A friend is having me replace all the tubes (both pre- and power-amp) in his Fender Twin Reverb. The original power-amp tubes are all miss-matched (two Fenders, a GE, and a Sylvania), and the pre-amp tubes are all original. I began by switching out all the tubes, and prepared to bias it. But the power-amp tubes are running very hot (the plates glow), and there is a horrendous hum and crackle sound when you put it out of stand-by. All the new power-amp tubes are matched Groove Tubes GT6L6GC #5’s. I put the original tubes back in and noticed the GE tube actually had the plate burned through! However it seemed to operate with the original tubes installed, and the hum was lessened (although not gone) and the crackling gone. So what have we got here?
I do not have a lot of experience with tube equipment, so I do not even know where to begin.
Thanks,
Mike
I do not have a lot of experience with tube equipment, so I do not even know where to begin.
Thanks,
Mike
This is going to take some probing around in the circuitry with a scope and meter. If you're inexperienced with tube circuits, don't even THINK about doing this- the voltages used are lethal, and electrolyic caps can explode. It's time to bring in a technician; you've done what you can do.
Primae non nocere, or however you spell that.
Primae non nocere, or however you spell that.
I had a similar case with an old Altec tube amp I was restoring. First I put in new tubes and noticed that they were running hot and there was a lot of noise. In my case (your case might be different) the problem was in the capacitors. They had deteriorated. When you used old tubes, because they had lost a lot of their power, they didn't draw very much current from the caps and it seemed to work. But with new tubes the problem became apparent and I had to change the caps to make it work.
Another note. When you change the caps, now you are stressing more the resistors and all parts of the circuit as well. So more problems might pop out.
Please be careful. Even after disconnecting the power cord, the capacitors can carry enough power to seriously injure you. Good luck.
Another note. When you change the caps, now you are stressing more the resistors and all parts of the circuit as well. So more problems might pop out.
Please be careful. Even after disconnecting the power cord, the capacitors can carry enough power to seriously injure you. Good luck.
Gah, be damn careful, there might be upwards of 400VDC inside that amp. 
That being said- after unplugging the unit, discharging all the caps with a 50 ohm resistor, and removing the tubes:
1) replace any old selenium diode rectifiers with a silicon diode 1n4007 etc
2) replace any burst or leaking electrolytic capacitors (replace all low voltage ones, high voltage ones if necessary)
3) replace all coupling capacitors (they usually range from 0.01- 0.47uF), and are commonly wax paper
Try these things if you are really keen. Try to find out the model number and download the schematic for this amp.
That being said- after unplugging the unit, discharging all the caps with a 50 ohm resistor, and removing the tubes:
1) replace any old selenium diode rectifiers with a silicon diode 1n4007 etc
2) replace any burst or leaking electrolytic capacitors (replace all low voltage ones, high voltage ones if necessary)
3) replace all coupling capacitors (they usually range from 0.01- 0.47uF), and are commonly wax paper
Try these things if you are really keen. Try to find out the model number and download the schematic for this amp.
I'm also having a problem with a similar amp, and I'm adding a pointer to another thread where I posted about it: http://www.diyaudio.com/forums/showthread.php?postid=1127133#post1127133
The answer he gave to the other person was that the grids were melted. That obviously cannot be the case here since the problem remained after I replaced ALL the tubes.
I'm not sure how much negative the bias should be. The schematic below shows it should be set to -54.6 V. Also, there's only one adjustment, not separate for both tubes.
http://www.fender.com/support/amp_schematics/pdfs/Hot_Rod_DeVille_Schematic.pdf
I'm not sure how much negative the bias should be. The schematic below shows it should be set to -54.6 V. Also, there's only one adjustment, not separate for both tubes.
http://www.fender.com/support/amp_schematics/pdfs/Hot_Rod_DeVille_Schematic.pdf
The Fender Twin Reverb restoration "bible".
1) Replace the 470 Ohm 2W Carbon Film screen resistors with 1K 5W wire wound (If you check the resistance of the old 470R ones you will measure anywhere between 180R and 850R on intact ones and you will often find ones that fall in half when you unsolder them - that is even though a quick inspection seemed to indicate that they were OK they were actually stuffed (as we say in Tech Talk). If a screen resistor to any output tube is "stuffed" that tube will conduct practically no current and you will get massive hum from inbalance currents in the output tranny.
2) During the step above, when you have the old screen resistors off and before putting in the new ones you have read acess to the 1K5 Grid stoppers, Replace those as well with 1W metal films.
3) I usually replace the entire bias supply chain at this time - all series resistors, the filter cap and thed diode - 70% of all Fender failures I've seen can be traced back to Bias Supply Failures.
