Hi everybody, this is my first post. I originally joined diyAudio for DIY speaker tips but now I really need some help from you Tube guys. I have an integrated using four EL34/6CA7-EH SOVTEK tubes. Today, when I wasnt using it but was powered on, one of the tubes went mad and a REALLY loud high frequency sound came from the right speaker. Then came the smoke, and the stench....I shall never forget it... The sound now wont go away. Something a bit strange though was that looking into the tube (like one does) on occasions before it blew, it looked different to the other three, it glowed brighter and somehow looked like it was wired back to front inside?! Does anyone have any idea what went on? And whether I've damaged the amplifier and/or speakers? Or would just replacing a duff tube do the trick?
The image shows the only visible break (though I assume the glass went if smoke could get out...)
Thanks, Stu.
An externally hosted image should be here but it was not working when we last tested it.
The image shows the only visible break (though I assume the glass went if smoke could get out...)
Thanks, Stu.
Your description sounds like an output tube in full thermal runaway.
You can't simply replace the tube, there is obviously a problem in the amplifier that needs to be fixed. Also you should not run it this way, your output transformer if not already fried could be the next item on the list to fail.
The smoke you saw came from burning resistors inside your amplifier, most likely the screen resistor and possibly a cathode resistor as well depending on the topology.
The envelope of the tube itself may have a scorched logo, but unless the getter (silvery stuff) on the inside of the tube has turned white you did not let the smoke out of the tube itself although it is definitely no good. Sometimes when an internal connection is vaporized by a short the sudden presence of a large internal pressure (shock wave?) may fracture the envelope and let in the air.. (And out the smoke as well
)
I would guess the tube developed an internal short possibly due to excessive dissipation or a manufacturing defect.
You should expect to replace the pair on the affected channel.
You will need to find someone to fix your amplifier, alternatively if you are willing to learn to do it yourself there are people here who can help you. Note that before proceeding further you should read the thread on high voltage safety as well as the newbie thread. I assume you can solder competently, if not find a technician to do the repairs...
The first step is to tell us what kind of amplifier you are using and to post the schematic if you have it.
You can't simply replace the tube, there is obviously a problem in the amplifier that needs to be fixed. Also you should not run it this way, your output transformer if not already fried could be the next item on the list to fail.
The smoke you saw came from burning resistors inside your amplifier, most likely the screen resistor and possibly a cathode resistor as well depending on the topology.
The envelope of the tube itself may have a scorched logo, but unless the getter (silvery stuff) on the inside of the tube has turned white you did not let the smoke out of the tube itself although it is definitely no good. Sometimes when an internal connection is vaporized by a short the sudden presence of a large internal pressure (shock wave?) may fracture the envelope and let in the air.. (And out the smoke as well
)I would guess the tube developed an internal short possibly due to excessive dissipation or a manufacturing defect.
You should expect to replace the pair on the affected channel.
You will need to find someone to fix your amplifier, alternatively if you are willing to learn to do it yourself there are people here who can help you. Note that before proceeding further you should read the thread on high voltage safety as well as the newbie thread. I assume you can solder competently, if not find a technician to do the repairs...
The first step is to tell us what kind of amplifier you are using and to post the schematic if you have it.
Fuses for tube cathodes
I have 5 EL34/6CA7 Amps and have auditioned heaps of modern production plus NOS tubes. I have had both NOS and modern production tubes fail for no apparent reason with symptoms of either screen to cathode short or (less likely) anode to cathode short.
All my amps now have 10 Ohm 0.25W resistors in the cathode circuit. Yeh - its a great place to measure tube currents for balancing BUT mostly they are great individual tube fuses. Cooked a few 10R resistors since but never an output transformer.
Next time you see a push pull EL34 schematic with 10 Ohm 5 Watt (or 2W) resistors specified in the cathode circuit - just ignore it and use 0.5W or less. Of course this does NOT apply to 270 Ohm cathode BIAS resistors - they need the full power rating BUT put the 1/2 or 1/4 watt 10 R in series with them.
Cheers,
Ian
I have 5 EL34/6CA7 Amps and have auditioned heaps of modern production plus NOS tubes. I have had both NOS and modern production tubes fail for no apparent reason with symptoms of either screen to cathode short or (less likely) anode to cathode short.
