I'll have to try out the schematic in a different program, the model for the MOSFET may be bad and in the program I'm using creating new device models is a real pain. One thing that confuses me about the circuit is the ring constant current source - if I remember correctly the current through transistor Q5 in the circuit is the base to emitter drop of transistor Q4 divided by R5, so with the values in the circuit that's about 6 mA...but 6 mA through 82k resistor R3 would give a voltage drop of 492 volts. 
Obviously the circuit works fine so if someone could point out where I'm going wrong in my thinking I'd appreciate it!
Obviously the circuit works fine so if someone could point out where I'm going wrong in my thinking I'd appreciate it!
Bitrex said:I'll have to try out the schematic in a different program, the model for the MOSFET may be bad and in the program I'm using creating new device models is a real pain. One thing that confuses me about the circuit is the ring constant current source - if I remember correctly the current through transistor Q5 in the circuit is the base to emitter drop of transistor Q4 divided by R5, so with the values in the circuit that's about 6 mA...but 6 mA through 82k resistor R3 would give a voltage drop of 492 volts.
What are you using to simulate? You can download ltspice (legally free) and I can give you working sim files.
Hi wuffer,
I'm actually in the process of building a FVP5a and if my simulation is right the circuit ill draw 30mA per channel - that for both phono and line stages. Each of these will draw close to 15mA each.
Let me know if you want more details on the result of my simulation.
Take care,
Luis
Hello everyone,
I've spent the last few days reading the whole thread! Thank you for sharing all the experiments and knowledge and a special thanks for Salas for sharing the great schematic!
As I've posted above I'm in the process of building a FVP5a by Allen Wright as seen here.
According to my simulation the circuit will draw 30mA per channel so I guess a 120mA shunt regulator is in order. Which will be powered by a > 200mA transformer if my EE is right.
A 230V transformer will do the trick, this will give over 300V DC so the voltage drop of 50V for the regulator plus the resistor in the CRC filter is covered. If my math is OK this is a 50 VA transformer or larger. Can anyone confirm this?
A few of questions:
- Would you make two regulators @ 60mA one for each channel?
- What about two regulators @ 60mA, one for the line stages and another for the phono stages? Making four regulators is out of the question for now!!
- How would you dimension the capacitors before the regulator? I guess a CRC filter will be required but is there a rule of thumb for the values required - say 150uF for 120mA, or in this case 2x 150uF with a resistor between them - something along those lines? How do I calculate the value of the resistor?
One last question (for now...) what would you recommend for the heater supplies? In a phono + line stage preamp AC heaters are not a good thing, are you just rectifying and filtering and then put a series resistor in order to get 6.3V DC at the given current? Maybe use a series regulator like the LM317?
Salas, how did you go about this since you're using the shunt regulator in a phono stage? How are you heating those tubes?
Maybe a low voltage shunt regulator is in order!!
Take care,
Luis
I've spent the last few days reading the whole thread! Thank you for sharing all the experiments and knowledge and a special thanks for Salas for sharing the great schematic!
As I've posted above I'm in the process of building a FVP5a by Allen Wright as seen here.
According to my simulation the circuit will draw 30mA per channel so I guess a 120mA shunt regulator is in order. Which will be powered by a > 200mA transformer if my EE is right.
A 230V transformer will do the trick, this will give over 300V DC so the voltage drop of 50V for the regulator plus the resistor in the CRC filter is covered. If my math is OK this is a 50 VA transformer or larger. Can anyone confirm this?
A few of questions:
- Would you make two regulators @ 60mA one for each channel?
- What about two regulators @ 60mA, one for the line stages and another for the phono stages? Making four regulators is out of the question for now!!
- How would you dimension the capacitors before the regulator? I guess a CRC filter will be required but is there a rule of thumb for the values required - say 150uF for 120mA, or in this case 2x 150uF with a resistor between them - something along those lines? How do I calculate the value of the resistor?
One last question (for now...) what would you recommend for the heater supplies? In a phono + line stage preamp AC heaters are not a good thing, are you just rectifying and filtering and then put a series resistor in order to get 6.3V DC at the given current? Maybe use a series regulator like the LM317?
Salas, how did you go about this since you're using the shunt regulator in a phono stage? How are you heating those tubes?
Maybe a low voltage shunt regulator is in order!!
Take care,
Luis
luis.martins said:
Salas, how did you go about this since you're using the shunt regulator in a phono stage? How are you heating those tubes?
Maybe a low voltage shunt regulator is in order!!
Take care,
Luis
They are heated DC in series, so 2X 317s at 12.2V out per tube pair.
The lowest voltage shunt I made is for Origin Live TT DC Motor mod on a Pink Triangle Anniversary. That one has a power Darlington as
shunt, a ring of two Vref like this one, and it is pre regulated by 317. Goes from 2-4.5V and runs at 1A. Two multiturn trimmers and a switch between them for 33/45rpm.
6 HV shunt regs like this one you wanna make I know are in work. 1 for me, 2 for Cygnus X1, 1 for Stixx, 1 for Marinos, 1 for Kofi Annan. We all use about double the audio circuit's steady state consumption for shunt's CCS set current.
