I have come across various PSU and noticed that the size of caps for SS amp is usually in the reagion of 10,000uf+ while the size of caps for B+ in output tube is in the order of 1000uF+. For tube that use output transformer, the current is low and this may explain it. But for OTL tube amp with a peak current of a few A, is it worthwhile to use higher capacitence?
Or would lower capacitance actually keept he amp faster?
Or would lower capacitance actually keept he amp faster?
I would expect to use the same capacitance in an OTL amp as in a SS amp as the currents are exactly the same. However there is a quite large variation in capacitance among SS amplifiers ranging from as low as 4700uF per rail up to >100000 uF in others, (if we consider most amplifiers which are class AB). That an ordinary transformer coupled amplifier has much lower capacitance is to be expected but also here you see quite a larger variation from classic amplifiers with as low as 8uF, (but with CLC) up to >1000uF which I think is unnecessary high.
In my own OTL I use 5500uF per rail for each channel and in amplifiers I have sold I have used 10000uF per rail for each channel, I don't see any reason to go higher than that. There is a difference in power supply rejection between SS push-pull amps and some OTL amps where the OTL amps are much worse, this can be solved in various ways but it is un-necessary to use larger power supply capacitors.
You have to explain to me the relation between capacitor size and amplifier "fastness", (which I assume is transient response) as I can not see it.
Regards Hans
In my own OTL I use 5500uF per rail for each channel and in amplifiers I have sold I have used 10000uF per rail for each channel, I don't see any reason to go higher than that. There is a difference in power supply rejection between SS push-pull amps and some OTL amps where the OTL amps are much worse, this can be solved in various ways but it is un-necessary to use larger power supply capacitors.
You have to explain to me the relation between capacitor size and amplifier "fastness", (which I assume is transient response) as I can not see it.
Regards Hans
Thanks for your kind reply. I am not sure if this is correct.
The transient response is related to how fast the PSU can deliver the neccessary current at the given voltage. This is important in the bass region when there is a large current drawn. Large capacitance is associated with the use of large caps which tends to charge/discharge slower than normal caps. So, when there are big swings in current demand, the large caps may not be able to catch up with the variations in demand, resulting in "slow" sound.
The transient response is related to how fast the PSU can deliver the neccessary current at the given voltage. This is important in the bass region when there is a large current drawn. Large capacitance is associated with the use of large caps which tends to charge/discharge slower than normal caps. So, when there are big swings in current demand, the large caps may not be able to catch up with the variations in demand, resulting in "slow" sound.
Actually my experience has pretty much been the reverse in solid state amplifiers, to a point larger caps should result in better regulation under short term transient conditions and the lower overall supply impedance (in amplifiers with modest psrr or feedback margin) subjectively results in faster, tighter bass.
Note however that the key here is to assure that both esr and esl are low in the supply caps chosen, and it may be better to use a number of smaller caps in parallel to achieve the overall desired capacitance.
Low impedance wiring techniques are a must as are low impedance/low resistance power transformers, (toroids or high quality ei types) and high current rectifiers.
There is a point of diminishing returns, and increasing capacitance just to reduce ripple on the supply rails can be counterproductive. Air core speaker cross-over inductors can make good chokes for high current pi filter based unregulated supplies in ss and otl applications.
As always YMMV.
Note however that the key here is to assure that both esr and esl are low in the supply caps chosen, and it may be better to use a number of smaller caps in parallel to achieve the overall desired capacitance.
Low impedance wiring techniques are a must as are low impedance/low resistance power transformers, (toroids or high quality ei types) and high current rectifiers.
There is a point of diminishing returns, and increasing capacitance just to reduce ripple on the supply rails can be counterproductive. Air core speaker cross-over inductors can make good chokes for high current pi filter based unregulated supplies in ss and otl applications.
As always YMMV.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.