Phase Linear 700 series II restoration

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Recheck all of the wiring you replaced, make sure all of NPNs and PNPs are in the correct locations, diodes in correctly. On a 400 (not sure on the 700) the DC fuses supply only the output stages so if those are removed and the Variac can be turned up without any problems you'll know the problem is in the output stage. It'll be something stupid, it always is. Oh did I do that? If I remember correctly the 700 has separate DC fuses for each channel.

Craig

Hi Craig,
Many thanks for dropping first thoughts. Some remarks to your suggestions:

  • I'll do the recheck again, but for 100% NPN/PNP outputs are installed properly.
  • There were no flyback diodes at all the model restored - see the previous posts.
  • Very good point on isolating the output stage with fuses taken out - didn't think about it :)
  • Yes, there are four separate DC fuses, two per channel - have a look pls at post #6 here with schematics attached.
I will confirm the situation bringing up the amp with no DC fuses in place.

Regards, Przem
 
Just checked - with no DC fuses, no current drain.

One more info to share is that before initializing the tests all outputs were checked in line with step 6-1.3 of service manual and no probs found. Also isolation of TO-3 sockets from chassis was OK. Maybe something wrong is with Vbe multiplier. I see from previous repairs that right channel uses BD139 while the left one 2N3404. Still measuring them with ohmmeter no warning readings observed.

I will now try to put in DC fuses channel by channel to see the outcome...

Any other directions to pursue?
 
Reason found and eliminated: broken leads at both bias transistors - again difficult to spot. Things back to normal. Thanks to the light bulb current limiter I saved newly installed outputs - a good lesson to remember!
Tomorrow final stage of bringing up the amp to normal operation, if no more surprises :)...
 
What a coincidence that was the last problem I had to find and repair. After a first initial run and adjustment everything looked good, screwed everything back together and it was drawing current. Leads broken on the bias transistors, that's when I decided to replace all of the PCB interconnect wires with something more flexible. I even went thru the trouble of drilling holes thru the solder pads and inserted the wires from the other side of the board. Tracked down some those once common transistors and everything was okie dokie again. Good to see you tracked it down, hope all goes well for the second initial firing.

Craig
 
Phase Linear 700 II - excessive DC offset

OK, need some help and guidance here. Didn't go as expected... :-(

I have way too much DC measured on outputs (ca. 40VDC). Cannot find the reason. Testing configuration:
- variac 0-240 VAC,
- light bulb limiter (60W/230V) in series between the vaiac and the amp,
- no inputs,
- no load,
- the amp doesn't draw current,
- outputs are good (re-checked)
- drivers, too.

What I might do wrong? Where to start diagnosis?

Tx a lot for all valuable advice!

/Przem
 
Perhaps the variac is not turned up high enough to get the circuit stable?

Hi Netlist,
Thank you for the advice. The VDC measured on big PS caps is slightly over 80VDC when variac is turned up to the max level (230VAC). I thought a good level for testing. However while measuring across each zener diodes (D1, D2) the reading is 0.7V B+ pole and 2.4V on B-. Maybe the reason for unbalanced voltage is that I hooked up speakers relay circuits to the B+ pole (takes ca. 80mA of current)? I will disconnect these two units and if that was my problem - another supply source needs to be designed.
Your view?
 
Did you stay with the LF351? That's the only thing connected to the LVPS (zeners), do you have good voltages into the dropping resistors? IC in backwards? Both channels acting the same? Until you have proper voltages to the IC you'll have DC offset. If you need more current from the LVPS for the relay circuit you'll need to adjust the dropping resistors, less Ohms.

Craig
 
Last edited:
Here is the next 'report':

Still light bulb limiter in the AC supply circuit - I am looking for 100W traditional bulb, but in EU that value is now not allowed (green policy...). Need to search a bit.

step 1:
- replaced D1 and D2
- replaced Q103 and Q203
-> result: proper offset level on right channel, the left one still too much DC (27VDC at 80VDC on PS caps), across each Zener diode similar voltage :)

step 2:
- temporarily replaced (only for re-check purpose) Z101 and Z201
-> result: no change, left channel gets DC

What else to check/replace?
 
Last edited:
PL36 ver.full complementary transistors equivalents

Had more time with 90VDC on PS caps - the PL36 PCB gets warm. Is that normal? Measured LVPS for IC across zeners - too low (1.8-2.4V) - why?
Doesn't look good... I think about 'en block' replacements of all transistors and diodes on the driver board...

Help needed, please!

In the meantime a question: I see MPSA18 on the PCB - is MPSA43 a good replacement? Or BC639?

Tx a lot...
 
Przem,
Yes, it is normal for certain parts of the PL-36 to get warm. I noticed in one of your pictures that the original Class-A transistor w/ steel heat sink has been replaced with a substitute transistor w/ a much smaller heat sink. This heatsink is too small and the transistor will run fairly hot. I would recommend a larger heatsink with more area. At the moment, this is not contributing to a problem your are having, but it may in the long run.

In order to troubleshoot the Zener circuit, I would take out the 4 output rail fuses. If you do not have sockets installed for the input op-amps, I would do that next. This makes taking the op-amps out very easy. If they are good sockets, you do not have to worry about any kind of degradation from the socket. I use sockets all the time and have never had a problem. With the output rail fuses and op-amps out, you can now troubleshoot your Zener power supply problem.

You say you have replaced both Zener diodes? 7.5k dropping resistors are good? At what location is the MPSA18 (Q103)? I don't recall this transistor in the original amp.
 