If you are in for the best subjective qualities, prefer 2 regs. One per channel at 50mA.
Hi Jan,
In my simulation is runs just a little over 27mA, just for the phono and line stages. Then I added the voltage dividers to get elevated heaters, two of them, one for 30v and another for 150v. That's another 2mA. I rounded it to 30mA... But like I said, it's my simulation and the models may not be very accurate.
The 60mA is the current going through the CCS in the shunt regulator, half of it goes to the preamp and half to the shunt itself. Maybe I wasn't very clear on that.
Take care,
Luis
Hi Salas,
Thank you for your reply!
2x 317s as in one for each channel?
Why do you run the heaters at that voltage? Shouldn't it be 2x 6.3 = 12.6V ? I must be missing something here....
I guess I'll go that way... Can I still power them from the same transformer, i.e. take the ~300V from the filtering caps? Or that would defeat the purpose of having two regulators?
Thanks again,
Luis
Thank you for your reply!
2x 317s as in one for each channel?
Why do you run the heaters at that voltage? Shouldn't it be 2x 6.3 = 12.6V ? I must be missing something here....
I guess I'll go that way... Can I still power them from the same transformer, i.e. take the ~300V from the filtering caps? Or that would defeat the purpose of having two regulators?
Thanks again,
Luis
Hi Jan,
Since you seem to know this circuit quite well, maybe you could help me here on a couple of things:
1. How should I get the 20V required to bias the lower tube on the line stage? I haven't figured that out. Will a voltage divider be OK? Say two series resistors: 230K + 20k or something along those lines? Maybe a 20V zener diode?
2. I can't seem to find 25K resistors for the plate loads, can I use 24K resistors (Kiwame 5W) or this is a critical value and will change the circuit bias?
Take care,
Luis
I got less hiss at 12.2V and there are statistics for longer life without real loss of transconductance after initial hours. I use one reg per tube type pair. Bottom-upper half of cascode. Common between channels.
Use 2 transformers 2 UF4007 bridges, 2 CRC pre filters. Not a big expenditure given your low power needs. You will appreciate the difference. 2 shunts with common power source still sound better, but only half better than full monoblock.
Use 2 transformers 2 UF4007 bridges, 2 CRC pre filters. Not a big expenditure given your low power needs. You will appreciate the difference. 2 shunts with common power source still sound better, but only half better than full monoblock.
Hi Salas,
Thank you very much for the quick reply!
Given the total cost of the project, the cost of 1 more transformer isn't going to make that much of a difference so I guess I'll take your advice.
Just checking the Hammond 182D240 - 30 VA / 240V @ 125ma - $53.76
I guess this will be OK to power the 60mA regulator.
Could you please provide some info for the CRC? What values of capacitors and resistor should I be looking for with this configuration?
Take care,
Luis
Thank you very much for the quick reply!
Given the total cost of the project, the cost of 1 more transformer isn't going to make that much of a difference so I guess I'll take your advice.
Just checking the Hammond 182D240 - 30 VA / 240V @ 125ma - $53.76
I guess this will be OK to power the 60mA regulator.
Could you please provide some info for the CRC? What values of capacitors and resistor should I be looking for with this configuration?
Take care,
Luis
CRC values
Hey Luis,
Strongly suggest you download Duncan Amps PSU Designer and use it to simulate your power supplies. (It is a free download) Don't forget to check both voltage AND amperage results. Big filter caps will give you least voltage ripple, but you will get a nasty shock when you look at the amperage results. Small caps, the opposite. It is a balancing act.... If you can afford it, CLC is generally better than CRC, particularly when there are so few filter stages.
Just my .02c worth.
Hey Luis,
Strongly suggest you download Duncan Amps PSU Designer and use it to simulate your power supplies. (It is a free download) Don't forget to check both voltage AND amperage results. Big filter caps will give you least voltage ripple, but you will get a nasty shock when you look at the amperage results. Small caps, the opposite. It is a balancing act.... If you can afford it, CLC is generally better than CRC, particularly when there are so few filter stages.
Just my .02c worth.
Don't use more than 100uF-Resistor-100uF for each channel IMO. No reason to have a heavy and strongly reactive pre filter. It will only stress the 50VA trafo, and create inrush peaks that can zap the CCS Mosfet erratically. Of course the Mosfets can be clamped with Zeners, but the less components the better. Its a regulator, it will add its own -non spectacular in this case- but enough ripple rejection too. Using a choke instead of a resistor is beneficial for RF rejection as well. Use one with at least 100R self resistance so to retain the slow charge of a CRC too.
Its nice to add a choke, but not essential. Double mono mainly adds to subjective goodness. This simple shunt takes some caution when setting up. It is not protected and can change current easily by LEDs different than spec, or enthusiastic trim pot turning. Please read my initial posts about calculating it and starting it. Don't probe around for no reason when live, use bleeder resistor across the last filter capacitor, and keep short twisted wiring. Its dirt cheap for main circuit, keep some spares. If not familiar and on a bad start, can be a PITA. Read Gary Pimm's lessons on nodes and grounds in this thread before you lay it out, it can save a lot of trouble and guarantee its intrinsic qualities. Ah, did I mention why bother then? People say that it sounds rather good.