All my amps now have 10 Ohm 0.25W resistors in the cathode circuit. Yeh - its a great place to measure tube currents for balancing BUT mostly they are great individual tube fuses. Cooked a few 10R resistors since but never an output transformer.
Next time you see a push pull EL34 schematic with 10 Ohm 5 Watt (or 2W) resistors specified in the cathode circuit - just ignore it and use 0.5W or less. Of course this does NOT apply to 270 Ohm cathode BIAS resistors - they need the full power rating BUT put the 1/2 or 1/4 watt 10 R in series with them.
Cheers,
Ian
First off, thankyou for replying to this post, it is very much appreciated. The amplifier is a Classic (Yushang Audio) No. 16 Integrated amplifier. I have emailed both Classic, and the distributor with no response.
Here are some pics:
http://www.fossilfarm.co.uk/stuebay/ouch0.jpg
http://www.fossilfarm.co.uk/stuebay/ouch1.jpg
http://www.fossilfarm.co.uk/stuebay/ouch2.jpg
http://www.fossilfarm.co.uk/stuebay/ouch3.jpg
http://www.fossilfarm.co.uk/stuebay/ouch4.jpg
http://www.fossilfarm.co.uk/stuebay/ouch5.jpg
Kevin you certainly know your stuff. It looks like that resistor has gone. Now, would replacing the right channel tubes and the resistor fix do you think, or does the problem lie deeper? Although it is annoying that I can't play music at the moment, I am keen to learn a bit...
Here are some pics:
http://www.fossilfarm.co.uk/stuebay/ouch0.jpg
http://www.fossilfarm.co.uk/stuebay/ouch1.jpg
http://www.fossilfarm.co.uk/stuebay/ouch2.jpg
http://www.fossilfarm.co.uk/stuebay/ouch3.jpg
http://www.fossilfarm.co.uk/stuebay/ouch4.jpg
http://www.fossilfarm.co.uk/stuebay/ouch5.jpg
Kevin you certainly know your stuff. It looks like that resistor has gone. Now, would replacing the right channel tubes and the resistor fix do you think, or does the problem lie deeper? Although it is annoying that I can't play music at the moment, I am keen to learn a bit...
Hi Stu,
Found this...which I assume youalready know, but might give a few more clues as to type of circuit...
http://www.yushang-audio.com/english/product/yushang-p160.htm
I would suggest replacing the tubes, replace all burnt components, check the output transformer primaries (probably best to compare DC resistance with the undamaged channels transformer, though the laminations may need degaussing if they were subjected to a sustained current imbalance. Not the end of the world if they do need replacing) and just ease up on the bias a little until the amp proves itself reliable. I think I would suspect a dodgy tube more than anything, but just keep a good eye on the bias currents and make sure none of the tubes are doing anything untoward.
Unfortunately I live all the way over in Hereford so it's a bit far to road trip over there.
Hope this helps,
Steve
Found this...which I assume youalready know, but might give a few more clues as to type of circuit...
http://www.yushang-audio.com/english/product/yushang-p160.htm
I would suggest replacing the tubes, replace all burnt components, check the output transformer primaries (probably best to compare DC resistance with the undamaged channels transformer, though the laminations may need degaussing if they were subjected to a sustained current imbalance. Not the end of the world if they do need replacing) and just ease up on the bias a little until the amp proves itself reliable. I think I would suspect a dodgy tube more than anything, but just keep a good eye on the bias currents and make sure none of the tubes are doing anything untoward.
Unfortunately I live all the way over in Hereford so it's a bit far to road trip over there.
Hope this helps,
Steve
Thanks Steve, I have ordered a pair of sovteks and several new resistors. Your post has reassured me a lot! I was not looking forward to a trip over China...
Another thing, that particular tube that died always made more mechanical noise than the others, and does the fact that it looked "back to front" mean anything? Or am I being a complete newbie....
To lndm, for now I choose to believe it was a faulty tube!
Thanks again everyone.
Stu.
Another thing, that particular tube that died always made more mechanical noise than the others, and does the fact that it looked "back to front" mean anything? Or am I being a complete newbie....
To lndm, for now I choose to believe it was a faulty tube!
Thanks again everyone.
