Ben Mah - thanks for the suggestion. You are exactly right. I found a short from one of the transformer secondaries to the chassis - it turns out that I put a small tear in the transformer's cover installing it. I fixed the tear and re-installed it with some additional insulation. Now there is no voltage getting to the chassis and the power supply is operating correctly with the thermistor to the chassis.
Thanks again! bmdduck
T
Thanks again! bmdduck
T
I see the PSU V3.0 accomodatea caps from 10-15.000uf. Can I use 22k (I know I can), without changig resistor values? Are there any drawbacks in doing this, as opposed to two banks on two different boards?
Baught a half-built project with dual mono, but I want to have full control as well as simplify, so considering using only one of the PSU boards and transformers.
Regards,
Andreas
Baught a half-built project with dual mono, but I want to have full control as well as simplify, so considering using only one of the PSU boards and transformers.
Regards,
Andreas
I see. Thats good news! Ehats your experience in using a 600VA transformer with the F5T V2? I know 6L6 built it with that. Having two 600VA trabsformers, I kinda feel using one is little, and using two is complete overkill as well as filling the chassis bottom to the point of zero flexibility.
Also, spending a lot of time studying how to succeed with the amplifier boards, i haven’t spent as much time on the allready finished PSU theory. So, I kinda also want to know why one would choose to do the following:
1: Use or not use the optinional resistors, and if using them: which of the optional should be used for what purpose? My boards have 1 unused resistor spot on each PSU board. But i don’t know why. This is explained in the PSU board V2 manual, but I cant get my head around why my board is done like this. So considering doing it over. Anyways I need to know how this works. I’ll happily find out myself, but need some guidance or a link to a manual or such. Or Ohms Law for dummies book =)
2: Output snubbers: generic values printed on the board OK, or should I leave them out if I don’t know the exact ringing frequency (if there is one)?
3: Input snubbers: can generic values for a 2x24V 600va (240volts primaries) be calculated, or do I need measuring?
Some say snubbers both in and out of PSU are unnescessary so long as the FE has a cap for doing off with ripple. Which the F5T V2 does. What do you think?
Sorry for all the questions. Truth is I like to know WHY I do things, at least where I can. So I might tend to complicate matters for my self.
Say I di the following:
1: I go with only one transformer.
2: I rebuild/edit the PSU to 22.000uF caps.
How would you set it up?
1: All snubbers, or only output, or none?
2: All resistors, no optional, or some optional?
Btw my bridge diodes do not have electrically isolation between diode and bolt, but i guess it is TO-247, so the diode itself is isolated.
1: Use or not use the optinional resistors, and if using them: which of the optional should be used for what purpose? My boards have 1 unused resistor spot on each PSU board. But i don’t know why. This is explained in the PSU board V2 manual, but I cant get my head around why my board is done like this. So considering doing it over. Anyways I need to know how this works. I’ll happily find out myself, but need some guidance or a link to a manual or such. Or Ohms Law for dummies book =)
2: Output snubbers: generic values printed on the board OK, or should I leave them out if I don’t know the exact ringing frequency (if there is one)?
3: Input snubbers: can generic values for a 2x24V 600va (240volts primaries) be calculated, or do I need measuring?
Some say snubbers both in and out of PSU are unnescessary so long as the FE has a cap for doing off with ripple. Which the F5T V2 does. What do you think?
Sorry for all the questions. Truth is I like to know WHY I do things, at least where I can. So I might tend to complicate matters for my self.
Say I di the following:
1: I go with only one transformer.
2: I rebuild/edit the PSU to 22.000uF caps.
How would you set it up?
1: All snubbers, or only output, or none?
2: All resistors, no optional, or some optional?
Btw my bridge diodes do not have electrically isolation between diode and bolt, but i guess it is TO-247, so the diode itself is isolated.
Last edited:
I strongly recommend dual mono power supplies. Sonic advantages are well documented, and I can confirm it is well worth the trouble/expense.
I started with a single transformer, shared PS for my Aleph J.
Upgraded to dual mono and will never build another amplifier that is not dual mono PS, or mono blocks.
I think you are on the right track. Papa typically builds the First Watt amps with a single transformer, and yours seems more than adequate. Many builders are constructing his DIY class A amps with single 300 or 400 VA transformers.
In my case, I built an F7 imitation using an Antek 300 VA with dual 20-volt windings, 2 separate "monolithic" bridges, and the DIY store Universal Power Supply V3 boards with 4 x 33,000 ufd 35-volt caps on each board. I used 5 x 0.47 Ohm paralleled resistors on each board for the R in the CRC. I did not use snubbers.
