F5 power amplifier

Jump to Latest
#9,801
6L6 said:
Which leads to the question - what would be the ideal choice for cap in that position? Also, would there be any benefit to having that pair of caps (the + and - rail) connected at the amp pcb's rail inputs? I.E., having it mounted physically as close as possible?
Click to expand...
the LF and HF decoupling are on board to meet the very fastest of audio transient current demands.
The main PSU smoothing is for VLF current demand and extra cabling inductance for a few inches of twisted 3core hardly makes any difference.
 
#9,802
It can't be for 500ms !
The length of a whole half wave of mains power is either 10ms or 8.3ms.
The longest pulse while trying to charge the first C from the transformer is very roughly half of those time intervals, say 4 to 5ms.
Click to expand...

Using PSUD, (and soft start) i can see a peak reached at 500ms (when charging 47mF from a 18v transfo) it is this peak (about twice the established ripple) i am speaking about

Edit:
It looks like the soft start introduces this peak
 
Last edited:
#9,805
I thought I should follow my own advice and check Tinitus link he so kindly looked up for us.

I can't see any value specified for peak one shot current. There is not a defined symbol for that parameter.

T,
where does 13.5A comes from for ripple current?
I see 6.5A or 9.5A ripple capacity and when corrected for 100Hz this drops to ~4A.

On that basis 4//3900uF would give ~16A or 24A of continuous ripple capability.
 
Last edited:
#9,806
AndrewT said:
I see 6.5A or 9.5A ripple capacity and when corrected for 100Hz this drops to ~4A.

On that basis 4//3900uF would give ~16A or 24A of continuous ripple capability.
Click to expand...

ah, yes, too good to be true
appears to be different when going to manufactors spec sheet insted of digikey's
but still a tough little cap
but small, and too expencive then

thanks for correcting Andrew
I dont know why they have specced current at 10khz, when they do 100hz for others

gosh, they have some rated at 150gr, and 500.000 hours at 75gr
 
#9,809
AndrewT said:
where does 13.5A comes from for ripple current?
I see 6.5A or 9.5A ripple capacity and when corrected for 100Hz this drops to ~4A.
Click to expand...

hey Andrew, Im not arguing, but trying to understand

and theres an even tougher 55V, 17.5A, 125gr, though taken at 105gr
Digi-Key - 495-3701-ND (Manufacturer - B41605A0368M002)

but its a bit messey as this one isnt in the spec sheet either
but theres a similar one with same can size, but slightly more capacity

do I understand the frequency/current chart correctly, if I say that the 40V ø25mm can size derates with a 0.7 factor, and the 55V ø25x50mm derates with a 0.6 factor

btw, could you by mistake have taken the current rating at 125gr
the listed current seems to be at the 105gr rating, which is higher
man, that kind of heat is like steaming hot water

well, enough from me on that subject
 
#9,811
Lerg said:
looks to be a bit excessive though.
Click to expand...

yeah, but as of now I have decided not waste more time, and not worry about it :dead:
"simple and easy" mostly works 🙂
 

Attachments

  • F5 simple cap TOWER.JPG
    F5 simple cap TOWER.JPG
    22.6 KB · Views: 626
#9,813
I and others have been saying for a long time that the first cap in a CRC and CLC must have a very high ripple rating.

I have recommended that a bank of paralleled cheap caps are the most economic way of providing that ripple capability. The smoothing of rectified AC to DC is a continuous duty. In a high bias amplifier (eg ClassA) that is a high ripple current and continuous duty. The caps will heat up. Keep them in a cool environment, probably at the bottom of a well ventilated amplifier. Design to meet the duty.
 
#9,816
building your own super cell caps 🙄

it should be possible to stack multiple small caps, and cast them into solid blocks using a 2-component molding silicone, or whatever its called

might be very practical "building-blocks", and easy to handle

and as a positive side effect, I think such silicone rubber is resistant to heat
and ofcourse its electrically isolating
 
You must log in or register to reply here.
Top Bottom