diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Pass Labs (http://www.diyaudio.com/forums/pass-labs/)
-   -   Summer / Consolation Amplifier: 1995 Adcom GFA-5200 (http://www.diyaudio.com/forums/pass-labs/197406-summer-consolation-amplifier-1995-adcom-gfa-5200-a.html)

Andersonix 27th September 2011 06:59 AM

Summer / Consolation Amplifier: 1995 Adcom GFA-5200
 
1 Attachment(s)
My two DIY amp projects, F5 and BA w BJT outputs have stalled for now, but look what fate (and craigslist) brought to me: a cute little Adcom GFA-5200 (an all-mosfet amp which NP designed, so that's why this post is here). I especially wanted the 5200, as it has a single output transistor per rail (whereas the 5300 has duals) and a single pcb with lots of empty space to work around. This will make a nice little summer amp to drive the top of my ESL hybrids, especially once I've completed my to-do list.
After a major vacuum, minor check-up, and revealing photo-shoot, I've punished it for over a week (making sure to test and re-test the thermal protect) playing the SL3's full-range. So far, no real guts in the bass and not so silky highs in stock form, but it shows promise.

Any chance of someone showing me a schematic?

Here's my immediate wish-to-do list:
Replace all the electrolytics
More PS caps (in the space thoughtfully reserved for them)
Thermistor to protect switch
New rectifier bridge made from 15A soft-recovery On-Semi MSR diodes and moved off-board
Re-route input leads (as they now take the scenic route close to the transformer!)

Long-term:
More heat sinking (and maybe redesign the cover to yield more space)
Bottom-mounted fan in silent housing
Transistor clamps + Kerafol insulators
More bias
Transformer-shielding and chassis-bracing steel plate

Andersonix 9th November 2011 03:57 AM

1 Attachment(s)
Stage One was completed about a week ago (see pic): I added two old 31mF caps to the PS, swapped all the electrolytics for a mix of Elna RFS and Panasonic FC, and squeezed in a 2u2 WIMA as input cap. Bass is noticeably tighter, but the highs are still not so smooth.

And Stage Two was completed a couple of days ago: I took the rectifier bridge off the pcb and made another bridge from 8A OnSemi MSR diodes. Also, the bias and DC offset pots were changed for Murata multi-turns. Now it sounds better in the highs, and I expect it will get even better when I add some heat sinking and raise the bias. Over the next few days I'll raise the bias a little with the stock heat sink and hear what happens....

Does anyone know the factory bias setting? It's currently set at 0.11A.

I'm considering swapping the 3W 0R1 emitter resistors for 5W Ohmite 10 Series axial wire element current sensing resistors (as they're available at DK and in my price range).

Then when Apex Jr gets in some 50V Panasonic caps I'll replace the four 6800uF in the PS and add a 0R25 in a CRC. And I have some old heat sinks to add to the stock hs and make another cover (and forgo the fan).

jsfinusa 9th November 2011 09:22 AM

I'm interested in your project. I am impressed that you squeezed those large electrolytics in that small package. I bought a beater 5200 off eBay and I am trying to get it operational. If I can get it working, I might try some of your upgrades. I don't suppose you have a schematic for this thing? I've searched in vane and found nothing other than an incomplete service manual from hifiengine. I'm forced to use the schematic for a 5300, which is similar. Good luck and I'll be following your progress.

Nelson Pass 10th November 2011 01:47 PM

I would bias it until the sinks run 50 deg C after an hour or two. If you don't
have a thermometer, this would be the temperature that you can leave your
hand on the sink for 10 seconds.

:cool:

Zen Mod 10th November 2011 01:55 PM

1 Attachment(s)
>ZM ; JB mode On<

Gotch ya !

<ZM ; JB mode Off>

Andersonix 13th November 2011 06:42 PM

1 Attachment(s)
Quote:

Originally Posted by jsfinusa (Post 2775147)
I'm interested in your project. I am impressed that you squeezed those large electrolytics in that small package. I bought a beater 5200 off eBay and I am trying to get it operational. If I can get it working, I might try some of your upgrades. I don't suppose you have a schematic for this thing? I've searched in vane and found nothing other than an incomplete service manual from hifiengine. I'm forced to use the schematic for a 5300, which is similar. Good luck and I'll be following your progress.

I'm glad you're interested, jsfinusa!
No, I have not seen a schematic for any of the Adcom mosfet amps in this series.

A future option is standing up the PS caps, so they will balance the look once there is more HS added (in the manner shown, but with an additional HS opposite it).

Andersonix 13th November 2011 06:48 PM

Quote:

Originally Posted by Nelson Pass (Post 2776526)
I would bias it until the sinks run 50 deg C after an hour or two. If you don't
have a thermometer, this would be the temperature that you can leave your
hand on the sink for 10 seconds.

:cool:

Thank you, Nelson! I think I'm pretty adventurous (and patient!) with dialing up the bias, so I'll report back my findings. (I will probably put in some Kerafol insulators before getting too wild...)

Andersonix 22nd December 2011 11:53 PM

1 Attachment(s)
Bias is at 0.25A (stock was 0.11A), and it sounds much better in the highs, but still not there somehow. So I'm going to R&R the 3W sand cast 0R1 emitter resistors. Would it be OK to go even lower in value here, say, 0R15//0R15 = 0R075? (That opens up options of using 2W resistors.) What is the emitter resistors' purpose?

Since this is an AB amp, I can't quite yet turn up the bias as far as NP suggests, as it gets quite hot while playing loudly. So I need to add more hs (and/or install a fan, which I am reconsidering).

While bass control has improved with the extra 31mF caps, since I had them on-hand, I put in a couple of 68mF caps instead (as shown), so it now has six times the original PS capacitance. This further improved bass control, but I feel there's still room for improvement. And it makes me wonder why Adcom chose to use so little originally....

jsfinusa 18th August 2012 03:46 PM

Andersonix

I know this thread is a little old but I wanted to follow up and say I finally got my beater 5200 working. It took replacing all output devices and several small electrolytics to restore operation.

In the discussion about bias, I set mine to what is called for in the service manual: 20 mV across 0R33 (R035/R135), which is about 60mA. That is about 1/4 of where yours is set. It makes me wonder if I should set mine a little higher? My amp sounds very good and seems to work fine though it does get quite warm after 1/2 hour or so.

BTW, I used the schematic from the 5300 and it is close enough for troubleshooting the 5200. At this point, I will live with mine the way it is though I may eventually follow your lead with some of your tweaks.

Thanks for posting your mods and pics.

Andersonix 6th October 2012 09:38 PM

Well done on getting yours working, jsfinusa!

If you are referring to the Adcom service manual for the 5200, then beware: Adcom was sloppy in editing the service manual for the 5300 to create the one for the 5200, and there are errors. Note that the source resistors (R035, R036, R135, R136) in the 5200 are 0R1 (on mine at least) and not 0R33 as in the 5300. Also, while the part numbers are correct, the biasing directions (20mV across R35) are taken from the 5300 and if followed will lead to more bias than intended. The bias on my 5200 when I got it was 0.1A. In your case, jsfinusa, 20mV/0R1 is 0.2A of bias...

With the stock heat sinking, I biased my 5200 to 0.25A, and it ran warm with the cover off. Now, with more heat sinking added, I have it adjusted to 0.4A (28mV over my new 0R07 emitter resistors), and it runs warm to hot.


All times are GMT. The time now is 03:34 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2