Help Identifying Vintage Component

[David Hess]

A 1 microfarad 75 volt Sprague 150D series solid tantalum is 6.79 dollars:

https://www.mouser.com/c/passive-components/capacitors/tantalum-capacitors/tantalum-capacitors-solid-leaded/?termination%20style=Axial&voltage%20rating%20dc=75%20VDC~~125%20VDC&instock=y&sort=pricing&rp=passive-components%2Fcapacitors%2Ftantalum-capacitors%2Ftantalum-capacitors-solid-leaded%7C~Voltage%20Rating%20DC

If I wanted to save some money, I might use a plastic film capacitor in series with a 1 ohm metal oxide resistor:

https://www.mouser.com/c/passive-components/capacitors/film-capacitors/?capacitance=1.5%20uF&termination%20style=Axial&voltage%20rating%20dc=100%20VDC~~250%20VDC&instock=y&rp=passive-components%2Fcapacitors%2Ffilm-capacitors%7C~Voltage%20Rating%20DC&sort=pricing

https://www.mouser.com/c/passive-components/resistors/metal-oxide-resistors/?resistance=1%20Ohms&instock=y&sort=pricing

[Drjaymz]

A 1 microfarad 75 volt Sprague 150D series solid tantalum is 6.79 dollars:

https://www.mouser.com/c/passive-components/capacitors/tantalum-capacitors/tantalum-capacitors-solid-leaded/?termination%20style=Axial&voltage%20rating%20dc=75%20VDC~~125%20VDC&instock=y&sort=pricing&rp=passive-components%2Fcapacitors%2Ftantalum-capacitors%2Ftantalum-capacitors-solid-leaded%7C~Voltage%20Rating%20DC

If I wanted to save some money, I might use a plastic film capacitor in series with a 1 ohm metal oxide resistor:

https://www.mouser.com/c/passive-components/capacitors/film-capacitors/?capacitance=1.5%20uF&termination%20style=Axial&voltage%20rating%20dc=100%20VDC~~250%20VDC&instock=y&rp=passive-components%2Fcapacitors%2Ffilm-capacitors%7C~Voltage%20Rating%20DC&sort=pricing

https://www.mouser.com/c/passive-components/resistors/metal-oxide-resistors/?resistance=1%20Ohms&instock=y&sort=pricing

Mouser doesn't work so well for me from here, but I can see these on RS.  I think 63V would be ok 50V probably not, having said that the transistor wouldn't survive above 50Vce - well it might but it would be annoyed.

[David Hess]

Mouser doesn't work so well for me from here, but I can see these on RS.  I think 63V would be ok 50V probably not, having said that the transistor wouldn't survive above 50Vce - well it might but it would be annoyed.

I think the circuit is using the capacitor as a snubber and not for its voltage breakdown.  If you want to do the later, then a 75 volt TVS diode will be more suitable.

[fzabkar]

That is just the regulator.

Doh, yes, of course. I was confused because your description and photo reminded me of the regulator I built for a mining application where there was no battery. IIRC, the field was grounded via a relay until the alternator's output voltage rose to a level sufficient to energise the relay and open its contacts. The transistor and zener then took over. The need for a battery (to excite the field) was obviated by the residual magnetism in the field.

BTW, your circuit makes no sense to me, either. :-?

[floobydust]

It won't work as drawn unless it's using an SCR. C3 is not on the schematic, part values would help.

[Drjaymz]

It won't work as drawn unless it's using an SCR. C3 is not on the schematic, part values would help.

I agree, because it latches on. (I think I mentioned that in previous post).

The mystery component which I thought was a zener has the markings 4545 in blue on a black case and no other markings except a ring for the anode.  I identified it as a zener because it doesn't do anything until just over 11.1V where it appears to clamp and as you raise the supply up to 20V its breakdown voltage only rises slightly to 11.4V.

Its common to have a zener reference that turns on a transistor in these type of systems, either over-voltage or the regulator (they are both the same).  But as drawn it wouldn't work because once you turned on the transistor then the transistor biasing circuit would turn it off again because its supply is now Vce which is ~0.8V.  And yet, here it is on my desk working.  Hence why I'm perplexed.  Quite often when you look at automotive electronics from the 60's and 70's the circuit isn't as obvious as it may seem until you figure it out and realise its very cleverly done.  The regulator someone else posted is very similar but that would be non-latching.

I noticed that I omitted C3 after I took the photo.  Its actually between Relay -ve and top of C1.  Its value isn't readable but its a high voltage small capacitance polyprop.  Its not measurable with my meter out of circuit so its less than 1n at this age, I think its supposed to be 4.7nF.