OPTIONAL:
While you have clear acess to the output tube sockets consider (seriously) removing the wire link from the cathode pin to chassis on each socket and fitting 10 Ohm 1/2 watt or even 1/4 watt resistors in place of the wire link. This does 2 things: It gives you a place to measure the bias current for each tube. The resistors act as fuses in the event of a bias failure or a tube short which can save your output tranny from damage. This also allows you to actually change some resistors in the bias chain if you are not happy with the idle currents although that generally won't be required (cause twin Reverbs dont have a bias level pot only a bias balance pot).
Once that is done:
With the output tubes NOT fitted and the amp still in standby use your multimeter to check the bias voltage on the Grid 1 pin to cathode pin off EACH output tube socket. NEVER refit the output tubes and switch out of standby until you are 100% happy with this.
The required voltage can be read of the schematics which you can find at Schematic Heaven.
When you are happy with the above fit the output tubes and get the amp up to operating (standby off) and then you can set the balance pot purely by ear - stick your head against the speaker and adjust the balance pot for minimum hum - that meams balance idle currents in the output tranny.
Dashed this off in a hurry but I hope there is something useful there.
Cheers,
Ian
1) Replace the 470 Ohm 2W Carbon Film screen resistors with 1K 5W wire wound (If you check the resistance of the old 470R ones you will measure anywhere between 180R and 850R on intact ones and you will often find ones that fall in half when you unsolder them - that is even though a quick inspection seemed to indicate that they were OK they were actually stuffed (as we say in Tech Talk). If a screen resistor to any output tube is "stuffed" that tube will conduct practically no current and you will get massive hum from inbalance currents in the output tranny.
2) During the step above, when you have the old screen resistors off and before putting in the new ones you have read acess to the 1K5 Grid stoppers, Replace those as well with 1W metal films.
3) I usually replace the entire bias supply chain at this time - all series resistors, the filter cap and thed diode - 70% of all Fender failures I've seen can be traced back to Bias Supply Failures.
OPTIONAL:
While you have clear acess to the output tube sockets consider (seriously) removing the wire link from the cathode pin to chassis on each socket and fitting 10 Ohm 1/2 watt or even 1/4 watt resistors in place of the wire link. This does 2 things: It gives you a place to measure the bias current for each tube. The resistors act as fuses in the event of a bias failure or a tube short which can save your output tranny from damage. This also allows you to actually change some resistors in the bias chain if you are not happy with the idle currents although that generally won't be required (cause twin Reverbs dont have a bias level pot only a bias balance pot).
Once that is done:
With the output tubes NOT fitted and the amp still in standby use your multimeter to check the bias voltage on the Grid 1 pin to cathode pin off EACH output tube socket. NEVER refit the output tubes and switch out of standby until you are 100% happy with this.
The required voltage can be read of the schematics which you can find at Schematic Heaven.
When you are happy with the above fit the output tubes and get the amp up to operating (standby off) and then you can set the balance pot purely by ear - stick your head against the speaker and adjust the balance pot for minimum hum - that meams balance idle currents in the output tranny.
Dashed this off in a hurry but I hope there is something useful there.
Cheers,
Ian
Nixie,
Pretty much applies to any Tube Guitar Amp. I've certainly have done the much the same thing with a Super Twin (Six 6L6s to give 180W Out), an old Fender Bandmaster (which I also converted BACK to tube rectifier) and a drop dead gorgeous DR412 Hiwatt I have restored recently.
That is just a good logical SAFE (keep the smoke inside the wires)sequence to use, that is:
Check / change the output screen resistors
Rebuild the bias supply
Fit bias current monitoring resistors in the cathodes
Check bias at the tube sockets before fitting output tubes (Note that bias can be checked with the amp in standby)
Fit output tubes, check adjust bias and bias balance
I used to use much the same sequence when fixing Surgical Diathermy units 30 years ago where the output tubes ran at 2000 plus volts and the HV supply was capable of 800mA - that capable of blowing me to Kingdom Come (they put out 700W at about 2MHz).
By the way - when someone gives me a vintage amp to restore and says results are more important than cost, which is most of the time as I'm selective about what jobs I take on, I use Winged C Svetlana output tubes. For the quick and dirty repair they get whatever I have on hand, Sovteks, JJs or whatever.
On the DeVille mentioned on your thread I would be replacing those phase inverter anode resistors as the next "experiment" in curing the crackle and pop. You may also need to try a new PI tube.
Cheers,
Ian
Pretty much applies to any Tube Guitar Amp. I've certainly have done the much the same thing with a Super Twin (Six 6L6s to give 180W Out), an old Fender Bandmaster (which I also converted BACK to tube rectifier) and a drop dead gorgeous DR412 Hiwatt I have restored recently.
That is just a good logical SAFE (keep the smoke inside the wires)sequence to use, that is:
Check / change the output screen resistors
Rebuild the bias supply
Fit bias current monitoring resistors in the cathodes
Check bias at the tube sockets before fitting output tubes (Note that bias can be checked with the amp in standby)
Fit output tubes, check adjust bias and bias balance
I used to use much the same sequence when fixing Surgical Diathermy units 30 years ago where the output tubes ran at 2000 plus volts and the HV supply was capable of 800mA - that capable of blowing me to Kingdom Come (they put out 700W at about 2MHz).
By the way - when someone gives me a vintage amp to restore and says results are more important than cost, which is most of the time as I'm selective about what jobs I take on, I use Winged C Svetlana output tubes. For the quick and dirty repair they get whatever I have on hand, Sovteks, JJs or whatever.
On the DeVille mentioned on your thread I would be replacing those phase inverter anode resistors as the next "experiment" in curing the crackle and pop. You may also need to try a new PI tube.
Cheers,
Ian
Nixie,
Went looking (quickly) for a DeVille Schematic but didn't find one.
If there is no balance pot then you need to use a matched pair of output tubes.
Adding a balance control is tricky and if you intended to mod the amp, then separate bias controls for each output tube is the way to go BUT we should be able to get the thing going without having to mod it.
Does it have bias monitoring points? Without those you are always guessing to some degree about the actual tube idle currents and the balance between them. If it has an overall bias adjustment then I would think it should have resistors in the cathodes so that you can measure the voltage across them and hence calculate the current (from knowing the resistor value). Once you know those idle currents you can multiple the idle current by the anode voltage to work out the power dissipated in each tube and check if the bias needs to be turned up or down. I generally run 6L6GCs at 25 Watts dissipation (max rating is 30 Watts).
Are the output tubes on a PCB mount socket or point to point wired. The 6L6 cathode is pin 8. Check for a resistor between pin 8 and 0V (or is just a short circuit?). If there is a resistor, then that will allow you to monitor the idle current by measuring the voltage drop across that resistor. The grid (G1) is pin 5. Before fitting new output tubes (and with the amp in standby) measure between pin 5 and pin 8 of the socket and set bias to maximum negative voltage you can dial up on the bias pot. Then fit the tubes and leave to warm up. After 10 minutes of operation adjust bias to required current and maybe trim it again after 1 hr or so.
Hope I'm not "teachin' granny to suck eggs" by telling stuff you already know.
Cheers,
Ian
Went looking (quickly) for a DeVille Schematic but didn't find one.
If there is no balance pot then you need to use a matched pair of output tubes.
Adding a balance control is tricky and if you intended to mod the amp, then separate bias controls for each output tube is the way to go BUT we should be able to get the thing going without having to mod it.
Does it have bias monitoring points? Without those you are always guessing to some degree about the actual tube idle currents and the balance between them. If it has an overall bias adjustment then I would think it should have resistors in the cathodes so that you can measure the voltage across them and hence calculate the current (from knowing the resistor value). Once you know those idle currents you can multiple the idle current by the anode voltage to work out the power dissipated in each tube and check if the bias needs to be turned up or down. I generally run 6L6GCs at 25 Watts dissipation (max rating is 30 Watts).
Are the output tubes on a PCB mount socket or point to point wired. The 6L6 cathode is pin 8. Check for a resistor between pin 8 and 0V (or is just a short circuit?). If there is a resistor, then that will allow you to monitor the idle current by measuring the voltage drop across that resistor. The grid (G1) is pin 5. Before fitting new output tubes (and with the amp in standby) measure between pin 5 and pin 8 of the socket and set bias to maximum negative voltage you can dial up on the bias pot. Then fit the tubes and leave to warm up. After 10 minutes of operation adjust bias to required current and maybe trim it again after 1 hr or so.
Hope I'm not "teachin' granny to suck eggs" by telling stuff you already know.
Cheers,
Ian
The schematic is on the fender.com site and I posted a link in post #7 on this thread: http://www.fender.com/support/amp_schematics/pdfs/Hot_Rod_DeVille_Schematic.pdf
Both tubes share the same cathode resistor (bypassed by a diode).
Both tubes share the same cathode resistor (bypassed by a diode).
Tried to respond last night (Oz Time) butDiyAudio was experiencing probllems.
I looked at the bias instructions on the schematic and they just do not make sense. Instruction say to adjust R82 to get a reading of 60mV at Test Point TP30. R66 is 1 Ohm so 60mV means 60mA or 30mA each tube. HT Volts at anode (B+) is 485V.
So each output tube would be dissipating 485 x 0.03 = 14.55 Watts - thats obviously way wrong.
Check carefully that R66 is indeed 1 Ohm. If so:
I think they must have intended to say that each output tube should idle at 60mA (Power Dissipated is then 29 Watts for 30watt rated tube). That would mean adjusting R82 to give 120mV at TP30.
