1979 Dodge Magnum Digital Chronometer clock repair

[440 PHIXX]

My experience with zener diodes is they burn out protecting the rest of the circuit. Usually a max voltage is used so it will burn out like a fuse. They, being diodes also protect from reverse polarity.

 

[Diesel01]

Yes, it is best if they open and crack. That usually doesn't burn up traces on the board. When they fail shorted, that's when PCB repair is required.

So a Zener has a 'knee' that once the current gets there it clamps the voltage. You can make a Zener by connecting the Collector to the Base of an NPN transistor, and depending on the transistor Beta a 'knee' will occur, etc.

More info than you want to know.

Try to power the circuit off of 5V and see what you get.

That is good to know. I have a couple old harddrives around where I have robbed SMD transisitors off it depending if NPN or PNP is needed. Also SMD resistors and caps.

I fixed a jetski module once. It was encased in hard expoxy. I boiled it for 10 minutes in water, peeled the expoxy away to expose the circuit board, drew a schematic, troubleshot and fixed the module with all SMDs. Then re-encased it in Marine gel. it is still working on my ski.

 

[Diesel01]

I removed the Zener from my working clock and tested it. Since my power supply only is variable to 15v, it wouldn't regulate either. I added a 12v battery in series and found the "knee" test on both Zeners from both clocks. They are 24v zeners. They both regulate at 24V. So the zeners are good.

 

[Billccm]

Interesting. So nothing should be dropping much voltage, which means coils and series resistors should not be generating heat. Very resourceful of you to stack the voltages.

 

[Diesel01]

Doing some more research and testing between my two clocks I have found that the coil, capacitor, switching transistor and clock chip form a voltage doubler from 12v to 24v using the 24 zener. On my working clock, I actually have 24.2v across the zener. So with my bad clock, I have to find out if the chip to base of the switching transistor is still pulsing. If it is, I have to figure out what size uH coil I would need to replace it. If it is not pulsing anymore, I think the clock may work but I don't know for how long.

I have attached a simple 12v - 24v doubler using a 555 timer. In my case, the clock chip is producing the pulses. Mine is also using a 47uf cap as well. I will have to get my oscilloscope out and look at the chip for pulses.

 

[Billccm]

Great work. I'm impressed.

 

[Diesel01]

I compared the working with the one with the burned coil

Good coil in working clock

Bad clock with coil removed. The chip is pulsing at 8khz so the chip is working.

I swapped the good coil and bad clock and circuit is working and no components getting hot.

Now to find the value of the coil since no value of any kind is on it. I dont have an inductance meter so have to do it old school with an AC source, resistor and algebra. I did it two ways to double check myself. One with match the inductance resistance and the other using voltage drop across the coil and test resistor. I have come up with 44mH. Looking at digikey and the physical size, it is close in size to mine 1/2"x 3/4".

If anyone wants, I can bore you with the math.

 

[Billccm]

Ringing is good! The inductor is doing it's job.

 

[Diesel01]

I thought I would update on my progress. I was able to fix the clock that had the burned inductor. However, my calculations were off. I purchased an Induction meter and found that the coil was 5.6mH. When I swapped the coil in the non working clock, I was only getting about 22v so I decided to get a 4.7mH and a new 3watt 82 ohm resistor for the filaments. 4.7mH gave me 24.2v.

I am posting my partial schemtics of the two different and the pinout of the VFD.

This is the 2 board version

This is the single board version

I troubleshot the 2 board version down to the 814 chip. Pin 1 is connected to pin10 of the VFD which is segment "b". Pin 1 has 20v and a 1v square wave which is causing all the "b" segments on all digits to be on. I have searched high and low all over the Internet and I cannot source the National Semiconductor 814. I cannot even find any information on it. The only info I have on it is from reverse engineering the circuit.

 

[Diesel01]

Single board version. I should mention I cannot source the 7837 chip for this board either.

VFD part#. I cannot source this VFD but I have found Russian ones that are not an exact replacement but by swapping pins around, I could probably get it working.

 

[Billccm]

Great work. I worked from National Semiconductor from1991 to 2003. I've got no contacts there as most people I worked with have passed away or was laid off after the acquisition by Texas Instruments.

It might be a ASIC ( application specific integrated circuit). National had a good ASIC business with many automobile suppliers.

Thanks for the update and sharing your findings.

 

[Diesel01]

Great work. I worked from National Semiconductor from1991 to 2003. I've got no contacts there as most people I worked with have passed away or was laid off after the acquisition by Texas Instruments.

It might be a ASIC ( application specific integrated circuit). National had a good ASIC business with many automobile suppliers.

Thanks for the update and sharing your findings.

That is my thoughts exactly is that these chips were custom made to be a real time clock and drive a VFD before cars were controlled by a CPU.

There are plans on the Internet and Github to do the same thing but with an Ardrino chip and a VFD driver


I did not know TI acquired them. Now I know why I kept getting TI results when I was searching.

 

[Diesel01]

Here is a back shot of the VFD that shows how the segments are connected.