My very own Beta 22 thread

[spritzer]

Spoken by a true member of team arcy sparky

[Smeggy]

I finally got the resistors, replaced the broken pair and now I'm reading +39/-37 V DC on each of the three pads. Any Idea why the reading would be so high. I'm getting 30V AC in.

[Beefy]

That sounds like it is not regulating. 30VAC is ~42V when rectified to DC, minus some losses, 37-39V output is probably appropriate.

Well beyond my skill level to debug it though

[inukage]

so it hasn't re-fried yet? interesting

whoops, just read your chat post.... sorry bout the troubles your having.

[Smeggy]

well it was just the power fuze when I connected a board. Replace the fuze and it's fine again.

Now worried about connecting the boards again. I'll try a different board

[Beefy]

Now worried about connecting the boards again. I'll try a different board

Yeaaaaaaahhhhhhh, I wouldn't connect ANY amp boards to it until I had the output voltage sorted out correctly.

I would also do some checking of things on the amp boards as well. All the resistances to ground, cap orientation, resistor values, etc etc.

[Smeggy]

the boards all measued fine before the resistors on the sigma popped before

[Smeggy]

Ok, It has a near perfect 30V AC going in but totally screwed V DC numbers coming out.

I officially have no idea what's wrong with it or how to fix the high output voltages or fix the voltage imbalance. I'll see if I can find another S22 to buy. This is above my level of DIY.

[Pars]

This might help:

http://www.amb.org/audio/sigma22/s22_30v_voltages.pdf

The AC voltage going in is somewhat immaterial. As was pointed out, 30Vac * 1.414 = 42.42Vdc - 1.4Vdc drop across rect. diodes = ~41Vdc which is then regulated down to 30Vdc. Checking thru the operating points should get you pretty close to isolating where the problem is. The PSU should put out +/-30V unloaded, and should be put to that point before any of the b22 boards are hooked up. Then as was suggested, one at a time for troubleshooting.

[Smeggy]

I tried using the pdf but I can't follow what is where on the board and what each bit is.

[amb]

Power off and measure the resistance across any pin of the sigma22's MOSFETs (i.e., D-S, G-D, G-S). You should get very high (~infinite) readings after the meter stabilizes. If you get low ohms readings then one or both of the MOSFETs on that bank is internally shorted.

[Smeggy]

How do you do that?

It appears both MOSFETS on that side are blown too. Is there anything else I should check before ordering more parts. Can I buy those from you?

[guzziguy]

You can buy the mosfets from AMB according to the sigma page.

[Smeggy]

Yeah, I was just looking at the order page. I'm just going to run my meter over a few more resistors etc to see if I should get anything else.

R4 and R5 are measuring 55 Ohm, R6 and R7 are measuring 100 Ohm. Two more things for the order.

Edited April 26, 2009 by Smeggy

[amb]

How do you do that?

Each MOSFET has 3 pins, you set your DMM to ohms mode (if manual ranging, set to 2K ohms scale). Put the probes on any combination of two pins (there will be six possible combinations) and read the meter display. The meter might show a brief reading due to capacitor charging but should settle quickly to over-range (either "OL" or "1", depending on brand/model of DMM). If you get an actual ohms reading of several hundred ohms or less, then a MOSFET is blown. It's difficult to determine which of the two is blown while they are mounted on the board (because they are in parallel), but if one is blown then there is a good chance the other one is too, and might as well replace them both.

Is there anything else I should check before ordering more parts. Can I buy those from you?

Check all the BC546B and BC556B transistors too. With the power off, use the diode check function of your DMM to verify each of the small signal transistors. For all the BC546Bs, put the red probe on the B pin and the black probe on the C or E pins. Both measurements should give you approximately 0.6V. You should read nothing when you reverse the probes. For the BC556Bs, the procedure is the same but the probes must be reversed.

The only transistors that should falsely "fail" this test are across the B-C junctions of Q8 and Q10. This is because they are purposely shorted (see the schematic), so you will get 0V reading for these.

Yes, you can get replacement transistors and MOSFETs from me.

[Smeggy]

Heh, I meant how did you know the MOSFETs were probably shorted. I suspect dark powers are involved

looks like R8 is only reading half what it should be too and the smaller transistors all seem to be in the 0.67 V area.

Thanks again for the help AMB

[amb]

Heh, I meant how did you know the MOSFETs were probably shorted. I suspect dark powers are involved

No dark powers. lol. Just some logic and experience. We know that your 0.47 ohm resistors are blown, so they've been subjected to overcurrents. That current goes through the MOSFETs, so there's a chance they're bad too. These MOSFETs typically fail with internal shorts.

The question is, what led to this failure? Either there was an accidental short circuit (in the wiring to the amp, or from other tools/objects), or a wiring error, or a short on an amp board..., etc. So be careful after you fix the power supply, to check the sanity of the amp boards and wiring before power it back up again.

looks like R8 is only reading half what it should be too

It should be 10K, and there is no reason why it shouldn't measure close to that. Try removing it and measure it off-board. If it's still not right, put in a new one.

[Pars]

I was hoping that Amb would pop in to help you

You might take your DMM on ohms mode and measure the V+ and V- to ground on your 3 amp boards with them disconnected. They should not measure shorted (0 ohms), but I'm not sure what a good reading would be, and it would probably depend on probe polarity.

[Smeggy]

Thanks guys, I'll go through them later today

[Smeggy]

*UPDATE*

So I have a kinda working B22 now. Of the three boards, one seems to be shot, two read properly but only one is making sounds. Great success!

I dunno whats going on with the board that measures well so I'll dig deeper tomorrow, could be pot or input selector or any number of things.

[Torpedo]

Congratulations and sorry hahahaha. Good luck spotting what's wrong on those two not properly working boards.

[Mister X]

http://www.youtube.com/watch?v=QUSDS9pkA2Y

Whooah, you're half way there

[Smeggy]

Bah!

I'm gonna do some troubleshooting of the simple stuff tomorrow and hope something obvious jumps out and hits me in the face.

At least the S22 and one board is working right

[Nebby]

Is it too late to say I found the epsilon board yesterday? Progress is looking good, looks like you didn't need it anyways

[Nebby]

hope something obvious jumps out and hits me in the face.

Hopefully that something isn't a cap gone nuclear.