I made more measurements today. Unlike yesterday when the output voltage increased when I decreased the load, today the voltage was dropping. No idea why. Here are the measurements:
All I get from this is that the constant current sources in both units appear to be producing the same current. All current limiting transistors are off.
In the bad unit the input to the op amp is lower as is the output. This turns the pass MOSFET off more thereby reducing the output voltage.
None of this is helping me diagnose the problem though.
| Initial setup: 45V in. Both setup for 25V out into 560R (44.6mA) | ||
| Measurements with load reduced to 220R (113.6mA) | ||
| Good unit | Bad unit | |
| Vout | 25V | 22.7V |
| AD8031 output | 30.2V | 25.4V and going down -> 24.6V etc |
| Q3 Vbe | 0.274V | 0.274V |
| Q2 Vbe | 0.2V | 0.2V |
| Voltage across R4 | 0.56V | 0.56V |
| Voltage across R11 | 24.9V | 22.8V and going down |
| Voltage across R1 | 0.56V | 0.56V |
| Q5 Vbe | 0V | 0V |
| Voltage at junction of R1, VR1, R4 etc | 22.7V | 23.7V and going down -> 23V etc |
All I get from this is that the constant current sources in both units appear to be producing the same current. All current limiting transistors are off.
In the bad unit the input to the op amp is lower as is the output. This turns the pass MOSFET off more thereby reducing the output voltage.
None of this is helping me diagnose the problem though.
Next if you want to post measurements please measure against Vout, not against Gnd.
The whole regulator floats on Vout so that's the reference.
It appears that the opamp is off.
Assuming it is not broken, it thinks that Vout is too high so it tries to shut down the pass device.
Check the inputs of the opamp (against Vout).
Check the polarity of the diodes around the opamp circuit, are they the correct way around?
Jan
The whole regulator floats on Vout so that's the reference.
It appears that the opamp is off.
Assuming it is not broken, it thinks that Vout is too high so it tries to shut down the pass device.
Check the inputs of the opamp (against Vout).
Check the polarity of the diodes around the opamp circuit, are they the correct way around?
Jan
Thanks for your patience Jan. I've re-done the measurements, referenced to Vo.
D7A and D7B are the right way round.
Referenced to Vo, there's very little difference between the measurements so I swapped meters to one with 4 decimal places.
I think the opamp is working. 3rd and 4th decimal places, but the problem unit has double the Δ across the input pins of the op amp. There's a 2nd decimal place difference in the voltages on the pass device gate.
Sadly the measurements still aren't identifying the problem for me. I thought maybe I had a different value of R11 in the problem unit but it's exactly the same.
Does anything jump out at you?
D7A and D7B are the right way round.
Referenced to Vo, there's very little difference between the measurements so I swapped meters to one with 4 decimal places.
I think the opamp is working. 3rd and 4th decimal places, but the problem unit has double the Δ across the input pins of the op amp. There's a 2nd decimal place difference in the voltages on the pass device gate.
Sadly the measurements still aren't identifying the problem for me. I thought maybe I had a different value of R11 in the problem unit but it's exactly the same.
| Initial setup: 45V in. Both setup for 25V out into 560R (44.6mA) | ||
| Measurements with load reduced to 220R (113.6mA) with reference to Vout | ||
| Good unit Vo 25.018V. Bad unit Vo 23.329V and heading down. | ||
| Good unit | Bad unit | |
| Vin | 22.5 | 24.3 |
| Junction of R1, VR1, R4 etc | 0.0010 | 0.0033 |
| Junction of D6, C11, Q8 source etc | 7.918 | 8.210 |
| AD8031 input pin 3 | 0.0011 | 0.0026 |
| AD8031 output pin 6 | 5.0470 | 5.0243 |
| Gate of Q7 | 5.0470 | 5.0243 |
| Source of Q7 | 0.2760 | 0.2805 |
| Q3 Vbe | 0.274 | 0.274 |
| Q2 Vbe | 0.2 | 0.2 |
| Q5 Vbe | 0 | 0 |
| Voltage across R4 | 0.56 | 0.56 |
| Voltage across R11 | 24.9 | 22.9 |
| Voltage across R1 | 0.55 | 0.55 |
Does anything jump out at you?
No nothing does, it's weird, the opamp seems to working OK.
The Vin value you noted, is that the Vin referenced to Vout? Seems OK to.
If it was a premature limiting of the current, you'd see in it the opamp node voltages.
One question: if you increase the current further say to 150mA, does the 'good' unit also start to drop?
Try this with the current limiting disabled (shorted limit set resistor).
I am at an industry convention today and tomorrow (IBC in Amsterdam) so I can't really work on it now, but will as soon as I get home.
Jan
The Vin value you noted, is that the Vin referenced to Vout? Seems OK to.
If it was a premature limiting of the current, you'd see in it the opamp node voltages.