The other odd thing is that R2 measures 1K but its colours are Pink Black Red Gold - now I have seen Pink before and I looked up a weird series where pink is 7.  But it could also just be age as its not uncommon for white to fade (although usually looks yellowish).  This unit would not get warm as it should only activate under fault conditions.

[Drjaymz]

If we ignore the relay for a moment, then the junction between R1 and VR1 would scale with the supply voltage.  Current does not flow through ZD1 until Vr1 > ~ 11.1V.  Then Vbe raises to about 0.7V at which point the Relay activates.

Then it stays activated.  And thats where I am confused.

If R3 took its supply from the input then that would work, but then the relay would de-activate when you reduced the supply voltage - and it doesn't, TR1 stays hard on until the relay lets go at 3v.  If you then raise again it activates again at about 7V demonstrating that the transistor has never switched off.  So its latched.

Now, to get further into it I could dismatle it further, but since it now works and all I have to do is replace C1, I'm probably not going to risk breaking it.  So it may end up keeping its secrets.

I have thought about this a bit more, T1 Must be an SCR.  I don't come across those much these days and all the more important that the capacitors have some capacitance because otherwise it could trigger on spikes and will by definition latch on.  I'll see if I can find markings before I stick it all back together.  I could also easily test this by desoldering the "base" and triggering it manually, if it stays on then thats what it is.

So whilst I know what an SCR is and how its triggered, what I don't know is what voltage will be present on the gate once its triggered (if its not connected).  I mean that we know you need a threshold to trigger it but we see 0.8V after the device latches on.  Now, since its a stack of silicon that may well be correct.  Hopefully someone understands what I meant there.

I had a look at whats available today and T092 case style Thyristors with a gate voltage of 800mV do exist and perhaps the part number is irrelevant from the 60's anyway.

[Drjaymz]

Logo "H" OK I found the manufacturer is Hilton Capacitor...

https://web.archive.org/web/20050210214206/http://www.tantalum-pellet.com/

This is it.  4070 is the date code.  Thats week 40 1970 which is correct because the aircraft was built in 1974.  So 75 would not have been it.
The rating capacitance and voltage rating then must be 15V75.  And it cannot be 15 volts so it must be 75V  and that leave 15 which is wither 1.5uF or 15uF.  15uF at 75V I think would be physically larger, this is 13mm long, 3.4mm diameter.  So that makes it most probable that its 1.5uF.  There is another cap next to it and I cannot see its markings but its slightly less diameter and length and its measures just under 1uF.

So its a Tantalum Pellet 1.5uF 75V capacitor.

And the weird circuitry - my diagram is correct, its not a transistor its a Thyristor, what threw me was that I can't access the markings and that in the on and off state the voltages were very close to an NPN. Its a T092 800mV gate voltage and would latch as it does.

I may just change both tants for a 1.5uF, I mean the tolerance of -10 / + 30 means they could overlap!  I can't understand why they are so expensive and why nobody has no stock unless as someone has said, they are indeed still hand made.

[Drjaymz]

This is fixed, tested and working nicely.

Now, going back to the circuit diagram.  Why does it use the zener at all?  You could just arrange the potential divider such that it was biased around the Thyristor trigger.

One reason could be that the zener is a stable voltage reference.  I don't really see it being any better than a simple divider and transistor.

[David Hess]

One reason could be that the zener is a stable voltage reference.  I don't really see it being any better than a simple divider and transistor.

It is much more stable than just the Vbe of the transistor.  Placing a zener in series with the base or emitter is an excellent way to make an accurate reference and error amplifier for a regulator or reference.  Pick the right zener voltage and the temperature coefficients of the zener diode and Vbe cancel.  RCA used to sell transistors with the zener built in.

[floobydust]

Note Prestolite is using the relay coil (resistance) as part of the voltage divider, giving this a tempco. Intentionally or not?
Copper wire at 20°C vs 50°C around 19% shift up, increasing the trip point.
System voltage at 16V for flooded lead-acid batteries is close but not quite a hard failure in very cold weather, you can find 15.5V below -30°C.
Older alternator voltage-regs many did not have a "shelf" at temp extremes whereby they limit the charging voltage temp compensation, clip it at the low temps.

The parts and circuit have a few discrepancies so it's not possible to extract the design secrets, but I would say there's enough for anyone to do a safe repair of the module.

[Drjaymz]

Note Prestolite is using the relay coil (resistance) as part of the voltage divider, giving this a tempco. Intentionally or not?
Copper wire at 20°C vs 50°C around 19% shift up, increasing the trip point.
System voltage at 16V for flooded lead-acid batteries is close but not quite a hard failure in very cold weather, you can find 15.5V below -30°C.
Older alternator voltage-regs many did not have a "shelf" at temp extremes whereby they limit the charging voltage temp compensation, clip it at the low temps.

