Power Supply Resevoir Size

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
This is a fantasy that a psu with transformer, drop only 1-5V. sorry for this. if you have measure, please show.
It seems from your various posts that you have gone down the regulated PSU route. It also appears that you have rejected all forms of linear PSUs as being suitable for audio power amplifiers.

If you have never seen Vdrops of 1V to 5V in a linear, unregulated PSU when driving full power into the specified load, then you have not investigated thoroughly.
They exist. I have some.
You obviously don't have any.

It further seems that if any Member dares to disagree with your SMPS "rules" view, that they are all wrong.
 
Last edited:
Originally Posted by AP2
This is a fantasy that a psu with transformer, drop only 1-5V. sorry for this. if you have measure, please show.

i am with AP2....in my super leach amp, i used a power traffo with 2inch center leg stacked 3inches....very heavy transformer....

full power testing with both channels putting out 45 volts ac into an 8 ohm load showed that standby rails of +/-85volts dropping to +/-72 volts.....

AmpTop-1.jpg
 
i am with AP2....in my super leach amp, i used a power traffo with 2inch center leg stacked 3inches....very heavy transformer....

full power testing with both channels putting out 45 volts ac into an 8 ohm load showed that standby rails of +/-85volts dropping to +/-72 volts.....

AmpTop-1.jpg

Hi, thank Tony...nice job. this your drop voltage 13V is familiar for me and reflect my last pic also.
AndrewT: no..i not have rejected linear psu (as concept),infact i have develop ultra fast regulator for it.
i reject unregulated, independant it is smps or linear.
Market push on new solution, then some companies invest in this direction.
at 50Hz i have calculate a hole of energy(min)=8ms.(just for acceptable 5V of drop under load,if we fill this hole of energy)
show me your solution (exclude big inductance for shift phase)
 
Last edited:
It further seems that if any Member dares to disagree with your SMPS "rules" view, that they are all wrong.

I'm sorry, I do not agree with what you say. I "accept disagreements about my work), they are clearly documented. I do not accept lies, especially if they come from incompetent people.
I do not force a smps for lamps into audiophile use of amplifier.
I do not hiden emi/rfi becouse are very poor.
 
Theoretically, or simple model if you prefer, 30mF will deplete ~33.3mV/ms/Arms.
For a 3.5Arms load (98W into 8r0) and a duration of 9ms, the smoothing cap will deplete by ~1V

Then you need to make a more sophisticated model to better reflect actual PSU performance.

Instead I simply measure the PSU voltage. 1V to 5V is pretty common for my unregulated linear PSUs.
As I said earlier
you have not investigated thoroughly.
or you too would have found similar by actual measurement.
 
Theoretically, or simple model if you prefer, 30mF will deplete ~33.3mV/ms/Arms.
For a 3.5Arms load (98W into 8r0) and a duration of 9ms, the smoothing cap will deplete by ~1V

Then you need to make a more sophisticated model to better reflect actual PSU performance.

Instead I simply measure the PSU voltage. 1V to 5V is pretty common for my unregulated linear PSUs.
As I said earlieror you too would have found similar by actual measurement.

I'm sorry, in this thread, there are people who can understand very well,a complex model, I have no doubt about it.
If you have this psu done well, is very easy to show the behavior during the peaks and the continuous absorption. maybe even a picture, just to see how big it is.
----------------
How You write "instead of simple measure volts?"
my last pic, can give you a lot of information also in ref to timing, perhaps you've only read the voltage?

and, one big capacitors, can help with the first impulse of mainstream (or single repeated in agreemet with period of recharge), after this, I want to see the charging time, especially in repetition (as is the music signal)
 
Last edited:
I am definitely not any expert or even close but it now appears that we are getting into an argument about two opposing trains of thought here. I see the latest picture of the power amp with what looks like a single large power supply capacitor. This seems to be going against what I thought that I had been following of smaller parallel capacitor banks,whether all together, in a CRC configuration or in a distributed large capacitor, smaller capacitor at the output device location. It would appear that this simple question could be answered with a simple test. One amplifier as a control with a fixed transformer size, large enough to handle the power demands. Then using one manufacturers capacitors with identical construction methods, make one experiment with a single large capacitor and another with multiple smaller capacitors of the exact same total capacitance. Select a time frame that is sufficient to show the results over time with a high demand and show the scope patterns so we can compare apples to apples. Just what I would like to see with only one variable changing in a scientific test.

Steven
 
This has been by far one of the most interesting threads I have followed. I think it would imply that there is still no consensus reached or a simple rule of thumb found. :)

Steven, your request is very valid and I do not think that any of us have amplifiers at the ready to make power supply alterations to without major effort and for that reason I would like to call on Harrison in this regard.