One question: if you increase the current further say to 150mA, does the 'good' unit also start to drop?
Try this with the current limiting disabled (shorted limit set resistor).
I am at an industry convention today and tomorrow (IBC in Amsterdam) so I can't really work on it now, but will as soon as I get home.
Jan
Really. Type the 1st part number in Google, and the first hit is:
https://eu.mouser.com/ProductDetail...T*MTcwNTI1NTcxOC4yLjEuMTcwNTI1NTcyMS41Ny4wLjA.
82 in stock.
Jan
https://eu.mouser.com/ProductDetail...T*MTcwNTI1NTcxOC4yLjEuMTcwNTI1NTcyMS41Ny4wLjA.
82 in stock.
Jan
If current is less than 350mA then I would suggest FCP190N60E or FCP190N65F. Mouser has plenty in stock and it's cheap.
Hello one and all. I come to this site from the audio vs engineering side of things. I have bought a couple of T-Reg boards for an amplifier project I am working on.
I have a question that I would like a little help on from those of you that are more knowledgeable than myself (Most of you).
Acknowledging that there are very few original thoughts - I got the bright idea to try to put together a modular, 3 enclose amp - one box for DC heater supplies, one box for B+ supplies and a 3rd for the tube circuits and output transformers.
I am attempting this with the goal of being able to use an adjustable power supply when switching between the 45 tube, 6a3 and 300b tubes. I am also using a Tent Labs fixed bias supply. I plan to mount remote potentiometers so that voltages can be changed without opening up the amp.
My question has to do with using the T-Reg to adjust the B+ voltage when changing tubes. The 45 tubes wants a B+ of potentially as low as 200 volts and the 300b could use up to around 400 volts.
Besides being wasteful of energy (I don’t feel bad about this because i have solar power) - will the T-reg be suitable for this application and is there a performance penalty of any kind by using this board vs separate B+ supplies?
Thanks very much for any thoughts on the matter.
Best,
Lou
I have a question that I would like a little help on from those of you that are more knowledgeable than myself (Most of you).
Acknowledging that there are very few original thoughts - I got the bright idea to try to put together a modular, 3 enclose amp - one box for DC heater supplies, one box for B+ supplies and a 3rd for the tube circuits and output transformers.
I am attempting this with the goal of being able to use an adjustable power supply when switching between the 45 tube, 6a3 and 300b tubes. I am also using a Tent Labs fixed bias supply. I plan to mount remote potentiometers so that voltages can be changed without opening up the amp.
My question has to do with using the T-Reg to adjust the B+ voltage when changing tubes. The 45 tubes wants a B+ of potentially as low as 200 volts and the 300b could use up to around 400 volts.
Besides being wasteful of energy (I don’t feel bad about this because i have solar power) - will the T-reg be suitable for this application and is there a performance penalty of any kind by using this board vs separate B+ supplies?
Thanks very much for any thoughts on the matter.
Best,
Lou
Hi Lou, the T-reg performance does not change with output voltage due to its design.
But be mindful of dissipation, assuming you are not changing the raw input voltage.
For 400V out you probably will have something like 420V input, leaving 20V across the series device.
If you change the output to 200V, you now have 220V across the output device. That's a dissipation of 22W per 100mA output current, which gets hot quickly.
Ideally you'd use a dual switch or something to switch both the input and the output.
Or use two T-regs, one set up for 200V with 220V in and one for 400V with 420V in.
With a bit of creativity you can use a center tapped transformer to set up 220V and 440V from a single transformer.
Jan
But be mindful of dissipation, assuming you are not changing the raw input voltage.
For 400V out you probably will have something like 420V input, leaving 20V across the series device.
If you change the output to 200V, you now have 220V across the output device. That's a dissipation of 22W per 100mA output current, which gets hot quickly.
Ideally you'd use a dual switch or something to switch both the input and the output.
Or use two T-regs, one set up for 200V with 220V in and one for 400V with 420V in.
With a bit of creativity you can use a center tapped transformer to set up 220V and 440V from a single transformer.
Jan
Thanks for your response Jan. I had thought of placing a switchable drop resistor before the T-reg to take a good amount of voltage drop (and heat) outside of your device - but I was just guessing.
I have 2 T-Regs with the idea of dual mono B+ and a separate power supply for the driver tube. I am an old audio guy with time on his hands (not enough) and am trying to make a dream amplifier. Usually I prefer simple vs complicated, but I am hoping that with care and a pre-flight checklist it will turn out well!
Thanks for your suggestion.
Lou
I have 2 T-Regs with the idea of dual mono B+ and a separate power supply for the driver tube. I am an old audio guy with time on his hands (not enough) and am trying to make a dream amplifier. Usually I prefer simple vs complicated, but I am hoping that with care and a pre-flight checklist it will turn out well!
Thanks for your suggestion.
Lou