Stu.
If it was brighter too, maybe the heater was shorter and passing more current? My KT88's jangle around a lot at startup with the inrush current into cold heaters. They're quiet as a mouse once they've warmed up and heater current has calmed down though. I had a little 6n1p that was like that, DC resistance of the heater was half that of normal ones, the heater was brighter, and with AC heaters used to hum like a *****, so I never used it again. Given time I'm sure it would have given up the ghost.
It could potentially be oscillations, but it looks like a very commercial product and I would hope they would iron all those problems out before release onto the market, but who knows...
Steve
It could potentially be oscillations, but it looks like a very commercial product and I would hope they would iron all those problems out before release onto the market, but who knows...
Steve
This document starts off on discussing catastophic, degenerative and subjective failures in tubes... maybe you find something of value...
http://z05.zupload.com/download.php?file=getfile&filepath=5441
http://z05.zupload.com/download.php?file=getfile&filepath=5441
Hi Stueyp,
Excellent pictures.. I recommend very carefully cleaning the board with rubbing alcohol (90% if you can get it) your local apothecary should have it.. Then once you have done that look at the traces connecting to the burned components, they hopefully won't be damaged. The carbon residue on the board is probably somewhat conductive and needs to removed, usually alcohol and a stiff acid brush will do the job nicely. Blot the area with paper towel and repeat until it is as clean as you can get it. Hopefully you won't have to remove the board to clean the other side.
Check all of the resistors connected to that socket (replaced or not) and the ones connected to the other valve and make sure that they measure to within a couple of % of the same location on the other channel.
Be sure to check the bias pots! (4 pin foot print with pot on other side of board.) Also it looks like there was (hehe) a capacitor on the screen grid of that tube that exploded? Can you check the voltage on the remaining caps, this should be marked somewhere on the capacitor. Are those relays used for switching from triode to UL mode, or do you know?
It might not be a bad idea to install fuses in series with the cathodes of the EL34.
The Siemens EL34 are wonderful, but IMHO they will cost more than what you paid for this amplifier. I am not incidentally a big fan of most Russian made EL34, and would recommend a quartet of new JJ instead. Their QC seems more consistent, although people do rave about the sound of the EH EL34. Make sure any you buy in the future are matched and burned in for at least 24 hrs - if you can't get them that way contact me as I can...
Need to think about how to power this amp up safely once you have made repairs. I recommend a variac to do this, or alternately just a ballast lamp in series with the power cord to the amplifier could be used to limit current in the event of a fault. (3 x 100W bulb in parallel, all of which should be connected in series with the amplifier power cord.)
Excellent pictures..
I recommend very carefully cleaning the board with rubbing alcohol (90% if you can get it) your local apothecary should have it.. Then once you have done that look at the traces connecting to the burned components, they hopefully won't be damaged. The carbon residue on the board is probably somewhat conductive and needs to removed, usually alcohol and a stiff acid brush will do the job nicely. Blot the area with paper towel and repeat until it is as clean as you can get it. Hopefully you won't have to remove the board to clean the other side.Check all of the resistors connected to that socket (replaced or not) and the ones connected to the other valve and make sure that they measure to within a couple of % of the same location on the other channel.
Be sure to check the bias pots! (4 pin foot print with pot on other side of board.) Also it looks like there was (hehe) a capacitor on the screen grid of that tube that exploded? Can you check the voltage on the remaining caps, this should be marked somewhere on the capacitor. Are those relays used for switching from triode to UL mode, or do you know?
It might not be a bad idea to install fuses in series with the cathodes of the EL34.
The Siemens EL34 are wonderful, but IMHO they will cost more than what you paid for this amplifier. I am not incidentally a big fan of most Russian made EL34, and would recommend a quartet of new JJ instead. Their QC seems more consistent, although people do rave about the sound of the EH EL34. Make sure any you buy in the future are matched and burned in for at least 24 hrs - if you can't get them that way contact me as I can...
Need to think about how to power this amp up safely once you have made repairs. I recommend a variac to do this, or alternately just a ballast lamp in series with the power cord to the amplifier could be used to limit current in the event of a fault. (3 x 100W bulb in parallel, all of which should be connected in series with the amplifier power cord.)