Harrison has a few threads running where members are building his amp designs with his freely supplied PCBs. (I commend you for that):worship:

I would ask of Harrison to request from the assemblers that is currently busy building his amps and have access to a scope to make some measurement with the aim to confirm the fact that a few smaller caps are more desirable than one larger cap. I think this will put that question to bed.:xfingers:

We can then focus on the next possibility and so eliminate the variables practically one by one on the recommendations from our esteemed DIY members that has been contributing so unselfishly to this thread.

I also think that Harrison's group will benefit from this exercise.:up:
 
Last edited:
I am humbled Nico :) . This being diy its a great place to meet, make new friends learn, research, share and have fun. You must have come across this :) Free PCBs

We just have to match our theory to practical implementation. A lot has been said about PCB layout in this thread and any member is free to offer what they think is the best layout for the designs.

AP thanks for sharing with us your amp. As you can see there is a big disjoint between theory and practical implementation. What would you say is the inductance of those interconnect cables, is the amp worse off ? AP also seems fine with film across the big boys. Does the amp display unruly behaviour ? What do the ears hear ?
 
I am humbled Nico :) . This being diy its a great place to meet, make new friends learn, research, share and have fun. You must have come across this :) Free PCBs

We just have to match our theory to practical implementation. A lot has been said about PCB layout in this thread and any member is free to offer what they think is the best layout for the designs.

AP thanks for sharing with us your amp. As you can see there is a big disjoint between theory and practical implementation. What would you say is the inductance of those interconnect cables, is the amp worse off ? AP also seems fine with film across the big boys. Does the amp display unruly behaviour ? What do the ears hear ?

??..may be you ref to amp that post Tony.
 
Harrison, since all your follower builders are using identical designs your PCB therefor amp design and pcb layout are not variables but constants and thus eliminated from the equation. It is not a test for your amp at all, just what can be concluded regarding multiple caps on power supply. Okay we will have to take power transformer into consideration.

Maybe Tom, Terry, AP, Frank, DF96, Daniel, AndrewT, et all can guide us what to measure and post it here for discussion of what we see and how it relates to the simulations.

So far we are speculating, theorizing and simulating, a few practical measurements would go a long way especially in substantiating some theories since the amps and layouts are identical and I doubt if we will ever have such an opportunity again.

Thanks in advance Harrison and those DIY members in the process of building SYMEF and 1DIFQC who is prepared and willing to contribute to this thread. I stand to be corrected but any findings here will take some guessing out of the power supply design of future amps.
 
Last edited:
Apologies AP :) . Thanks Tony for sharing your amp ;) .

thanks, the amp was built in the 80's, it uses computer grade 3inch can type cap with 56000ufd/100vdc caps bypassed with rifa 2.2ufd/250volt stacked film cap....this was my last super leach amp.....

it will interest you to know that the traffo i built for this amp used #16 primary and #13 secondary wires....
 
Hi, thank Tony...nice job. this your drop voltage 13V is familiar for me and reflect my last pic also.
AndrewT: no..i not have rejected linear psu (as concept),infact i have develop ultra fast regulator for it.
i reject unregulated, independant it is smps or linear.
Market push on new solution, then some companies invest in this direction.
at 50Hz i have calculate a hole of energy(min)=8ms.(just for acceptable 5V of drop under load,if we fill this hole of energy)
show me your solution (exclude big inductance for shift phase)

the 13v rails drop was for full power sine wave testing....i suspect that with music, the drop will be smaller since music has less energy demands on the psu in comparison....
 
Thanks Tony. That was a big film at 2.2ufd ;) you may not have encountered any problems. Just for curiosity what is your current amp ? :cheerful:

it's an EL34 pp amp....:D i am comfortable with both tubes and ss amps....

this super leach amp was well liked by those who heard this amp play....i don't usually tell others what to expect, i just wait for their comments.....and the leach super amp is indeed a winner....i am going to rebuild this 20 year old amp soon...
 
Nico,
I am one of the people who is building the 1DIFF amplifiers. I must say that what I can already see is that some people will be modifying the board in a myriad of ways. Different biasing settings and changes in output devices are already appearing it seems. I am trying to build as per the original design unless there is a reason to change what is proposed. I am installing standoffs so I could potentially change the output devices without having to De-solder the devices from the board. I have been watching this thread thinking that the final power supply will have a major impact on the final result to the amplifier. If I can make more than one power supply with the simple test that I proposed I will see if I can do that. I am just not one of the people who will build this amplifier in a couple of days and have the answers fast enough I think. Perhaps one of the people who is a consummate DIY builder and designer could answer this question for all of us.

Steven
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.