• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Spec 4 Surprise Overhaul and restoration

So I put the second rebuilt amp in and still couldnt get the damned DBT to stop buzzing. So I went for broke and plugged it directly to the wall. Nothing blew!....I checked the voltages to each amp and get -66.8/66.8 and -66.5/66.5 at both. However, my left channel heated up pretty quickly. Right channel just got warm. Any ideas?...I am a bit confused as to what direction to set the VRs. Top one is a no brainer as two legs are connected. The bottom one (VR1) is less obvious. Are the power limiter VRs to blame here?
 
measure the current idle current. both channels, post it. that's probably why.

watch your idle current for a channel, and carefully adjust it.
more detail could be wrong (CW/CCW etc)
just get the reading showing and carefully and slowly change the pot position. NO BIG changes!! Give it time to settle, thermally settle, takes minutes to do...

don't fool with the power limiter pots...
 
Last edited:
Sorry I havent posted these currents yet Mark...I blew the 6 amp fuses yesterday (actually 1)...my hardware store only carries automotive, so I had to order some from Amazon. Will get back to you as soon as I can. In the meantime, I am going to pull my outputs and test them. I hope I didnt damage them when one channel got so hot. I am also thinking about recapping the meter board while I wait. Does anyone know offhand what transistors need to be replaced on this board?...was going to replace all the power trannies anyway. Thanks.
 
Another quick question. I am trying to test and dial in one amp at a time, but I cant get a good DBT reading because of the relay buzzing. Light flickers like crazy. Tried every wattage I got. However, the DBT behaves normally with no amps hooked up. Is this an indication the amp still has a short?...
 
NO dbt.

Are you trying to "tickle the dragon"? If that is gobbledygook to you, NO.

This distortion of the use, and the overuse of DBT's is making me sorry people ever heard of it.

It is a seasoned tech trick, useful only in certain circumstances.
 
So how can I check my work without frying 110 bucks worth of new components?...can I at least check the voltages without the outputs installed?...I did manage to ruin one 96G when I powered on, as I feared. But only one, which is surprising. If I could measure voltages without them installed, I could post them and maybe determine if there is a problem first.
 
When I rebuilt my SPEC-4, I used a DBT to check everything was okay. I needed a bulb wattage of around 300W to get it to pull in.

The relays in the centre of the SPEC-4 will chatter like crazy.

It sounds horrible but all that is happening is:

1) The relays are energizing, closing their contacts and connecting the second amp channel power supply.

2) The second amp power supply starts to charge quickly, pulling the voltage back down under the operating voltage of the relay.

3) The relay now opens, the voltage now jumps back up to working voltage.

4) The relay energises again and the second amplifier power supply is reconnected again and starts to charge again. Back to 3) until the main filter caps are nearly charged.

In my experience, using 200 - 300W of lamps means after a few seconds the power supplies are charged enough to stop this behaviour and then everything calms. That is, IF your output transistors aren't shorted.
 
Thanks Qsilver, I did buy a 150 Watt bulb but it has not arrived yet, so I am using 100W with one amp installed...I am just trying to get one tested/adjusted at a time. Would it be safe to bypass the relay and disconnect one side of the power supply to at least get working voltage to one amp to test these junctions found on the scat?...also, I fear I did not match hfe on Q8, Q9 and Q10, Q11 now that I read the scat again. How disastrous is this? I matched the 992's and 1845's. Well, back to the soldering iron.
 
I wouldn't go fiddling with the relay myself.... feels like asking for trouble. I assume you used transistors from the same gain code for the transistor positions. If that's the case (and it probably is if you bought them all at once from one place), then it won't have much effect. The DC offset adjustment will counteract any mis-matching at this level. The gain matching Pioneer refer to AFAIAA, is getting transistors that are near like 300 and 350 rather than 300 and 100 for example.
 
Yes, I bought 20 of each 1831, and 3505 at the same time. I still think I should check but, as you know, the more times you release components from the board, the more damaged the pads get. The MJE's I had bought at different times, so that concerns me. Also what worries me is the power limiter VR's. I was told to leave these be when rebuilding the boards, but wouldn't these cause issues if you replace everything with new components?...should I set them to zero also?...I have a buddy on here that is going to do the limiter adjustment for me because he has the necessary equipment, but I want to get him a DBT passing unit at least.
 
Until I can check (reverse engineer & actually get hands on) those power limiter circuits, I don't know what setting a pot at an extreme will do.

One wants to go to the turned off condition (your "set them to zero"), not the maxed out condition.

I suspect that is the minimum resistance from the pot's wiper to the output terminal pin 11 .

At least the pattern, while mirror imaged, does NOT mirror image the pots. CW and CCW are consistent.
 
Thanks Mark. I will look into it also. In the meantime, I am going through my resistor values again and validating. I have new 1/4 watts to change all out just to be safe. Will also try to match my gains on the transistors I missed in the manual. Check and re-check while I wait for Mouser to deliver my new output.
 
Quick question.. I dont have any matching MJE's for the amp boards!..I have 10 of each and they are all 109 vs. 233...can I still use them?...how far apart can they be in value?
 
So, I managed to get 1 amp working and it comes out of protection...but I can only dial the idle down to 83mV before the VR ends. Manual says 50mV. What can I check?
 
For future reference, I had to turn VR2 (25 turn) all the way clockwise to bring the idle close to specs. This is probably why I burnt up one output when I was told to start everything at CCW. Worked for me anyways. The outputs were only warm after about 5 minutes of power up. If I could get that idle down to 50mV, I would be happier.
 
Ok, measure the voltage across R15 (R16 right). one dmm probe on opposite sides of the resistor at the same time.

R15, R14, R13, D1 & Q5 form a constant current source that is fed to the STV diodes for a particular voltage across it at a particular temperature. Drop the current, drop the STV voltage and thus drop the idle current.

voltage divided by resistance is STV-x current. EXAMPLE: 1.35 v / 220 = 0.006136363636 or about 6 mA
if we go to 240 ohms, 1.35v / 240 ohms = 0.005625 5.5 milliamps
if we go to 270 ohms, 1.35v / 270 ohms = 0.005000 5 milliamps

which will drop all the stv-x voltages correspondingly. Which will drop the idle current range adjustable by the VR.

This all depends upon your R15 (R16) voltages and currents. What are they?

The question is, why did this happen - what did you use for Q8, Q9, Q10, Q11 and the outputs?
modern transistors (on-semi it seems) have dropped their Vbe's a few millivolts, and it all adds up rather quickly. They are apparently shifting their purchased Fairchild transistor lines into their proprietary process which is different (better?) than the old Fairchild process.
I tried to avoid the on-semi's that did this in my replacement lists, but that change is becoming too widespread.

edit - r16 brain fart...
 
Last edited:
Hi Mark!...Happy Easter!...let me digest this a minute and get back to you. By the way I got amp number 2 up and running. Perfect idle and bias. I am still waiting for my new output (mouser is slow). Swapped one for now so I need to put it back in the other amp to run this check...BRB
 
I used 3503/1381 combo for Q8 and Q9, tested my original SD608 and SD628 and re-used them since they tested good. And their Vbe were a match. I didnt have any MJE's that were even close. Let me look into that R15/16 value and get back to you.
 
I really wish Mouser had a way to order matched pairs of transistors...would have saved me a lot of time.
 
You must log in or register to reply here.
Back
Top Bottom