Its nice to add a choke, but not essential. Double mono mainly adds to subjective goodness. This simple shunt takes some caution when setting up. It is not protected and can change current easily by LEDs different than spec, or enthusiastic trim pot turning. Please read my initial posts about calculating it and starting it. Don't probe around for no reason when live, use bleeder resistor across the last filter capacitor, and keep short twisted wiring. Its dirt cheap for main circuit, keep some spares. If not familiar and on a bad start, can be a PITA. Read Gary Pimm's lessons on nodes and grounds in this thread before you lay it out, it can save a lot of trouble and guarantee its intrinsic qualities. Ah, did I mention why bother then? People say that it sounds rather good.
You are such a modest guy...Have 20+ hours playing time on my shunt now, and I am liking it better and better. Going trough very well known recordings makes me realize how different it is (in a positive way) from my previous psu. Music is presented absolutely effortless and natural, it is lightning fast ('xcuse the big words...) and on good recordings details are rendered so precisely you want to reach over and grab them...
Built No.2 with tighter layout is in the works
.
Since the question, how much CCS current is enough has been asked, and I steadily recommend about twice their bias current for servicing class A circuits, here is a good post on the matter from AndrewT in a relative discussion:
Hi,
determine the continuous quiescent load current.
Determine the maximum increase in current for worst case output conditions.
The CCS must pass more than that total.
The shunt needs to pass more than the increase.
If the minimum current falls to near zero then the shunt must be able to pass the whole CCS current. The two big questions become:
What is the average dissipation in the CCS pass transistor?
What is the average dissipation in the Shunt transistor.
eg.
A single ended ClassA output stage draws 50mA of quiescent current. The drive circuits draw a further 10mA.
The total current in the quiescent state is 60mA.
The Output stage varies it's current draw by +-20mA.
The minimum current is 40mA and the maximum is 80mA.
The Shunt and CCS use 15mA to ground.
The CCS must pass > 80+15>95mA. Set to 100mA.
The minimum current is 40+15-(CCS current) ~=50mA.
The shunt must be able to pass >100-50> 50mA.
Set the CCS to 100mA, set the amp to work and in the quiescent state the Shunt will pass 100-40-10~=50mA.
Check dissipations in the CCS and Shunt and from there calculate the heatsinking requirements.
To design any Shunt regulator you MUST know the maximum and minimum current draw of the load.
__________________
regards Andrew T.
Original post's link
Hi,
determine the continuous quiescent load current.
Determine the maximum increase in current for worst case output conditions.
The CCS must pass more than that total.
The shunt needs to pass more than the increase.
If the minimum current falls to near zero then the shunt must be able to pass the whole CCS current. The two big questions become:
What is the average dissipation in the CCS pass transistor?
What is the average dissipation in the Shunt transistor.
eg.
A single ended ClassA output stage draws 50mA of quiescent current. The drive circuits draw a further 10mA.
The total current in the quiescent state is 60mA.
The Output stage varies it's current draw by +-20mA.
The minimum current is 40mA and the maximum is 80mA.
The Shunt and CCS use 15mA to ground.
The CCS must pass > 80+15>95mA. Set to 100mA.
The minimum current is 40+15-(CCS current) ~=50mA.
The shunt must be able to pass >100-50> 50mA.
Set the CCS to 100mA, set the amp to work and in the quiescent state the Shunt will pass 100-40-10~=50mA.
Check dissipations in the CCS and Shunt and from there calculate the heatsinking requirements.
To design any Shunt regulator you MUST know the maximum and minimum current draw of the load.
__________________
regards Andrew T.
Original post's link
I don't know if Andrew's reading this thread, but I'd like to add something. The shunt regulator will have an increased idle current when the load current goes down. IMHO, to be on the safe side, one should assume a minimum load of 0 amps, i.e. no load. Most likely you're powering a tube amp/preamp with a HV regulator. Any tube can fail, resulting in either half the load current, or, if failing on both channels, zero load current. Then, can the shunt element take it? Because under no load the shunt element will reach the highest current. Even if you're willing to take the risk of failing tubes, when you turn on your circuit chances are that for quite a few seconds, until the filaments heat up, there load is quite low on the shunt regulator. Just my 2c, please correct me if I'm wrong.
Missed that :/ ... must have been too 
But then, if one must safeguard against the worst case scenarios, zero and max load, these two should probably be emphasized, which I guess, I was trying to do for zero load. Anyway, all I'm trying to say to people out there is, watch out for a ton of dissipation on the shunt when there's no load.
But then, if one must safeguard against the worst case scenarios, zero and max load, these two should probably be emphasized, which I guess, I was trying to do for zero load. Anyway, all I'm trying to say to people out there is, watch out for a ton of dissipation on the shunt when there's no load.