• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Kuba Serenade Stereo Console - Part II - New development - amp debug question

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

I am starting a new thread on old project as it's been abandoned for couple of months and nobody will follow back. I am sorry if that is not advisable, I'll correct that if needed.

The original thread is here:
http://www.diyaudio.com/forums/showthread.php?t=242064

Sorry dropping the project for a while.
Here is the new development and big questions (in red) :scratch1:

Last couple of days I finally spent some time in restoration and debugging the console.

What I have done – I’ve started rebuilding completely the power supply rail - seems that the power transformer either have been replaced or some gimmicky repair job(s) have been done, extremely dangerous! The transformer have been mounted pressing the 110V cable and the heat “ate” the insulation to almost running bare 110V to transformer’s enclosure (see photo 5).

I’ve rebuild all the wiring, installed new fuse holder (see how was done before on photo 5), replaced both the can caps (50uF/350V) with Samsung 150uF/400V and made new enclosure for standard 3-pin female power connector, grounding the chassis to power ground (see photos 1, 2 and 3).

I also broke-in all the pots and switches with good contact/silicon spray.

I run standard Line-In (CD-player) with nice jazz CD and power it up. The sound appeared loud and clear all the way! AM with internal ferrite antenna works absolutely clear IN THE BASEMENT, so no complains about AM at all.

The bad – I still have FM dead (no hissing noise at all) and I still have one of the channels distinctively quieter than the other one.

I started with troubleshooting the amp as that would be the main use anyways.

Please refer to the “imperial_j611k_sch.pdf “file from the thread (attached again below) from now on – it is NOT the exact schematic, but seems to be close enough to troubleshoot.

What I have checked and done:
- swapping the tubes makes the things even slightly worse (meaning – the quieter channel kept being quieter, but louder channel become more loud, very little, but I can tell the difference – e.g. – tubes are fine.
- if I remove the ground connection from the speakers (pin 1 and 6) and leave speakers in series (fed between pins 3 and 8) – the output it equal (the way it should be).

I’ve checked the WHOLE passive rail with BASS/TREBLE and some funky corrections (everything west of the triode of ECL86 on the schematic – pots, switches and passive components) with signal-generator and scope – everything works fine, all controls and balance work.

I’ve decided to go to the back end – pentodes of ECL86:
- replaced the 470pF caps (C66 and C79) – no difference
- replaced the 10nF caps (C65 and C78, actually – 3.3nF on the board) – no difference
- all surrounding resistors are fine and exact

I’ve checked the DC on the anodes – seems that the problem is somehow there – voltages are as follows:
- main power DC (from rectifier) – 221V
- same DC is on the OTs – pin “rt” (rot = red)
- on the quieter channel – anode of the pentode (pin 6) has 204V on power up and goes slightly down do 201V after warming up.
- on the louder channel – anode of the pentode (pin 6) has 218V on power up and stays 218V, EVEN after warming up.


I may have missed some caps or other components to check on the back-end as the schematic is not exact and the mess on the board is big enough, but I can’t see any reasonable explanation why the DC is different. So far, excluding obviously shot big-can electrolytic caps – every single passive component I’ve checked is dead on the value and I can’t see any other electrolytic cap to change...

I’ve even swapped the output transformers (see photo 5) – the result is the same, but the power voltage rose up to 230V, of course – quieter channel still has lower voltage on the anode, but is 3-5V higher before the swap. Go figure now...

I don’t have extra tubes to play further, but I feel that I reached dead end for now.

Any ideas would be appreciated, guys, thanks in advance!
j2xgwDMBDSsnVA69OtCogRQzID4UOTOFEwAHkGJSmhPzgRIAFU7lagXmBJGuXwEwEMPlz8AaBEagAjDrBQAkcUj5iCtQwK0FSzhciBsQADs=
 

Attachments

  • Receiver - New Power - Top.JPG
    Receiver - New Power - Top.JPG
    428.8 KB · Views: 77
  • Receiver - New Power - Side.JPG
    Receiver - New Power - Side.JPG
    368 KB · Views: 94
  • Receiver - New Power - Side 2.JPG
    Receiver - New Power - Side 2.JPG
    357.4 KB · Views: 81
  • Receiver - OTs Before.JPG
    Receiver - OTs Before.JPG
    499.5 KB · Views: 71
  • Power Components - Before.jpg
    Power Components - Before.jpg
    673.2 KB · Views: 70
  • imperial_j611k_sch.pdf
    581.1 KB · Views: 43
Zibi, the caps are more than fine, both showed 90nF (note the schematic is not exact one). Moreover - Ive checked ALL similar caps in the triode and pentode rails - no difference. I've checked the surrounding caps - all are good.

