would be. My homemade 4mOhm shunts or even 2mOhm shunts seems to develop plenty of voltage to monitor with an AD22057 current Isolation amp for getting a Low side current display working and the shunt also matches the LM311P comparator input voltage capabilities nicely. My copper wound shunts can handle 50 amps however a 100 amp shunt could be used to reduce heat.
Hooking the voltage monitor point up after the B+ schottky barrier diode is one way to further stabilize voltage regulation and certainly worth investigating. My design monitors the voltage drop across shunts alone for accurate current regulation.

Tue Jan 14, 2020 1:02 am

Yes, the whole circuit is connected to the B+, also the supply section (based on a 78L05), so the circuit is always on. This is the Bosch-style regulator, I was inspired by this.

If anyone interested, this is the LT-spice schematic.

Out of the dashed rectangle there are:

- a model for the Battery
- a on/off load
- a generator model

Those will be useful to evaluate the circuit behaviour.

Tue Jan 14, 2020 12:43 pm

What's RGen?
It took me a second or two to wrap my brain around the layout however yes it's going to do the same thing as my design! It does a better job of regulating the supply voltage than my design. Probably the LM193 is a little less current draw. I think my circuit is 7ma minimum where the LM193 pulls 4ma minimum but when you add the LT1086-5 to regulate its internal power the design will pull more current at rest than mine.
I wanted to see how simple I could make mine so avoided unnecessary components. I use voltage spike protection rather than a 5 volt regulator. Not sure about the LM193 Comparator, my LM311P Comparators don't care if the power is regulated or what the power supply voltage to them is as long as it's over 3 volts and I think they go up to 30 volts the power does need to be 1.7 volts or so greater than input bias voltages for them to operate properly.
The Schottky Barrier 4045 diode is a good choice however it's only rated for 40 amps probably not a problem though with a 12 volt unit. I used the same type of diode on mine a 4015 (40 amp 15 volt) on a 6 volts generator and a bit to my surprise with the more than ample heatsink it's holding up fine even with my 2 X 60 Watt 6 volt Head Lamps on High with my generator regulator adjusted to over 7 volts.
Instead of powering from B+ I am powering mine through another Schottky Diode as its a lower forward voltage at low current with only about 100mv drop when I turn the key. (Powered from the 15 wire.) Using a 4015 or 4045 this way provides Or-ing (Powering one load from two supplies) for the unregulated internal power rail.

The real test of a voltage regulator design is to use it in the car.
How well does it start charging on real cold mornings, (Not sure mine would work at 40 below Zero!)
Did it burn out while trying to make it to Denver driving all night with the Headlights on itc.
Does it burn out if someone forgets and leaves the key on.
(I accidentally hooked the battery up backwards on my Bosch 12 volt Electronic regulator for a few seconds. It did draw a lot of current but it didn't damage it somehow!)
"herbie1200" I noticed they eliminate the big field resistor to ground, did you? I have a 10 ohm power resistor hanging on mine. Wouldn't D4 make it overcharge? No I guess not, it's Ok!

Thu Jan 16, 2020 1:30 am

I used the LMx93 series because of its "open collector" output, that is the optimum when the gate of final MOSFET is fed via a pullup resistor, and when you have to use another operational to control the current; you can short operational outputs because open collector acts as an AND port.

And LMx93 is specific for low operating voltage.

The IC voltage regulator is not essential, opamps do not need stabilized voltage; I used it to give more stability to the circuit. A zener suffers with big temperature variations, the top of the generator (where my regulator is mounted) can vary from 0°C to 100°C.

For the cut-off I used a different solution than the schema, an array of 2x2 TO220 schottky diodes.

I've LED headlights so I have no issue with current, and, to be not only theoretical, an old 6V generator outputs 20-25A at maximum, current over 30A is an illusion.

PS. RGen is the internal resistance of the generator. In LT SPICE if a generator is modeled via a formula the internal resistance should be externally modelled.

The coil original resistance has to be disconnected; it is only to protect the original contacts from spikes. The two diodes are not only to protect from spikes from the field but also from negative spikes generated by the generator collector.

Please note that this shematic can be destroyed (parasitic) from a battery inversion.

Thu Jan 16, 2020 12:39 pm

I took a really close look yesterday at the LM311P vs the LMX93 DataSheets I think they could be directly interchangeable in our applications. Everything your saying I do agree with.

herbie1200 wrote:
Please note that this shematic can be destroyed (parasitic) from a battery inversion.
Yes, I think reversing the car battery would wipe mine out too as that forward biases the Cut Off Diodes and the Field Fly-back Diode and maybe the MOSFET internal Zener protection I should check that. The one I reversed that lived was the off the shelf Bosch 12 Volt Electronic Reg. (It Looks like it has 3 maybe 50 amps diodes for cut off.)
As far as current output of the (36 Horse generator 45 amps at 7 volts) and I have a (356 Generator that says 200 Watts @ 6V) that would just be (33)amps wouldn't it. I think they will put out much higher current than their rating however they are derated to protect them. I see surges over 55 amps on my current readout all the time. But Just when the battery is low. I agree though that it's possible current readings the way I am doing them could be off however I calibrated them with known 120 Watt Head light draw/6volts=20 amps: The copper shunts do change value as they heat though and this means that it's a curve not a linear thing.
I couldn't find acceptable LED 6 Volt Headlights bulbs?
And.
"herbie1200" do you or does anyone have any ideas about further simplified 6 volt electronic dynamo regulator designs? .