Arcade Otaku - アーケード オタク

I have a Nano MS9-SU on the CRT for my Astro City cabinet that recently died and I think the transformer on it is bad. I checked B+ voltage and was getting 0v so I took the board back out and connected it to A/C power on the bench so I could troubleshoot. I checked I had 120V A/C going into the rectifier at BD901, and it seems to only be putting out ~68V DC when the manual says it should be 75V. Not sure if that is a problem or not. Next I checked the big cap at C911 (while it was off the board) for capacitance and get ~875uF when it should be closer to 1000uF. I have a new one on the way but wanted to do diagnosis in the mean time in case I need to order more parts. Lastly I checked the transformer at T901, on the primary side I'm getting ~68V DC going in and on the secondary side I'm getting 0V on all pins. I even checked post-diodes and still get 0V.

I desoldered the transformer from the board and checked all the windings. None of them appear to be open, and the results I got testing it with a meter reflect what I can see on the ms9 schematic. I desoldered pin 2 on the flyback and put a load on B+ using a 40W light bulb and got nothing, no light and meter reads 0V. If I connect the light to the primary side it lights up and I get a reading. So I guess the transformer is bad? I attached a picture that shows the points I tested and the voltage readings for reference. Any help is appreciated, thanks.

Switching transformers fail only under exceptional circumstances like corrosion caused by soda spills or rat ****. So the advice to a novice is don't suspect the transformer in a switch-mode power supply.

I think you're reading 68V because you are using the ground of the secondary side aka cold ground or chassis ground (all the heat sinks are connected to this ground on the MS9). Voltages in the primary side are referenced to the hot ground. This is the negative pin of the bridge rectifier, the negative of C911 and one side of R903 (probably the best point for attaching a hook clip probe). With this ground the reading will be around 140V. And there's no DC voltage on the output pins of a transformer, only at the cathodes of the rectifying diodes.

When you find there's no voltage on a power line, the first step is to check the resistance to ground of that line. If you have zero or low ohms there will be some part shorting the line so it would be pointless to connect a light bulb. Typically, when the HOT (Q506) fails, the collector and the emitter (which is grounded) are shorted and this causes zero ohm on the B+ line
which connects to the HOT collector through the flyback primary winding (pin 1-2). As you disconnected the B+ from the flyback, the HOT even if shorted wouldn't cause a short on the B+ so a good power supply should light up the bulb and give a correct voltage reading. Call me skeptical but I would need to see how you're doing the bulb test before questioning the power supply. You need to use an old style incandescent type bulb. Newer bulbs don't work.

Thanks for the reply. You're right, I was using the secondary side ground to test the rectifier. I retested using the +/- of the rectifier and I'm getting 170V. When I was testing the secondary side with a light bulb I was connecting to TP2 and the negative side of C515. The light bulb is definitely incandescent, I had to run out and buy one because I didn't have one on hand. FWIW I removed the transformer from the board completely and connected only the primary side using some spare wire. Then I connected the light bulb to +/- terminals for the B+ winding and get no light. If I connect it to the +/- side of the primary side I get light and 160V reading on the meter. I attached two pictures for how I performed these tests. Ignore the diode, I tried with and without it and it makes no difference. Am I performing the test wrong?

Yes, put everything back on the chassis and use the heat sink for ground and TP2 or R967 for B+.

I already did that before I ever removed the transformer from the board. The negative side of C515 is connected to the same trace the heat sink is connected to as is CN509 (which I also used as well), it makes no difference which point you use. No matter which point I use, the light bulb does not light.

To get DC voltage from a transformer you need a rectifiying diode *and* a decoupling cap. The way you did the test in the above pic isn't correct as you only have the diode. It may be entirely possible that the power supply is at fault but until I see a test done properly I can't be sure. There's no reason to keep parts off circuit anyway.

PrincessPrinPrin wrote: ↑May 16th, 2022, 3:24 pm To get DC voltage from a transformer you need a rectifiying diode *and* a decoupling cap.

That's the info I was looking for. The point of isolating the transformer from the secondary side is to remove any variables it introduces. There could be a number of things wrong with the secondary side circuits, but by testing the transformer while it's isolated you know if it tests good it's probably fine, and if it tests bad it's probably bad. Like I said earlier I already tested B+ three different times and the bulb never lights and the meter never gets a reading, I don't see a point in putting the transformer back on just to take a picture showing I can connect the light bulb properly.

Anyway, since I'm fairly sure the transformer is bad I bought another MS9-SU chassis off YAJ. Once it gets here I will test B+ on that to confirm it works and then swap the transformer over to this board and report back.