Roll Call: Who's building, built, or thinking of building a beta22?

[luvdunhill]

Quote:

Originally Posted by cfcubed /img/forum/go_quote.gif NP test scale

I didn't want to invoke the almighty name in case amb had another opinion, but yes indeed you caught the reference quite well

 

[cfcubed]

Trying to build up a full 4-ch b22 & have run into a snag on one of the boards. I try not to call for help, but would appreciate some direction here.

During setup cannot get board to maintain > 1.8mv across R34/R35 (R9/R11 = 4.5v, R10/R12 = 4.8v).

As I've an identical perfectly working board, I moved VR2 on the problem board up a few turns - close to where VR2 is set on the working one.
Problem board starts out exactly as all my other b22 boards do - like 80mv settling to 60mv BUT then falls to 1.8mv (all within about 3 secs).

What I've tried:

1. 3X review of all parts & orientation. Full check of board top & bottom w/magnifier. Full reflow of all joints, wash & more visual checks.
2. Compared resistances of top exposed leads to gnd, V-, V+ to working board. Allowing caps to equally charge between the boards, measurements are comparable between them... IOW don't think there are any unintended opens or shorts..
3. Since Q25/Q26 come up a lot here, pulled tested & replaced them.

My guess is a temperamental/bad part... In the old days I would have reached for Freon spray & tried that on the semis.
About to embark on checking live Vs but it'd be great if I knew where to focus my efforts.

 

[amb]

cfcubed, what you describe does sound like a thermal-related issue with a certain part. A cold spray here and there sounds like a good idea. At any rate, also check R25-R27, Q17/Q18 and the 0.47 ohm output resistors. With power on, measure the DC voltage across the wire jumpers at R30 and R31 as you turn the VR2 trimmer clockwise, that voltage should increase. Don't let it go much over 7V without also checking the voltage drop across R34/R35 to make sure you don't overbias the MOSFETs. If that voltage does increase to 7V without a problem, then the problem is isolated to the output stage alone, otherwise, it could be Q17/Q18 and surrounding parts or further upstream.

 

[cfcubed]

Thanks a bunch amb... I'll give your suggestions a go this weekend.
Hopefully I can pick up some sort of cold spray @ a decent elec store nearby as well.

 

[LeMat]

Finally finished my 3 boards B22... fired up... 2 boards working perfectly but the ground channel .. quite the opposite. Leds are on, r9 is 4.5v but i have 0mv across R34... damned !
If there is something obvious coming to your mind, please share ! Otherwise, what's the better way to debbug it / narrow down the possible issue ?

 

[amb]

LeMat, tried looking back at past posts in this very thread?

 

[cfcubed]

Quote:

Originally Posted by amb /img/forum/go_quote.gif ..check R25-R27, Q17/Q18 and the 0.47 ohm output resistors. ..wire jumpers at R30 and R31 as you turn the VR2 trimmer clockwise, .. If that voltage does increase to 7V without a problem ..otherwise, it could be Q17/Q18 and surrounding parts or further upstream.

Could not find bad part using cold spray (well, upside-down RS dust-off - same thing)
R25-27 & 0.47R resistors checked out fine (w/a little variance for being in-circuit)
R30 - R31 Vdiff starts @ 8v within 4 secs drops to 2.5v & VR2 will not raise it.
Replaced Q17,Q18,Q12 & Q16 w/known good spares

All my actions to date on the board have not changed its behavior.

The "worst" variance in operating Vs for this board is on the right side (Q22/Q24), most notably V @ Q22 gate drops from 3.3v all the way to 0V.
Seems something is dragging down the whole line on schem from Q16 collector -> Q22 gate to 0v after 4 secs or so.

 

[amb]

cfcubed, check the rail voltages on both sides of Q25 and Q26 before and after your 4 second interval to see if the supply rails are still where they should be.

 

[cfcubed]

Rails are holding fine:
Q25 coll=24v,base=22.5v,emit=22v...Q26 coll=-24v,base=-23.4v,emit=-22.6v
with +/- 24v PS after 3 mins (PS running cold/little draw)... As a reality check swapped in working board & all is still fine w/it & PS is not cold powering it (expected draw)

Note: I've a yet-to-be-build b22 kit I can pillage for parts if we want to try things

 

[amb]

There has to be a bad part somewhere that only goes bad when powered up for a short moment. Time to do more exhaustive voltage measurements and compare to the operating points PDF file found at the β22 website's "initial setup" section, under the "Troubleshooting" heading. Maybe that will help identify what's going on.

 

[oneplustwo]

What's the best way to install a zobel network for headphones? My thought is that it would be easiest to just go from the positive and negative tabs of the headphone jack to the ground tab. Of course, with speaker posts, it's straight forward, but I'm wondering if folks have had good luck with a particular method installing for headphone jacks. (This is for the beta that I had problems with oscillating with very low impedance headphones.)

 

[amb]

Directly on the headphone jack would be best, but depending on the model of jack you have, it may or may not be easy to do.

 

[simwells]

Looking into building one of these once I finish my Gamma2, I'm somewhat confused as to the actual advantage of the active ground version over just the two board design. Would be for headphone useage and I don't want to go as far as a balanced unit. Also am I right in thinking I need a sigma22 per beta22 board?

 

[oneplustwo]

No... one sigma is enough for 2, 3 or 4 beta boards. Usually, people use 2 sigmas for 4 in order to distribute the load and manage heat. But 2 or 3 board setups typically use 1 sigma. At least for headphone use.

As for the advantage of an active ground, some people say it's significantly better than a 2 channel, others not. You can search for previous discussions on this topic, but IMHO, the decision is probably based on things like budget, enclosure size, etc as much as your belief on the merits of active ground.

 

[DemonicAngelz]

Hi guys. I've read this thread all the way from post #1 to #2384 and I'm seriously contemplating building a two board b22 with a s22. I've built the pimeta a couple of cmoy and the mini3 from amb. However I am relatively lousy at troubleshooting if any problems pop up. I am willing to learn of course. Time is not an issue, have lots of time to spare from now till october.

I intend to upgrade by building another set of 2 board b22 and s22 running balanced out to my hd650. Doing this way because i'm not totally confident that I can fix up one, if I happen to give up half way it wouldn't be too expensive either.

Do you guys think this is a good idea for me to start off or should I hold off till I have more experience?