Well, just to throw a different view in.
I am using 6CA7EH's in my own amp, better than the EL34's I tried. I have used 6CA7EH's in Mesa Baron amps that eat EL34's (pick your brand), if you know that amp you also know what I mean. It should be stuffed with 6L6GC's or better.
So I am very happy with Electroharmonix as some of you know. You may have had a defective tube, can happen. In general I find the fat bottle 6CA7/EL34 more robust. These are beam tetrodes as opposed to a true EL34 which is a pentode.
Be aware too that a defective output transformer can ruin the tube, so can an open in the load (not in your case as you heard it oscillating). Also check the original solder joints to the plate and screen leads, and grid for defects. It would be nice to know the why of this.
-Chris
I am using 6CA7EH's in my own amp, better than the EL34's I tried. I have used 6CA7EH's in Mesa Baron amps that eat EL34's (pick your brand), if you know that amp you also know what I mean. It should be stuffed with 6L6GC's or better.
So I am very happy with Electroharmonix as some of you know. You may have had a defective tube, can happen. In general I find the fat bottle 6CA7/EL34 more robust. These are beam tetrodes as opposed to a true EL34 which is a pentode.
Be aware too that a defective output transformer can ruin the tube, so can an open in the load (not in your case as you heard it oscillating). Also check the original solder joints to the plate and screen leads, and grid for defects. It would be nice to know the why of this.
-Chris
Hi Stu,
Those relays switch G2 between the transformer and a resistor to plate (triode to pentode mode). One wonders .... many places where you could have a problem here.
Can you clean up the damaged area and wash it with alcohol, then take another picture?
Also, try to get a schematic so you can service the amp in the future.
-Chris
Those relays switch G2 between the transformer and a resistor to plate (triode to pentode mode). One wonders .... many places where you could have a problem here.
Can you clean up the damaged area and wash it with alcohol, then take another picture?
Also, try to get a schematic so you can service the amp in the future.
-Chris
Hi guys,
I will check the solder points, the ouput from the transformer, and clean up the bombsite. I'll take another picture of the relay. Installing fuses sounds like a good idea, but what pins should I put them between, and what sort of fuse? The reason I ask is I have a surplus of 3.15A glass fast-blow types from an Audio Analogue Puccini. (Lf.F3.15AL250V)
By the way, the amp as you can see in the pictures has two sets of speaker posts, one for 4 Ohm and one for 8 Ohm. If I have a pair of speakers running at a nominal 6 Ohms, which post do I put them in??
Thanks again,
Stu.
I will check the solder points, the ouput from the transformer, and clean up the bombsite. I'll take another picture of the relay. Installing fuses sounds like a good idea, but what pins should I put them between, and what sort of fuse? The reason I ask is I have a surplus of 3.15A glass fast-blow types from an Audio Analogue Puccini. (Lf.F3.15AL250V)
By the way, the amp as you can see in the pictures has two sets of speaker posts, one for 4 Ohm and one for 8 Ohm. If I have a pair of speakers running at a nominal 6 Ohms, which post do I put them in??
Thanks again,
Stu.
Lordy, lordy, what a bonfire! The last time I had a bonfire like that was inside a Tektronix curve tracer. I would echo all of kevinkr's advice and add that fibreglass brushes are good for scrubbing burnt printed circuit board away. You really do need to get rid of all the damaged parts and to wash the affected area with isopropyl alcohol (sometimes available from chemists but always available from Farnell or RS).
After that, a circuit diagram is essential. It is not enough to replace the damaged parts - you need to know what caused the problem. I would suspect a failure of negative grid bias on the destroyed valve.
After that, a circuit diagram is essential. It is not enough to replace the damaged parts - you need to know what caused the problem. I would suspect a failure of negative grid bias on the destroyed valve.
amperex said:
Ahh..ah and also the price ? Nup.
I can't say much for siemens nos EL34's, the ones I got (fawn coloured box with blue end) tend to have stars glowing inside on the grids. This implies non cleansiness in assembly.
Re burnt out amp..I wasn't able to download all pics.../details.....but it sounds if only a solid state power supply could do that kind of damage, if a GZ or 5Uxx was doing the rectification, they would immediately glow red hot on a short circuit and the drama would stop there. From the details I gleaned it mentions a ring transformer/or toroid...capable of high current at low impedance.
