Page 2 of 2

Re: Ignition Waveform

PostPosted: Wed Oct 21, 2015 2:13 pm
by JGeezer
First, there are early and late Smith's tachs, that have two transistors or just one respectively. Here is the circuit diagram for the "late" with one, which is what I have:

72JagSeries2TachA.jpg and


I replaced the capacitor, and the needle is steady but reading way too low (like maybe 1/3, I have an external meter I can calibrate against). Now I have to calibrate, and looking I see two adjustment screws:

IMG_2436.JPG and


(WHY didn't I trace circuits before putting it back together, you ask?) I only see one variable (plus the thermistor) in the diagram - can somebody tell me how to properly calibrate?

JGeezer wrote:
bob_rich wrote:Hi JG

I think there are a number of reasons why your waveforms look odd and suspect there is not a real problem ( with the ignition at any rate ) when the points open , close, and the spark occurs at these 3 time points current around the circuit changes rapidly -- this can induce voltages in what might be thought of as "0V or earth" connections. Depending where the scope probe earth or 0V is connected this can give the appearance of some ringing as well as appearing to produce a voltage that seems to go in the other direction. the leading edge of the first pulse seems to show this effect.

The capacitance of the hIgh voltage plug leads determined partly by how long they are and how close to earthy stuff, it is not unusual to see each cylinders pulse looking a bit different. Rarely are all four plug lead identical in layout. Also in the dizzy the volts have to jump the dizzy gap as well and they may be variation here

Not sure what coil you have but they typically step up the voltage around 50 to 60 times so to get say 10000 volts, (probably enough volts at fast tick over with a high vacuum) at the secondary probably means around 180V at the primary. So your scope vertical amplifier system would be overloaded. Depending upon the scope how it responds to this overload may give false readings AFTER the high voltage pulse has occurred.

From your waveform it looks though the points closed period is around 2/3 of the time between pulses and this seems about right for a 60% or so dwell.

Regarding the tacho if you have an RVI type with the inner trigger loop ( when you open it up if there is only a single transistor then it is) then it is probably the capacitor across the base - emitter of the transistor. I seem to recall a 2.5uF at 10V tantalum or electrolytic will substitute for it OK. When mine went iffy I changed this capacitor and it is now OK.

Hope this helps best of luck

Bob


Thank you Bob, very clear explanation. Very cheap little scope (half the size of a phone), so I do only take it's waveforms to be a very general guide. I will be pulling the tach tonight.

This tread turned out to be very informative for me, I never thought about the capacitor interacting with the inductance of the coil on the rising signal, makes great sense.

Re: Ignition Waveform

PostPosted: Wed Oct 21, 2015 5:12 pm
by billwill
Unfortunately I think most of us have the other 2-transister model, so I for one don't know which screw is the adjuster.

Tthe second screw probably just secures the board to the frame or case.

Re: Ignition Waveform

PostPosted: Sun Oct 25, 2015 1:07 pm
by ricarbo
There is only one adjuster and I'm fairly sure it is the slotted plasic part at the left of your picture. There are 4 metal screws and it is none of those. The two with large metal washers hold the movement in the case. The other two hold the electronics board to the movement so you don't undo these until the movement is out of the case.
Unless you want to do a lot more work, you can get the calibration roughly right by attaching an external rev counter to the car, then twiddling the adjuster to match them up. On some variants, the adjuster is not accessible without dismounting the movement from the case, but it looks like yours is.
I say roughly right, because calibration means more than just adjusting the electronics. The movements have mechanical correctors in them and you should really start with those. They compensate for the many different weights of indicating 'needle' used in different cars with different instrument styles. Also the circuit with the single transistor seems overly simple to me. It uses a thermistor for temperature compensation, but thermistors are non linear. It has no regulation of the supply voltage, unlike the earlier two tansistor version.
My opinion is that it is better to change the ancient electronics for something more modern, compatible with whatever ignition system you may present it with. Even if you do all this, it will still not be accurate at all speeds, because of non linearity of the printed face compared to the movement. The best achievable is accuracy at two nominated speeds, whatever you do. But the errors are not enormous.
Originally, there was a rubber grommet over the adjuster and you can see the mark it has left. There to keep dust out.
regards
Richard

Re: Ignition Waveform

PostPosted: Mon Oct 26, 2015 5:48 pm
by bob_rich
Hi JG

You have a unit with the single transistor so it is an internal current loop type . There is only a singe adjuster. If it has not been moved for a while it might be a bit scratcy so might need a spray of contact cleaner As you have a scope you could count the timing pulses in a given time and that would give a good idea ( most scopes read timing or pulse frequency pretty well) if you trigger the scope off the coil lead just dangle the probe near the coil HV lead should pick up the pulses OK, or connect the scope as your original waveform at the contact breaker. If you read the coil HV pulses then you get 2 pulses per rev for a 4 cylinder engine so 200 pulses per second would be 6000rpm.

There are tricks to calibrate tachos of this type out of the car and if you want a circuit idea I can sketch one up for you as you have a scope I assume you might be in a position to knock up a simple calibration circuit. Yuu may find some on the web as well.

Incidentally I found my tacho ( which is the circuit you show) was quite accurate and linear over the range to 6000rpm and the reading was fairly insensitive to battery voltage from 11 to 16V

Hope this helps

best of luck

Bob