I am going to replace them all, just to be sure that no leakage occur.

But before doing that - I've made the following observation:
The problem with the amp channel occur about 3-5 minutes AFTER INITIAL WARMUP - when amp tubes are warmed up after 20-30 secs - the sound is CLEAR AND EQUAL on both channels, after another 3-5 minutes - the problem channel start to fade out and stays about 60-70% of the good channel. After cool-down - the procedure repeats always the same way.

What it might be so heat-sensitive - I've already swapped the tubes and output transformers, there is no visual sign of bad/cold solder...

What components might be more sensitive to heat - if I replace all the caps (probably around 15-20 in the amp area) - do you think that resistors may behave this way as well? Or sockets.... I don't know what can be...

Any advice would be appreciated.

Old Hush
 
don't think heat has anything to do with it ...
the ECL tiodes have true grid leakage biasing by means of the very high R 10Meg; that is indeed the primary suspect; any charge slowly building up here - or not building up as intended - will cause a phenomenon as you describe;
if leakage in the coupling C is not the case, it could be the 10Meg either open (check) or leakage across it due to dust or residue from cleaner ...
 
Zibi, the caps are more than fine, both showed 90nF (note the schematic is not exact one). Moreover - Ive checked ALL similar caps in the triode and pentode rails - no difference. I've checked the surrounding caps - all are good.

I am going to replace them all, just to be sure that no leakage occur.

Old Hush

NO, THEY ARE NOT FINE. Those stock caps are notorious for leakage, and if you haven't replaced them, do so now. Checking them on a capacitor checker will often show a high reading. This is characteristic of a leaky cap. ANY and ALL paper capacitors in that set should be changed.
 
Jmsent is right.
old64goat: "One thing you must remember a modern cap tester like what you will find in some digital multi-meters can read a high capacity reading IF the cap is leaky."
Testing Capacitors Part 2 - YouTube

Oh, I didn't know about the leakage in case of higher reading (I have tested them on about 20 years old good digital multimeter). :scratch:

Thanks a lot, Jmsent, Zibi, guys, that may explain the weird behavior...

OK, that's what I thought needed to be done at the end - replacing ALL caps around triodes and pentodes (around 25 pcs - a little tricky as they are mounted one over another in futuristic mesh with resistors, but at least I'll have nothing more to blame... :rolleyes:

I have to go shopping for caps then, but I am a little rusty now a days, haven't shop for audiophile caps for over a decade...so I have few questions:
1. Any particular brand I have to aim for (better spend a little more than do that again)?
2. What type should I use - is modern ceramic good enough?
3. What voltage - on original caps it's written: "1000V DC, 400V AC, 250C AC (b)" - what the (b) stands for and will 400V rating suffice (highest voltage around the console is 235V)?
4. Any link for, let's say, typical 22nF from big vendors like DigiKey will be appreciated!


Thank you very much for the advises!

I'll need some time to get the caps and to replace them so, probably I'll have new update around Christmas time, when the snow will be deep and soldering appears very appealing...:D

Cheers!:drink:
 
Steer clear of ceramic caps., if at all possible. 716P series Orange Drop film parts are a reasonable choice. Another reasonable option is Soviet surplus paper in oil (PIO) parts. You are probably safe in using 400 WVDC rated caps. in the signal path. Parts rated for 600 or 630 WVDC are surely safe.

I have a sneaking suspicion that the 1000 WVDC part(s) you asked about are being used as "death caps.", attached to the AC mains. Get that sort of stuff out of there! Make certain to install a proper, 3 wire, safety grounded, power cord.
 
Last edited:
Eli, thanks for the tip. Whoa, they're expensive - it'll cost me over $50 to replace the whole rail... will keep searching alternatives.
And... no, it is NOT death cap, I'll wrote that I already replaced the whole power rail - all wires, fuses, can caps, installed 3-wire standard 15A female PC-type socket. No death-caps, I've learned my lesson from you, following your advise, when started with another console!
Thanks!
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.