You mention triode cum pentode c/o switch. A defective switch may not show up on a meter, because internal arcing could be via switch body to chassis and would certainly re-track/arc internally the next time B+ appears and the problem is ready to destroy the replacement tube.. Switch replacement is the only option or hard wire the problem out. The other point is if the anode was shorted to chassis via int arcing from a duff switch, the screen (g2) may have it's supply derived from another low impedance source and the current immediately destroy the tube. One needs the circuit diagram.When checking the output transformer resistance with some digital multimeters, the ohms range may not settle due to the slow rise time v.s inductance....simply shortout a secondary winding. If the tranny is duff this problem is automatically solved.
Again, with Sstate psu's, this emphasises the correct fuses.
richj
Stueyp,
Oh man, my heart burns for you! As a man of many years and several such experiences in my youth, I know how you must feel. (Assure you it is worse if one discovers afterwards that all the devastation was due to one's own stupidity....)
Just to add a bit about the output transformer. I did not download the circuit but presume that the connections are known. One way to test is to remove connections (removing power tubes would in most cases be sufficient), and feed a small voltage into the secondary and measure on the primaries. If you do not have a signal generator the mains will do - I mean suitably reduced by a transformer, of course!!
The ratio of secondary/primary would be very roughly of the order of 20 - 30:1, so a few volt should do. Avoid shocking yourself! You should notice if there is a short somewhere in windings, by not measuring equal voltages where you should, e.g. the anodes, G2 taps, etc. This will be more revealing than a d.c. ohms test - the latter will, of course, show any open circuit. One problem might remain undetected, i.e. a winding short only on high (signal) voltages.
Regards, and success!
But I also note with concern the comments on the different tube makes. I have just purchased 10 x 6L6EH (supposed to be 6L6GC), and carefully read the different experiences. They will work under 460V G2; 550V Va (fixed bias). I hope I will not need to open a similar post some time from now! It might be a little off- thread, but can someone kindly comment on any experience with JJ tubes, please?
Thanks.
Oh man, my heart burns for you! As a man of many years and several such experiences in my youth, I know how you must feel. (Assure you it is worse if one discovers afterwards that all the devastation was due to one's own stupidity....)
Just to add a bit about the output transformer. I did not download the circuit but presume that the connections are known. One way to test is to remove connections (removing power tubes would in most cases be sufficient), and feed a small voltage into the secondary and measure on the primaries. If you do not have a signal generator the mains will do - I mean suitably reduced by a transformer, of course!!
The ratio of secondary/primary would be very roughly of the order of 20 - 30:1, so a few volt should do. Avoid shocking yourself! You should notice if there is a short somewhere in windings, by not measuring equal voltages where you should, e.g. the anodes, G2 taps, etc. This will be more revealing than a d.c. ohms test - the latter will, of course, show any open circuit. One problem might remain undetected, i.e. a winding short only on high (signal) voltages.
Regards, and success!
But I also note with concern the comments on the different tube makes. I have just purchased 10 x 6L6EH (supposed to be 6L6GC), and carefully read the different experiences. They will work under 460V G2; 550V Va (fixed bias). I hope I will not need to open a similar post some time from now! It might be a little off- thread, but can someone kindly comment on any experience with JJ tubes, please?
Thanks.
Hi Johan,
I am told the plates are a little heavier than spec for a 6L6GC. They do look in between a 6L6GC and 7581A.
Where did you get them and did you buy any matched? As for me, the hardest service I have seen them in are Fender Twins. They seem to be happy and no complaints. Even the guitar guys like the tone.
I have used quite a few, so let me know how they work out for you. I've heard the JJ's are very good too. Since I am happy so far I have not bothered to try the JJ's.
-Chris
I am told the plates are a little heavier than spec for a 6L6GC. They do look in between a 6L6GC and 7581A.
Where did you get them and did you buy any matched? As for me, the hardest service I have seen them in are Fender Twins. They seem to be happy and no complaints. Even the guitar guys like the tone.
I have used quite a few, so let me know how they work out for you. I've heard the JJ's are very good too. Since I am happy so far I have not bothered to try the JJ's.
-Chris
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.