How to fix PCBs - A Beginners Guide

Thank you. Although I'm used to Arduino and ESP programming it is still a bit overwhelming at this stage.

Was looking at a TL866 with pin attachments. Any other recommendations?
 
Thank you. Although I'm used to Arduino and ESP programming it is still a bit overwhelming at this stage.

Was looking at a TL866 with pin attachments. Any other recommendations?

Use the Advanced Thread Search. There are many threads on this topic, including recent ones. Search for TL866 or rom programmers or programming or recommendations, and you'll find them.

In general, it's best to try searching before asking questions to the group here, as there are 20 years of archives, and you'll actually find better answers than from whatever people happen to be online when you ask. It also saves folks here from having to answer the same questions repeatedly.
 
Is there a specific way to "jump" power to other parts of a PCB? (e.g. if the chips on the back does not get enough volts).

Can you just, for example, solder wires from the JAMMA harness to the Vcc and GND pins of a EPROM or is there a cleaner way?
 
Is there a specific way to "jump" power to other parts of a PCB? (e.g. if the chips on the back does not get enough volts).

Can you just, for example, solder wires from the JAMMA harness to the Vcc and GND pins of a EPROM or is there a cleaner way?


The chips on a board should all be getting the same voltage. There shouldn't be a big enough difference to matter.

If that's not the case, you're measuring things wrong, or are missing something in your understanding.

Make sure the pin you think should be 5V is actually a 5V pin. Some chips do not have the 5V power pins in the corners of the chips.

What board and what pins are you measuring that makes you think the voltage is low?
 
The chips on a board should all be getting the same voltage. There shouldn't be a big enough difference to matter.

If that's not the case, you're measuring things wrong, or are missing something in your understanding.

Make sure the pin you think should be 5V is actually a 5V pin. Some chips do not have the 5V power pins in the corners of the chips.

What board and what pins are you measuring that makes you think the voltage is low?
Or you have some incredibly corroded sockets or chip legs, or chip legs (e.g., Namco customs) that have legs which seem to dematerialize at times.
 
Or you have some incredibly corroded sockets or chip legs, or chip legs (e.g., Namco customs) that have legs which seem to dematerialize at times.

Yes. In which case adding more voltage is not what you would want to do. (And that would also fall under 'missing something in your understanding').
 
Yes. In which case adding more voltage is not what you would want to do. (And that would also fall under 'missing something in your understanding').
I agree.

Work the problem. Don't make it worse by guessing.
 
The chips on a board should all be getting the same voltage. There shouldn't be a big enough difference to matter.

If that's not the case, you're measuring things wrong, or are missing something in your understanding.

Make sure the pin you think should be 5V is actually a 5V pin. Some chips do not have the 5V power pins in the corners of the chips.

What board and what pins are you measuring that makes you think the voltage is low?
I'm referring to this:
Check the voltage at the chips as far away from the power input as possible, volt drop across a board is normal, you many have 5V at the input of the board, but if that's dropped to 4.5V at the farthest reaches of the board then things will get screwy. Either up the voltage a tad, or run a power wire to the far side of the board, operators often did this back in the arcade heyday
I'm struggling with this concept.

I'm working on a Namco Tekken 2 board where the PSU is set at, for example, 5V. When the PCB is not connected to the JAMMA harness, the Vcc and GND of the JAMMA harness measures 5V too. When the PCB is connected, the voltage measurement at on the JAMMA pins of the PCB measures 4.7V. The chips on the PCB also measures more or less 4.7V.

Is this drop to be expected or does it mean there is a problem that needs fixing (or to be "worked"), e.g. checking caps?

There is a lot of info on the internet that say to simply increase the voltage on the PSU, so I get the impression the voltage drop is to be expected...

In my example, do you up the PSU a tad, or do you run a power wire to the back? If not, when would be a good example to run a wire to the back?

Also, where do you run the power cable from (and where to)? From the power supply or the JAMMA input pins.

Lastly, do you only run Vcc or do you run a GND wire too?
 
I'm referring to this:

I'm struggling with this concept.

I'm working on a Namco Tekken 2 board where the PSU is set at, for example, 5V. When the PCB is not connected to the JAMMA harness, the Vcc and GND of the JAMMA harness measures 5V too. When the PCB is connected, the voltage measurement at on the JAMMA pins of the PCB measures 4.7V. The chips on the PCB also measures more or less 4.7V.

Is this drop to be expected or does it mean there is a problem that needs fixing (or to be "worked"), e.g. checking caps?