These are the caps I used: 380LX333M035A052 Cornell Dubilier - CDE | Mouser
The measured conditions are:
Voltage out: +/- 23 volts
Bias current each channel: 1.2 amps
A/C input current @ 120 VAC: 1.4 amps RMS (~160 Watts) - I am on 120 VAC/60 Hz mains.
Ripple current into each capacitor bank: ~ 4 amps RMS
DC supply output ripple: 6 mV RMS each rail, quiescent
Your mileage will vary, of course, but hopefully, this will give you a calibration point.
Naturally, going true dual-mono with 2 transformers and separate cap boards is the ideal situation, but like everything in DIY audio, there are tradeoffs and I am happy with the results I am getting with the above.
I strongly suggest you try out @duncanamps very nice power supply calculator to make sure you are comfortable with all the tradeoffs and component selection:
PSUD (Power Supply Unit Designer)
Good luck!
Last edited:
I would not think so but the board may be purchased from the diyAudio Store:
Universal Power Supply – diyAudio Store
Universal Power Supply – diyAudio Store
Rectifier / Heatsink confusion
Hi friends
Long story short:
Can I file down a TO-247 package to fit into a TO-220 heatsink?
The case is plastic and relatively thick... but it has to withstand heat-stress?)
(The Difference is approx 0.6 mm (rectifier 16.4 mm versus heatsink 15.88 mm) so that I'd have to remove like 0.5 mm on each side... does this work?)
The whole story goes like this: When I placed my order almost a year back, I didn't spend attention to the details (I didn't quite knew what to look at), and thus chose a basically incompatible combination of the rectifier (Vishay FEP30DP-E3/45 TO-247AD) and the heatsink (Aavid 513102B02500G, for TO-220 Packages)
Thanks!
Hi friends
Long story short:
Can I file down a TO-247 package to fit into a TO-220 heatsink?
The case is plastic and relatively thick... but it has to withstand heat-stress?)
(The Difference is approx 0.6 mm (rectifier 16.4 mm versus heatsink 15.88 mm) so that I'd have to remove like 0.5 mm on each side... does this work?)
The whole story goes like this: When I placed my order almost a year back, I didn't spend attention to the details (I didn't quite knew what to look at), and thus chose a basically incompatible combination of the rectifier (Vishay FEP30DP-E3/45 TO-247AD) and the heatsink (Aavid 513102B02500G, for TO-220 Packages)
Thanks!
Buy the proper heatsinks.
Since I had a Mouser-cart waiting for delivery (1 part was due to arrive in 5 days, so worth the wait), I asked if I somehow could add the missing parts (heatsinks _and_ corrected resistor-values
), and well, mouser just threw it in the bag without hesitating (or asking for more money)...If that's not cool, then...
Is there such a thing as too much transformer VA?
Hi! I was wondering whether there is such as thing as too much transformer VA.
I understand that
1) there may be sufficient VA, i.e. 300-400 VA for an MX2 or aleph J.
2) there may be diminishing sonic returns on further VA
3) that diminishing sonic returns may not be worth the investment in even more VA
However,
- is there any averse effect of too much VA? I am under the impression that VA is simply the potential that can be delivered if the circuit requires it
- does more VA require any changes to the PSU circuit? I thought no.
Hypothesis: Apart from wasting money on VA, too much VA can do no harm. Even though it might be silly, especially from a serious engineering standpoint. However, as stated elsewhere on this forum "only the absolute overkill is ever good enough".
(I suspect that this question has been asked before... though I have not seen it.)
Hi! I was wondering whether there is such as thing as too much transformer VA.
I understand that
1) there may be sufficient VA, i.e. 300-400 VA for an MX2 or aleph J.
2) there may be diminishing sonic returns on further VA
3) that diminishing sonic returns may not be worth the investment in even more VA
However,
- is there any averse effect of too much VA? I am under the impression that VA is simply the potential that can be delivered if the circuit requires it
- does more VA require any changes to the PSU circuit? I thought no.
Hypothesis: Apart from wasting money on VA, too much VA can do no harm. Even though it might be silly, especially from a serious engineering standpoint. However, as stated elsewhere on this forum "only the absolute overkill is ever good enough".
(I suspect that this question has been asked before... though I have not seen it.)
It seems you analysed the pros and cons pretty well.
With too much VA and a substantial capacitor storage bank, the inrush current might be high.... the inrush current surges may loosen the winding/isolation, so the transformer could become noisy over time.
