Yes, replace the diode if it tests bad. Is it shorted?

Test voltages at connector. Do you have a schematic?

Attach a picture of the schematic from the totem manual that board replaces.

With the connectors shown on the schematic, it's possible to test all the output voltages coming out of the board and see if any are missing.

DISCLAIMER: I have ZERO experience with Gottlieb machines (other than occasionally playing them). I also learned my electronics from reading and doing rather than any higher education institution. You would be advised to double check the information provided.

It looks like Rottendog used the MC34167 on this board. GPS001 shows that D3 is 1N5400. GPS002 shows that the same diode is now a Schottky diode (1N5825) instead of a standard rectifier diode (1N5400). My understanding is that a standard rectifier diode cannot switch fast enough forcing a Schottky diode to be required. If you replace the diode you should probably replace it with a Schottky diode (1N5825). That diode can also be substituted with 80SQ045N. That diode has higher ratings than the 1N5825.

Documentary information follows. Information drawn from https://www.onsemi.com/pub/Collateral/MC34167-D.PDF

mc34167_diode.jpg

mc34167_design.jpg

Answer to 2nd question of post 1 is here, measure output voltages from this diagram. Believe this is the correct board:

http://www.pinrepair.com/sys1/ps1.gif

http://www.pinrepair.com/sys1/index.htm#power

"Testing the System1 Power Train."

As DumbAss stated (get a chuckle every time I see that name) -- There should have never been a standard rectifier used in this position. This is the catch diode for the supply --> a huge amount of current goes through that diode and it MUST maintain a low forward voltage drop. The 80SQ045N is a great choice for a replacement.
Looking at the input and output capacitors - it appears there is another issue waiting to happen. It looks like those are standard Nichicon capacitors, probably UVR series. Nothing wrong with Nichicon UVR/UVX caps but they are the wrong type to be used as input and output caps for a switching power supply. Those two MUST be low ESR type caps. If they are not low ESR, the ripple current through these two caps will substantially reduce a standard cap's life span. Is there a "VR" or "VX" marked in small letters on the sides of the input and output caps (C1 and cap near L1)?

Here is a schematic of the sys 1 power supply with voltages marked at each component.

http://www.stevechannel.com/images/tech_tips/System1PowerSupplySchematic.jpg

C1 (Cinput) is a Nichicon UPW series cap. This is a low impedance cap which is a good thing.

C3 (Coutput) is a Nichicon UVX series cap. This is a standard but high temp cap - which is NOT a good type to be used as a switching regulator output cap.

Standard practice on switching supplies is to always use a low impedance caps at the input and output of a switching regulator, usually paralleled caps are used to reduce ESR and increase ripple current capabilities.
For the input cap, low-ESR is often optional but a high ripple current rating is a must. Increasing the ripple current rating is often achieved by using a low ESR cap.
For the output cap - both high ripple current rating and low ESR values are a must.
The cap he used for C3 is no longer available so I can't look up diameter and lead pitch. I'm guessing that would be 16mm diameter with 7.5mm lead spacing. Assuming he used the correct capacitance and voltage values, I would go to a Nichicon UPW1C472MHD or Panasonic EEUFC1C472.

Regarding R21 -- 100 ohms is a bit low:
If the DC in is 12V (normally higher for this board)--
Voltage across R21 = 12V - zener voltage 8.1 = 3.9V
Current through R21 = = 3.9V / 100 ohm = 39mA
Power dissipated by R21 = 3.9V x 39mA = 152mW. A bit high for a half watt resistor but acceptable.
Power dissipated by CR21 = 8.1 x 39mA = 0.31W. That works but is quite high for a 1 watt zener diode. It will get hot but shouldn't burn - heat created will result in shortened life span.
In real life - the DC voltage on a System 1 supply is normally a bit higher than 12V which increases the above wattage values and makes things worse.

Going to a 470 ohm resistor will reduce zener power to about 20% of what it was or about 0.062W - MUCH better.
I normally use 390 ohms which would reduce power to about 25% of that 0.31W rating or about 0.077W -- also much better.

Going to too large of a resistor value will cut off current flow to the zener diode (below zener diode "knee" current rating).
The zener diode requires a minimum of just under 1mA to operate properly.
100 ohm resistor provides 39mA -- high enough... actually too high.
390 ohm resistor provides 9.75mA -- plenty of current to keep zener operating.
470 ohm resistor provides 7.8mA -- also plenty of current to keep zener operating.
So a resistor in the 390 or 470 range will work. Can actually go higher but I never bothered to calculate where the limit is.
Going too high will also create an excessive voltage drop across R21 depending on load from displays (normally quite low).
To prolong life of R21 and CR21 and stay in proper voltage ranges - I recommend you change R21 to 390 to 470 ohms rated at 1/2 watt.