The parts and circuit have a few discrepancies so it's not possible to extract the design secrets, but I would say there's enough for anyone to do a safe repair of the module.

Its odd that they tend to use remote regulators. That means they more often than not end up with oscillations due to resistance and interference, especially when you consider the average GA aircraft is over 40 years old.  The over voltage relay is optional and unfortunately you dont know its failed until your radios explode one day which happened to a friend of mine in february as it fried the garmin glass g<something>.  A good 12k of damage. You can't easily test it in place and its never tested on annual but worth doing.
Theres nothing wrong with the relay and its not really a surprise that its a capacitor at fault. Later models will unlikely be repairable.

The later models claim better accuracy but I think the important thing here is that the capacitor is really there to prevent it triggering spuriously so it does need to be there especially as its using a thyristor, once triggered it will stay on.  I didnt know they latch on and can't find that information anywhere. But after reverse engineering that's the only way the circuit works. 

[fzabkar]

Its odd that they tend to use remote regulators. That means they more often than not end up with oscillations due to resistance and interference, especially when you consider the average GA aircraft is over 40 years old.

It's odd that you should say that. :-)

My parents had a 1967 Chrysler Valiant (car). The regulator was a remote type, but it was mechanical, so there were no stability issues. However, when I retrofitted a modern solid state regulator at the same position, the lights flickered with a 1 second period. IIRC I cured this oscillation by installing a relay so that the regulator was able to directly sense the alternator output (rather than via the ignition switch).

[floobydust]

If you ever look at alternator field winding waveforms, from an old school linear solid-state regulator, they are actually low freq. PWM the field is bang-bang controlled. I expected a nice linear control but there is more gain than that, and the field winding just toggles on/off. That was a surprise to me.
People running old aircraft retrofitted with electronics, ultralights as well, are finding you need bulk capacitance 10,000-150,000uF added in order to lower system ripple and noise, transients. I'd be adding some huge TVS they use for load dump as well.

The 1Hz oscillations of car charging system voltage I have only seen when the battery is pooched and has a problem. A few cars with lights surging up and down in brightness at 0.5-1Hz.

[Drjaymz]

If you ever look at alternator field winding waveforms, from an old school linear solid-state regulator, they are actually low freq. PWM the field is bang-bang controlled. I expected a nice linear control but there is more gain than that, and the field winding just toggles on/off. That was a surprise to me.
People running old aircraft retrofitted with electronics, ultralights as well, are finding you need bulk capacitance 10,000-150,000uF added in order to lower system ripple and noise, transients. I'd be adding some huge TVS they use for load dump as well.

The 1Hz oscillations of car charging system voltage I have only seen when the battery is pooched and has a problem. A few cars with lights surging up and down in brightness at 0.5-1Hz.

"Panel light flicker" is a very common complaint, if you look at the regulator part number you'll see other manufacturer equivalent parts and they all state "reduced panel light flicker" which tells you why people look to change it out.  It is as you say either all the field current or none and there's a capacitor in the resistor chain which looks like they were trying to build an oscillator!  I have a new revision M regulator which does use PWM which I never tried so might ground run with that and see if it helps.  The older regulator uses a Zener reference a Darlington pair then a beefy transistor so yeah, lots of gain there.  The transient suppression is taken care of with a diode and standard neon protects the main transistor.  Prestolite's maintenance manual diagram attached if you are interested.  This type of arrangement is probably common to 1960's automotive designs.  Note that despite the diagram attached this is potted and so it really is impossible to fix.

The thing about aircraft wiring and batteries for those not familiar is that they are lightweight, so both the wiring and the battery have huge resistance compared with a modern car. I feel a short alternator cable as thick as your thumb going to a good battery is infinitely superior but on the plane the alternator lead is both long at 5' and no thicker than your multimeter lead and its expected to carry 40A.  Unlike a car they then have to go through a fuse or breaker, and a master switch.  Which means that even when it was new it wasn't really ideal.  For balance reasons some aircraft have the battery in the tail as far as possible from the engine and twins have two alternators to worry about.  Then you could have 24 / 28 volt systems event on smaller singles like Cessna decided was a good idea.  Ultralights use gel batteries and I think I've see lithium phosphate at aero expo but I'm not an ultralight aficionado. 

On certified aircraft you have to stick with the 1960's design even if they are massively deficient, again, the requirement for STC's and the impenetrable paperwork and cost requirement basically discourages improvement.