There's weight factor in play as well... but, again, it really comes down to what you deem, personally -> as being too heavy...
With too much VA and a substantial capacitor storage bank, the inrush current might be high.... the inrush current surges may loosen the winding/isolation, so the transformer could become noisy over time.
There's weight factor in play as well... but, again, it really comes down to what you deem, personally -> as being too heavy...
Hi Boky,
thank you for your response!
My understanding is that a high inrush current would sometimes require an inrush current limiter and that this in turn can affect the dynamic output of the amplifier aversely.
Therefore too much capacitance is not necessary good. I remember reading a post from 6L6 that he thought 18.000uF? capacitance in may be the sweetspot for the Aleph J. I guess this is because of the trade off between the benefits of additional capacitance and the averse effect of an inrush current limiter.
However, if I got a toroidal transformer with lots of VA, I do not necessarily need to build a larger capacitor bank, right?
Of course, in terms of financial balance, it would make sense that one that a transformer with lots of VA is matched with an amp that draws lots of current and a massive PSU with lots of capacitance.
An analogy: In the same way, one would not want to pull a small trailer with 1000 horsepower truck.
However, the trailer design is not affected and it does not hurt the performance of the trailer.
thank you for your response!
My understanding is that a high inrush current would sometimes require an inrush current limiter and that this in turn can affect the dynamic output of the amplifier aversely.
Therefore too much capacitance is not necessary good. I remember reading a post from 6L6 that he thought 18.000uF? capacitance in may be the sweetspot for the Aleph J. I guess this is because of the trade off between the benefits of additional capacitance and the averse effect of an inrush current limiter.
However, if I got a toroidal transformer with lots of VA, I do not necessarily need to build a larger capacitor bank, right?
Of course, in terms of financial balance, it would make sense that one that a transformer with lots of VA is matched with an amp that draws lots of current and a massive PSU with lots of capacitance.
An analogy: In the same way, one would not want to pull a small trailer with 1000 horsepower truck.
However, the trailer design is not affected and it does not hurt the performance of the trailer.
The inrush current limiters are indeed limiting the sound quality quite severely. All your observations are spot-on. It is possible to go with a lower capacitor bank and try to omit the CL60 or similar thermistor entirely. Many DIY-ers have done it. There are many solutions that use relay bypass, after the initial charge done through a resistor. With a careful choice of the relay (multiple silvered contacts), the hit on the sound quality should be minimal.
Apart from optimising the power supply with slightly larger VA transformer rating and lower capacitor bank to try to avoid inrush current limiters, there are many other ways to improve the sound of this lovely amplifier.
Use balanced inputs (this will immensely shorten the ground loop that has a potential to pick the buzz/hum), careful placement of the PCB's and transformer to minimise the length of the critical wiring, use short and thick wires for power and speaker connections. You could also omit the 1uF capacitor from the -IN; this will bring major sound improvement. Also, remove the short circuit current limiter transistor. You'll end up owning an amplifier that can go head to head with anything out there, cost no object, because of its minimalist approach in design (only two gain stages, with output working in pure class A, single-ended config, with no transformers nor capacitors in the signal path), and the minimalist approach in its execution when it comes to PCB placement and the choice of parts, wire and the short signal lengths.
However, you'd want to be willing to accept few risks (no short circuit protection, DC amplification)... I always valued the minimalist approach and ultimate sound reproduction quality over anything else, but others may not. The risk might be too high.
Apart from optimising the power supply with slightly larger VA transformer rating and lower capacitor bank to try to avoid inrush current limiters, there are many other ways to improve the sound of this lovely amplifier.
Use balanced inputs (this will immensely shorten the ground loop that has a potential to pick the buzz/hum), careful placement of the PCB's and transformer to minimise the length of the critical wiring, use short and thick wires for power and speaker connections. You could also omit the 1uF capacitor from the -IN; this will bring major sound improvement. Also, remove the short circuit current limiter transistor. You'll end up owning an amplifier that can go head to head with anything out there, cost no object, because of its minimalist approach in design (only two gain stages, with output working in pure class A, single-ended config, with no transformers nor capacitors in the signal path
), and the minimalist approach in its execution when it comes to PCB placement and the choice of parts, wire and the short signal lengths.However, you'd want to be willing to accept few risks (no short circuit protection, DC amplification)... I always valued the minimalist approach and ultimate sound reproduction quality over anything else, but others may not. The risk might be too high.
Last edited: