So, I assembled the TubelabSE for 300B.
I am using a transformer 360-0-360 VAC, with 40 ohm resistance.
To lower the voltage I am using a 10 uF for C4, choke 93 ohm, and 5R4WGB that has a bigger drop than the 5AR4.
Checked filament voltages: OK
Only with rectifier I have B+ 510 V
Connected the drivers, I manage to have one at 175, but the other does not go over 165 V. Swapped the tubes problem changed side, so I guess the tubes are pretty different. Put both at 165 V.
Connected the 300BC.
Now they conduct heavily, the B+ goes to 274 V and does not vary much does not matter what I do with the trimmer. I have 1.2 V on R18 and R29.
I added two 300 ohm resistors (20W) between the transformer and the board, less B+.
Now B+ 211 V and voltage on R18 and R29 to 1.1 V.
So did I blow anything ?
Please help,
Davide
I am using a transformer 360-0-360 VAC, with 40 ohm resistance.
To lower the voltage I am using a 10 uF for C4, choke 93 ohm, and 5R4WGB that has a bigger drop than the 5AR4.
Checked filament voltages: OK
Only with rectifier I have B+ 510 V
Connected the drivers, I manage to have one at 175, but the other does not go over 165 V. Swapped the tubes problem changed side, so I guess the tubes are pretty different. Put both at 165 V.
Connected the 300BC.
Now they conduct heavily, the B+ goes to 274 V and does not vary much does not matter what I do with the trimmer. I have 1.2 V on R18 and R29.
I added two 300 ohm resistors (20W) between the transformer and the board, less B+.
Now B+ 211 V and voltage on R18 and R29 to 1.1 V.
So did I blow anything ?
Please help,
Davide
Something doesn't sound right. With the drivers, I had a similar issue with getting close to the "recommended 175 Volts". To correct that, I increased the driver current by dropping the CCS reference resistor values. As designed per the spec sheet, the CCS, it will hold ~7mA. I increased the current setpoint to 12mA. Well within spec of the 417a driver tube. You can find the value by looking at the curve of the CCS. I don't have them available to me now otherwise I'd check on it.
Now on to the output tubes. Once you set up the driver voltage, you need to set your bias voltage on the 300B. When you start out (before you insert the 300B tubes) it needs to be as negative as possible. What reading do you have at this point? Recalling from memory I was able to get to something like -140Volts. You should be okay if you can get it to -100. If you can't then I would start to trouble shoot the bias circuit. You might have a wrong value or a bad connection.
As far as the tube(s) are concerned, It is hard to say. If you didn't overload them for long, they probably are fine. They aren't bullet proof but are able to handle more abuse than semiconductors.
As far as the 300Ohm resistance, that seem like alot prior to the rectifier. Follow the recomended practices for the rectifier tube on the spec sheet. If more voltage drop is needed you can add more resistance after the choke.
Now on to the output tubes. Once you set up the driver voltage, you need to set your bias voltage on the 300B. When you start out (before you insert the 300B tubes) it needs to be as negative as possible. What reading do you have at this point? Recalling from memory I was able to get to something like -140Volts. You should be okay if you can get it to -100. If you can't then I would start to trouble shoot the bias circuit. You might have a wrong value or a bad connection.
As far as the tube(s) are concerned, It is hard to say. If you didn't overload them for long, they probably are fine. They aren't bullet proof but are able to handle more abuse than semiconductors.
As far as the 300Ohm resistance, that seem like alot prior to the rectifier. Follow the recomended practices for the rectifier tube on the spec sheet. If more voltage drop is needed you can add more resistance after the choke.
Last edited:
You are blowing R5? That resistor doesn't dissipate that much...it is acting more like a fuse. I suspect you have a shorted cap...possibly C6. I actually put 450V caps in the C6 and C7 positions because I was concerned about the voltage across them before B+ comes up and starts drawing current though the bias circuit. My parts list shows:
C6,C7 - Capacitor - 100uF - Aluminum Electrolytic - 450V, 105^C, 650mA - 40x18mm, 7.5mm - Nichicon - PT Series - Mouser - 647-UPT2W101MHD - $3.31 - 2 - Driver bias supply filter (360V < B+ < 400)
EDIT: you already have 450V caps here, so nevermind about that.
If you are drawing enough current through the bias circuit to fry that resistor, then I would look more closely at the bias circuit. Like Gregory said, check the actual grid voltage going to the output tubes. Also double-check R6. On my amp, this resistor dissipates far more than oany of the others in the bias circuit. I put a 7W part there.
I wouldn't worry about the 5842 operating point. Have 16 of them on hand and the NOS Ratheon ones are all over the map.
C6,C7 - Capacitor - 100uF - Aluminum Electrolytic - 450V, 105^C, 650mA - 40x18mm, 7.5mm - Nichicon - PT Series - Mouser - 647-UPT2W101MHD - $3.31 - 2 - Driver bias supply filter (360V < B+ < 400)
EDIT: you already have 450V caps here, so nevermind about that.
If you are drawing enough current through the bias circuit to fry that resistor, then I would look more closely at the bias circuit. Like Gregory said, check the actual grid voltage going to the output tubes. Also double-check R6. On my amp, this resistor dissipates far more than oany of the others in the bias circuit. I put a 7W part there.
I wouldn't worry about the 5842 operating point. Have 16 of them on hand and the NOS Ratheon ones are all over the map.
Last edited:
So It looks like I did not blow anything beside R5. I bought today a new resistor that actually looks 3W that actually looks much bigger than the one mouser sent me labeled as 2W. This did not blow...yet.
I am running with 5k of impadence at 384 V and 66 mA of Bias. Tomorrow I am going to have a matching session in the shop where I got the drivers to find two that perform alike.
Let's see...
Few questions: What would you upgrade the cathode bypass caps with ?
And what about the PSU caps for the bias ?
What size of choke do you recommend ?
Thanks,
Davide
I am running with 5k of impadence at 384 V and 66 mA of Bias. Tomorrow I am going to have a matching session in the shop where I got the drivers to find two that perform alike.
Let's see...
Few questions: What would you upgrade the cathode bypass caps with ?
And what about the PSU caps for the bias ?
What size of choke do you recommend ?
Thanks,
Davide
This is what puzzle me, I have never seen more than 2 V, so the power dissipation is nothing. Maybe the resistors I had were just defective.
Now I saw another problem, I think that the 5V DC regulator goes onto temperature protection. As all of a sudden the current went to zero. I switched off and on after one minute and it was fine. Although this is again not good for the tubes. I'll buy a bigger heatsink tomorrow and try again
Let's again hope that tube are safe, they run for one minute with high voltage and no heaters :-(
D.
Now I saw another problem, I think that the 5V DC regulator goes onto temperature protection. As all of a sudden the current went to zero. I switched off and on after one minute and it was fine. Although this is again not good for the tubes. I'll buy a bigger heatsink tomorrow and try again
Let's again hope that tube are safe, they run for one minute with high voltage and no heaters :-(
D.
R5 should not be critical.
A hole in the plate would be a bad sign. Loss of the getter another. The rest of the damage would be that the grid is warped or broken. It is nearly impossible to see that however. The result would be tube runaway (broken grid) or distortion (warped).
I doubt you did any damage with the filament going cold. For testing purposes you could just blow a small fan on the heatsink until the bigger on arrives.
A hole in the plate would be a bad sign. Loss of the getter another. The rest of the damage would be that the grid is warped or broken. It is nearly impossible to see that however. The result would be tube runaway (broken grid) or distortion (warped).
I doubt you did any damage with the filament going cold. For testing purposes you could just blow a small fan on the heatsink until the bigger on arrives.
Same thing happened to me when I started testing with 300Bs. The stock HS could manage with the 45's heater requirements, but the 300Bs were too much. It shutdown several times before I figured it out. There were no ill effects. I clamped a squangle to it to get by until I started designing the chassis.
I'm trying to imagine how the heater shutdown could cause R5 to go. The only thought I have is the same as startup: excessive B- voltage. You have 450V caps like I do, so I doubt it.
I'm trying to imagine how the heater shutdown could cause R5 to go. The only thought I have is the same as startup: excessive B- voltage. You have 450V caps like I do, so I doubt it.
The problem with R5 and the filament are not connected. Probably the resistors I was using were not so good or the had a lower rating then declared. They looked a bit small to me. With the new one I got no problem.
Putting the bigger heatsink fixed the filament voltage shutdown too. I am going to do some serious listening tomorrow.
Now I am running the 300B at 380 V B+ and 69mA bias. The driver are both at 175 V. I found out that one 5842 takes just much more time than the other to go up with the voltage.
I hope I did not damage any other component, as I have 770 VAC, so when I start up I have a peak around 480 V. I am worried about the mosfet. I don't know how they behave if they are broken.
I had the valve tested today and they should be ok. For the damage I was more worried about me passing 130 mA when the max rating is 100mA, although it was for a very short time.
Thanks,
D.
Putting the bigger heatsink fixed the filament voltage shutdown too. I am going to do some serious listening tomorrow.
Now I am running the 300B at 380 V B+ and 69mA bias. The driver are both at 175 V. I found out that one 5842 takes just much more time than the other to go up with the voltage.
I hope I did not damage any other component, as I have 770 VAC, so when I start up I have a peak around 480 V. I am worried about the mosfet. I don't know how they behave if they are broken.
I had the valve tested today and they should be ok. For the damage I was more worried about me passing 130 mA when the max rating is 100mA, although it was for a very short time.
Thanks,
D.
MOSFETs are a little weird in how they can fail. Most silicon is working or shorted or popped. Large FETs can be overloaded where damage occurs to the junction but the FET still functions to some degree. I've done this in a regulator circuit and usually the operating point becomes unstable when that happens. In the case of this design I would expect the output tube bias to drift a lot. If the bias is stable, the FETs are probably fine.
As far as overloading the tube briefly...I wouldn't worry about it. Tubes can handle short durations outside of their safe operating conditions. One nice thing about their physical size, I guess.
As far as overloading the tube briefly...I wouldn't worry about it. Tubes can handle short durations outside of their safe operating conditions. One nice thing about their physical size, I guess.
So after one morning of listening, I do not think I blowed anything. The listening is very plesant.
Due to the higher voltage of the power supply, I am running at 380 V B+, that put around 365 V on the plates. The B- is -70V. The drivers have 175 V on the plate.
What I noticed, is that this amp needs a much higher signal level compared to the Simple SE. With my CD player I could turn the volume knob to the maximum and it still was not to loud. My speakers are not the most efficient ones.
Is it like this also with your amps ?
Best Regards,
Davide
Due to the higher voltage of the power supply, I am running at 380 V B+, that put around 365 V on the plates. The B- is -70V. The drivers have 175 V on the plate.
What I noticed, is that this amp needs a much higher signal level compared to the Simple SE. With my CD player I could turn the volume knob to the maximum and it still was not to loud. My speakers are not the most efficient ones.
Is it like this also with your amps ?
Best Regards,
Davide
The Simple SE has a lot of gain, especially in UL mode. Mine will go well into clipping at about 80% of "line level" signal. The Tubelab SE can take 100%, depending if I am running 45s or 300Bs. What tubes do you have in the SSE? I found that the SSE with EL34s in triode would start clipping at about the same volume level as the TSE with 300Bs, but the SSE still needs less input to get there.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.