AND - rant mode on...
Another strange thing I noticed - there is no bulk capacitance for the low voltage input!
The input capacitance is the switching regulator's value used for when you already have a DC input to the circuit, it is NOT intended to be part of the DC rectification! This is a low ESR, high ripple current cap placed 'close to the device'. *The rectified voltage is supposed to already be filtered prior to the switching regulator's input cap*. This board goes straight from bridge rectifier to the regulator input cap and then the regulator. There is no dedicated bulk capacitance,he combined the bulk capacitance and the switcher input capacitance! No wonder these were cheap.

Normal design of this type would go from rectification circuit CR1 & CR2 to a bulk capacitor. Bulk capacitance is used for basic DC filtering and are sized to supply average DC current without sags and absorb current swings.
From bulk capacitor - voltage the goes to the Cin cap which are placed as close to possible to the regulator. The Cin cap is used to facilitate the high current spikes normally created by the switching type regulator.
The bulk and input caps are normally not combined into a single cap - let alone one with such a small value. Yet, somehow his board works. In the electronics world, we call this type of design "Muntzing". Muntzing is the practice of reducing the components inside an electronic device to the minimum required for it to function long enough to get past warranty.

Never heard that expression before. We are dealing with machines from the late 70's and 80's that I think were never thought of still being here today when they were manufactured.
With that being said, what do you think of the original design of the sys 1 ps. I see many still in service today. Granted most are rebuilt but the design seems to work.
-Mike

Muntz had a series of TVs in the 1950s. These TVs were known for being service nightmares. When a design was complete - he would open the TV and start cutting out components until it stopped working and then put the last part back. Adjustable parts - they would measure the adjustment and then make the adjustable part into a fixed value for all TVs. People in fringe areas were screwed. The whole idea was to make it as cheap as possible and not care about servicing the product after the short warranty expired.
BTW - it was Muntz that made popular the abbreviation "TV" for Television.

Regarding the original System 1 and System 80 power supplies -
The original design was mostly a sound design (for that period in time) but poorly implemented. The biggest problem with these is the heat sink design where the heat sink had to be removed to get to other parts. That made it a nightmare to work on.
Once you got these working and stopped futzing with them - they tended to work well and kept working. They got quite hot - but they worked.

Low voltage is good.

High voltage is not good.
These would be down in the high voltage section of diodes (near J1) and up to the LM317AHV regulator in the middle.
I believe this is due to how you are reading the voltages. If I remember correctly, the old System 1 supplies had separate high voltage and low voltage grounds. You need to reference the correct ground for old system 1 supplies. I have no idea as to what this replacement board did regarding grounds.

One important detail - the HV supply has about 90V rectified voltage on it. The LM317AHV can withstand an absolute maximum of 60 volts between input and output. In a perfect world, this would be fine. In the real world, the output is often loaded down to a low enough voltage to take out the LM317AHV regulator. Poor design. Obsolete part, I have sold out of replacements for these long ago.

One way to determine if good voltage is getting to board:
Disconnect P1 from board, meter in AC voltage range, measure between pins 6 and 7 of the plug while P1 is still disconnected from board.

Yes - plug removed from A2 J1.

I need the voltage directly between pins 6 and 7.
Based on:
Pin6: 37
Pin7: 32.50
Sounds like you measured these with respect to ground?

Which fuse is blowing? low voltage? High voltage?

That HV input is correct.
With the plug to J1 being pulled and the correct voltages above, your problem lies within the power supply itself.

When the LV regulator catch diode crapped out, it may have taken one or both of the full wave rectifier diodes with it. These are the two black diodes closest to the 390 ohm resistor that you installed. Check those two with meter in diode setting with plug to J1 pulled. Also possible that the voltage regulator itself may have been taken out - MC34167TVG.

The MC34167 is a pretty complicated integrated circuit - no way to test him on board other than to power it up and try it.

The output transistors inside your voltage regulator (MC34167) appears to be shorted.

But first make sure you have good connections to the catch diode (80SQ45N):
Make sure you have a good ground connection to the anode (non banded end).
And make sure you have a good connection between the cathode (banded end) and the inductor (L1) at the end closest to the regulator.

For the continuity to the diode - visual and continuity.
Some pretty good current surges go through that diode during normal operation.

Not sure if I missed it: What is the Ni-Wumpf issue? I recently got mine repaired and working.

Unfortunately, if you ran that CPU board with the 14V from the power supply instead of 5V - that board may be toast.