Power Supply Construction Issue

[omnesilere]

Hello there forum! Maybe you can help me? I've done some circuit bending including an RX-17 but nothing like this x0xb0x.

I bought a kit through ebay from Technology Transplant, saying it was possible to follow the kit instructions at adafruit.

I started the power supply and had little to no issues. However, I did get readings that were a bit off and haven't been able to figure out what is wrong. All the IC's and resistors are in correct. I'm getting 10.8V on the wallwort. The diodes are all going to correct way and are the correct ones. From what I can tell the soldering seems alright but I think also that might be the weakest link here. Someone did use the multimeter on it not knowing anything and said there was a spark so I'm really worried that something might be fried but the readings at the end seem too close to make that an issue. It was unplugged at the time (I'm pretty sure) too so I think the spark might have just been from a capacitor.

Across C3 I am getting 5.0V DC and across C5 I am getting 14.2V DC.
I should be getting the same reading out both C3 and C5 which should be about 15V DC or higher.

I'm not sure what else to check for but voltages off the 6562 are:
1: 11.7V 2: 5.31V 3: 5.31V 4: -0V 5: 5.31V 6: 5.33V 7: 18.18V

I know the out voltage is supposed to be 5.333V but if I turn the trim pot all the way up it wavers between 5.0V and 5.3V.

Is this something I should be worried about? Does it sound like cold soldering joints? Is it possible that a resistor or capacitor was fried and still get that close?

Thank for your replies I hope I am on the right path.

[Brassteacher]

It appears close enough that I would get/borrow another meter to double-check. I've never seen a multimeter that was good at every function it has built into it. I keep three different ones, different brands even, around for this purpose, so that if I get a reading on one meter that doesn't make sense, I'll check it against the other two. Also, make sure the batteries in your meter are fresh, that the probes are clean, and you have a good, solid ground connection to whatever circuit you're measuring. It is also helpful, since your readings are not far off, to let the power supply sit (powered up of course) and have a chance for all the components to get to operating temperature. Say ten minutes or so.

So, basically, I'd recommend getting a "second opinion" before worrying about those readings too much.

[omnesilere]

Thank you for the quick response and sound advice.

Maybe I do need to let it warm up, as I always test right after plugging it in again. I think this meter is pretty cheap - a 'second opinion' is usually a good idea. Thanks again, I'll get back to this after I continue some more work on it.

[omnesilere]

Well I finally got around to finishing the x0xb0x. It runs and sounds GREAT... for like 20 seconds... ugh and then stops. every thing seems to work in the meantime. Then I have to wait a few minutes to use it for another 20 seconds. I hope its a simple fix. No mods or anything either. I wanted to make sure it was good before screwing with it even if its easier to build them in.

[guest]

when it stops working
do the leds no longer light up

check if your 5v regulator is overheating

[omnesilere]

The LEDs go out except the one above the tempo knob which just stays on. There is a buzzing sound when it goes 'dead'. I only have time enough to check it in random mode. The 5 v regulator (Thats the thing with the heat sink right?) isn't getting hot by the time it goes out. The tempo knob doesn't seem to work either, but I have it in random mode - not sure what that entails completely.

Why would it work for a little bit and then stop?...hmmm


Thanks. Any advice available...

[guest]

after it goes dead
check the powersupply
and see if you got good 5v 5.33v and 12v

[omnesilere]

Yea I still get 5.3 and 11.8 volts at J4, even after it goes out. The 5v regulator did get hotter quicker this time. Tried it again and it didn't get hot quick, and this time LED 1 on the top row and LED 2 on the keyboard stayed on, which never happened before. Other lights have stayed on I think I explained before, but usually no LEDs beside the one over the tempo stays lit.

My ICs are all in the right way. Ive checked capacitors and and some transistors. Nothing out of place yet but I haven't been able to re-check everything thoroughly. I don't have an oscilloscope, and would have to learn how to use it...
Ive rechecked my solder joints and my buddy went over them, they all seem good and we retouched those that might have been 'if-y'. I hope its not a faulty part.

I might have hit a button while it was set to bootload mode, but I think I read it needs a USB signal too to actually change anything.

[omnesilere]

So I tried it in keyboard mode and it acted similar. I could get off a few notes, but I didn't have time to reach to the knobs before it froze and stopped. The last note I played stayed lit and the buzzing started. No other functions worked after that.

[guest]

if in keyboard mode
no keys light up anymore
then i would suggest reloading the firmware

if that doesnt work
then perhaps there is sometihing flakey
on the reset pin or crystal oscillator

[omnesilere]

How would I go about reloading the firmware? USB time is it?

[guest]

yes
connect and bootload

[Sokkan]

Yea I still get 5.3 and 11.8 volts at J4, even after it goes out.

*snip*

I might have hit a button while it was set to bootload mode, but I think I read it needs a USB signal too to actually change anything.

Yes, but do you still have 5V? It sounds just like the 5V-line goes out as guest suggested. Only having 5.333V and 12V is not enough.

You cannot really damage the firmware by button pushes in bootload mode, no problem there.

[omnesilere]

Yea the middle one is zero! Not even having the J unctions connected (but the large J3 is). Is that a power supply issue then? Looks like progress... Thanks!!!

[omnesilere]

That is on J4 one is 11.8 V the middle is 0 V and the other is 5.3 V testing with the red and the black is on the 5v regulator.