While troubleshooting, you may also want to disable the protection circuit temporarily by unsoldering one lead of D101 / 105. This will be one less area to be concerned with until you get the amp stable and working correctly.

It is also very easy to chase your tail when repairing amps... if the amp has a failure and you know it has been repaired, most times it is best to replace all the diodes and transistors with new ones (and the correct ones!) since there are not that many of them. Some people would argue that this is not true troubleshooting, but in my experience, when bad components are left in the circuit, and new components are installed, the new parts may be damaged. All it takes is a weak diode or transistor, and these components may check OK with a meter, but turn on the power and they show their symptoms.
 
Przem,
Yes, it is normal for certain parts of the PL-36 to get warm. I noticed in one of your pictures that the original Class-A transistor w/ steel heat sink has been replaced with a substitute transistor w/ a much smaller heat sink. This heatsink is too small and the transistor will run fairly hot. I would recommend a larger heatsink with more area. At the moment, this is not contributing to a problem your are having, but it may in the long run.

In order to troubleshoot the Zener circuit, I would take out the 4 output rail fuses. If you do not have sockets installed for the input op-amps, I would do that next. This makes taking the op-amps out very easy. If they are good sockets, you do not have to worry about any kind of degradation from the socket. I use sockets all the time and have never had a problem. With the output rail fuses and op-amps out, you can now troubleshoot your Zener power supply problem.

You say you have replaced both Zener diodes? 7.5k dropping resistors are good? At what location is the MPSA18 (Q103)? I don't recall this transistor in the original amp.

THD, many tx for responding!

Yes, the heatsinks are not the same as in earlier models, still it is factory mod in fully comp models. Such heatsinks are obsolete now, but I agree - need to find a better cooling stuff for the mirror transistor.

Great idea - i will pursue the problem in line with your guidelines!

Yes, both zeners are new now and so are the dropping resistors (old were indeed drifted).

Here is the location attached (the picture is taken before repairs started). I think Q102...
 

Attachments

  • PL36 comp version.jpg
    PL36 comp version.jpg
    158.6 KB · Views: 381
All it takes is a weak diode or transistor, and these components may check OK with a meter, but turn on the power and they show their symptoms.

I believe that's exactly the case here... Writing my shopping list then, still I have some new parts for replacement for PL400/700s2 amp, but indeed for instance those MPSA18 surprised me as well...

Thank you again, Przem

P.S.
on the previous picture I forgot to mark electrolytic caps - all replaced too...
 
Last edited:
PL36 refurbished

To continue my 'progress report':

  • output rail fuses out and OPA IC out too -> across each Zener diode: +/- 15V;
  • left channel rail fuses in and no OPA IC -> LVPS network with +/- 15V;
  • both rail fuses in and no OPA IC -> as above.
Consequently, my focus was PL36 components' problem. I decided to follow THD+N advice (thank you!) and replaced all diodes and transistors (picture attached). Q106/206 exchanged into MJE340 with better/equal heatsinks. One diode was soldered wrongly before I put the new one. I found one resistor with cold solder and slightly burned (2.7K) and two resistors with 4.4K instead of 3,9K (for the moment I've put two .25W in parallel 7,5K+9,1K as could find the proper value in my drawer). Also had to beef up some PCB traces - someone who repaired the amp before used too hot solder gun for sure...

Didn't have time to re-connect the PCB to output stage - tomorrow will do...

Tx for all suggestions!
 

Attachments

  • PL36 after all semi's replaced.JPG
    PL36 after all semi's replaced.JPG
    317.2 KB · Views: 353
Not much progress: after reconnecting the driver board almost all symptoms remained with the key one being:
- when OPA IC's are in the voltage across Zener diodes is less than 2V
- when OPA IC's are out the voltage level goes up correctly.

I only increased PS to 40V measured on the big caps and rail fuses were out. All semi's are new including bias transistors. Several more replacements as stated in earlier posts. What's left are resistors, but those remained look rather good.

Without getting the right level of voltages for OPA IC's cannot move forward. Honestly I am puzzled - a big dark hole...

Every suggestion is appreciated!
 
You've replaced the zener bias resistors with a higher value, right? The opamp's current draw may be too high to bias through 7.5k at 40V depending on what type you're using. Temporarily put back the original value and see if you get more across the zeners. You can go back to the higher value when you get done troubleshooting and know that you can run safely at full voltage.

You may still need to go to 60V rails before it "snaps" on. I would be using the light bulb limiter *without* the varaic to make sure there's enough to get the shunt regulators started. With the rail fuses out, the outputs and drivers are effectively disconnected and the only thing you could really hurt are the predrivers, which are cheap in the grand scheme of things.
 
You've replaced the zener bias resistors with a higher value, right?

Hi Wg_ski,

Rail fuses are out until fixing the PL36 driver board issues. The dropper 7,5k resistors are new and with proper unchanged values. In the meantime replaced C113, C102, C103 and C106 (and shadows for the left channel) in case having a shorted capacitor at nominal voltage. Finally changed IC's and .... bingo! I have realized that actually the IC's were replaced before changing all the semi's, some resistors and caps (just mentioned). So clearly they both were damaged by the origin cause of wrong zener PS voltage levels. I've put new ones and the readings across both zeners at 100VDC measured on main PS caps was +/- 15.07V. Success!

I am ready to proceed with further stage of starting up/ testing the amp.

Tx a lot for your contribution!

/Przem
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.