If the amp were mine I would be setting about 55mA per tube (27Watts) or 110mV at TP30. That will make the tubes last a bit longer and should not impact the sound.
Cheers,
Ian
I looked at the bias instructions on the schematic and they just do not make sense. Instruction say to adjust R82 to get a reading of 60mV at Test Point TP30. R66 is 1 Ohm so 60mV means 60mA or 30mA each tube. HT Volts at anode (B+) is 485V.
So each output tube would be dissipating 485 x 0.03 = 14.55 Watts - thats obviously way wrong.
Check carefully that R66 is indeed 1 Ohm. If so:
I think they must have intended to say that each output tube should idle at 60mA (Power Dissipated is then 29 Watts for 30watt rated tube). That would mean adjusting R82 to give 120mV at TP30.
If the amp were mine I would be setting about 55mA per tube (27Watts) or 110mV at TP30. That will make the tubes last a bit longer and should not impact the sound.
Cheers,
Ian
Bias adjust for each 6L6GC tube and even gain on 12AT7
I just finished modifying my Fender Twin 135RMS and it sounds better than the day I bought it in 1976. All work
or service on any Tube Amp should be done by qualified
Service Techs because of the deadly voltages. With that said if you replace all the caps and add the 10 Ohm Cathode bias resistors to each 6L6GC with a test point to access this bias point with bias adjust for each 6L6GC you will be able to run and adjust your amp to have crunching sounding chords and warm singing leads. You will also need to remove your reverb and vibrato jacks and replace with Bias 10 turn pots so you don't have to pull the chasis out every time to adjust bias on all 4 tubes.
The Lenard Audio - Education - Guitar Valve Amps.
has excellent info which I used
I added a 10 turn pot to the inverter 12AT7 and also added 2 test jacks on the back so I could balance out the 12AT7.
I just finished modifying my Fender Twin 135RMS and it sounds better than the day I bought it in 1976. All work
or service on any Tube Amp should be done by qualified
Service Techs because of the deadly voltages. With that said if you replace all the caps and add the 10 Ohm Cathode bias resistors to each 6L6GC with a test point to access this bias point with bias adjust for each 6L6GC you will be able to run and adjust your amp to have crunching sounding chords and warm singing leads. You will also need to remove your reverb and vibrato jacks and replace with Bias 10 turn pots so you don't have to pull the chasis out every time to adjust bias on all 4 tubes.
The Lenard Audio - Education - Guitar Valve Amps.
has excellent info which I used
I added a 10 turn pot to the inverter 12AT7 and also added 2 test jacks on the back so I could balance out the 12AT7.
30ma or 14.55 Watts per tube sounds about right for for a 6L6GC in a guitar amp. It might not be the best sounding bias point for the guitar amp but the tubes will last a long time; longer than the warranty in most cases, which is the most important thing to the amp company!
Most guitar amps that I have serviced in the past bias the tubes at about half of what each output tube's dissipation rating is. Again, not the best sounding point, but a safe operating point where the chances for tube failure is slim.
Daniel
IME, guitar amps generally sound best biased somewhere between 50 and 70% of the power tubes' dissipation rating. Any more than that makes them kind of mushy sounding. Too low and you get audible crossover distortion. Fenders generally sound better biased a little hotter than Marshalls do. I generally set the bias by ear to find the point where it sounds best to me, then measure to make sure that I've set it within 75% of the tubes' dissipation rating. It always is.
BTW, not all Twin Reverbs have bias balance controls. The blackface ones (65 - 67) had a true bias adjustment pot. Also, a lot of silverface Fenders will have been converted from bias balance to true bias adjustment. It's a very simple conversion. I think it's a good idea to convert the bias balance pot to a true bias adjustment pot in any silverface Fender amp you have and buy matched tubes.
I think Fender ditched bias balance controls for good in the late 70s or early 80s. The bias balance control was intended to be used for hum balance adjustment -- IIRC, it might even be labeled as a hum balance control on some schematics. In the late 70s or early 80s, Fender put a hum balance control on the back the chassis, which was a 200-ohm 2-watt pot across the heater circuit with the wiper connected to ground. (That was also when it switched back to blackface cosmetics, but those aren't the amps that are normally referred to as blackface Fenders). I think Fender dropped the bias balance circuit at the same time (which would only make sense). However, if I recall correctly, it did not go back to using a bias adjustment control in all of its fixed-bias amps. I think it made some of them with a fixed, nonadjustable bias circuit, like it did in the 50s. If you have one of those, it might be a good idea to replace one of the resistors in that circuit with a trimmer to make the bias adjustable, because amps with non-adjustable fixed bias are usually set pretty cold from the factory.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.