There is a lot of info on the internet that say to simply increase the voltage on the PSU, so I get the impression the voltage drop is to be expected...

In my example, do you up the PSU a tad, or do you run a power wire to the back? If not, when would be a good example to run a wire to the back?

Also, where do you run the power cable from (and where to)? From the power supply or the JAMMA input pins.

Lastly, do you only run Vcc or do you run a GND wire too?


What Womble said is wrong. He's mixing up two different things.

You don't get voltage drop across the board. What happens is there is voltage drop in the wiring harness, as well as the edge connector.

When you measure voltage at the PS, or at the JAMMA connector *without* the game board plugged in, there is no load on the PS, so there is no current being drawn.

When you plug the game board in, you have several amps flowing through the 5V wires of the harness. Per Ohm's Law, volts equals current times resistance (V = I*R). Even though the harness is just copper wire, it does not have zero resistance. So there is some voltage drop across the harness and connector.

This will cause the voltage measured on the board to be about half a volt (give or take a couple of tenths) lower than when you measured without the game board plugged in. But the voltage at all points on the game board should be the same.

As obitus said above, the solution is to turn up the 5V so you measure 5.00v on the game board when it is plugged in. If you measure at the PS after that, you'll have roughly between 5.5 and 6.0V, depending on how much current the board draws and how good your harness is. (It will very per game and per setup, but it should be in that range.)
 
Last edited:
What Womble said is wrong. He's mixing up two different things.

You don't get voltage drop across the board. What happens is there is voltage drop in the wiring harness, as well as the edge connector.

Actually you do. Measure the +5v at the JAMMA connector pads. Measure it at a chip close to the JAMMA connector, measure it at a chip at the other end of the board. Measure it at a chip on a daughterboard. They will vary slightly in a well designed board and greatly in a poorly designed one. It's common to see 1 to 2 tenths of a voltage drop on the video PCB vs the CPU PCB on Galaga.
 
Actually you do. Measure the +5v at the JAMMA connector pads. Measure it at a chip close to the JAMMA connector, measure it at a chip at the other end of the board. Measure it at a chip on a daughterboard. They will vary slightly in a well designed board and greatly in a poorly designed one. It's common to see 1 to 2 tenths of a voltage drop on the video PCB vs the CPU PCB on Galaga.
Is this a case scenario where you jump wires to the back of the board?
 
Is this a case scenario where you jump wires to the back of the board?

No. You never want to do that.

In some multi-board stacks, you can get differences between boards, when some boards get their power from other boards, through additional wiring and/or connectors. However you want to adjust the 5V so you have at least 4.90V on all boards. (You might need to set one board slightly low, so the other is slightly high, as you often cannot adjust them separately.)

TTL chips can operate across a range of voltage, typically 4.75-5.25V at least. You don't want to set the voltage at those levels, but as long as all chips are getting voltage in a range within that (e.g., 4.90-5.10V), everything should work fine.
 
Never, ever?

Yes, pretty much never. It's just a bandaid that hides the real problem(s).

Once you get a stable 5V onto the PCB at the connector it should only drop 100-200 mV across the PCB from corner to corner. If it drops more than that you likely have a board problem that you need to solve first. Most likely a failing component that is drawing more current than it should. Or it could be a damaged trace that exhibits a higher than normal resistance (and therefore a higher voltage drop). In rare cases it could be a low impedance sneak path...current flowing through contaminants, etc. That's why you should always clean a board before you start troubleshooting it.


Getting 5V onto the PCB is almost always a connector problem. But it can be bad wiring too. Ideally your wiring drop should be below 200 mV at 4A. And the connector losses maybe 200-400 mV. But it all depends on the game. Wire gauge and length, connector type, and the number of parallel connections are all factors. But beyond the ideal case if there is oxidation or corrosion anywhere along the conduction path or broken/missing wire strands the resistance will creep upward resulting in extra voltage drops of 100's of millivolts up to several volts in extreme cases. BTW, corrosion can travel several feet up the wire inside the insulation too. Check it!

If you have more than about 500 mV of drop between the regulator output and the IC nearest the connector then you have a wiring/connector problem and you need to track it down and fix it because it will only get progressive worse. And the the worse it gets the faster it will get even "worse than worse". Because things will heat up more and speed up the chemical reactions that lead to permanent damage. It's a runaway process.

EDIT: For multiboard stacks, the same thing applies. It's a connector issue and that's where you fix the problem.
 
Back
Top Bottom