F5 power amplifier

Jump to Latest
#4,921
To minimise that, either twist the wires or use a shielded cable or both.
I am initially going to try just twisting wires.
I use 4mm^2 wire then twist the pairs together and then put a couple of layers of heat shrink tubing over them.

What ever is left over will be taken care of by the second capacitor bank in the amp.
 
#4,922
While I work on getting more suitable heatsinks, a quick Q: I notice that the tech-diy parts kit uses 22K for R21/22 instead of 10K. What is the reason for this, and what changes can be expected from the protection circuitry as a result of this?
 
#4,926
thanh1973 said:
I use 4mm^2 wire then twist the pairs together
Click to expand...
1sqmm will pass 3A continuously. Since there is a pair carrying the +ve & -ve supplies that equates to 6A continuous at the output.
4sqmm has a capability of ~24A continuous and probably five times that for transients.
What kind of amplifier requires that current carrying ability in the wire between the two banks of capacitors?

CRC is effective if the resistance between the two banks of caps is not zero.
What's the point in using 4sqmm and then adding an extra resistor to create the R of the CRC?

0.5sqmm would probably do for a big ClassAB amplifier and 0.6mm diameter (0.3sqmm) for a small ClassAB amplifier.
 
Last edited:
#4,928
Correct me if I am wrong, Noob here.
I thought current carrying ability is not the same as the R in CRC.
The cable are used for carrying the current during transient and normal activities. While the R is used for filtering purposes isnt it? No?

If the current carrying ability is not good enough, wouldnt the amplifier sound slow/muddy? Worst case, the wire burn up?

So to my understanding, using 4sqmm wire is meant for current and R is for filtering. Am I correct to think this way?
 
#4,929
fff0 said:
Correct me if I am wrong, Noob here.
I thought current carrying ability is not the same as the R in CRC.
The cable are used for carrying the current during transient and normal activities. While the R is used for filtering purposes isnt it? No?

If the current carrying ability is not good enough, wouldnt the amplifier sound slow/muddy? Worst case, the wire burn up?

So to my understanding, using 4sqmm wire is meant for current and R is for filtering. Am I correct to think this way?
Click to expand...

I see your point, but remember that a low ohm resistor is usually just a small wire inside some ceramics. A longer wire do however have more self-inductance than a short wire, but in this case we actually don't mind the features of an inductor. 😀
 
#4,930
Andrews comments, were based on using a separately cased power supply, with a bank of caps in the power supply and then another bank of caps in the amp case.
His idea was incorporating resistance in the umbilical power cable between the two cases.
Which I also agree is a good idea if you want to use a CRC powersupply.
At this point in time I am trying to avoid a CRC setup.
 
#4,931
R value in CRC power supply

Hello,
How does one determine the value of R in a CRC power supply for the F5 or the Aleph J? I've seen PS schematics with 4 .47 ohms resistors in parallel for an effective value of .117 ohms and in another 4 1 ohms in parallel for a value of .25 ohms. So what's the acceptable range of R value that can be used in a F5 or Aleph J power supply? Thanks,
David
 
#4,933
TerribleT said:
Hello,
How does one determine the value of R in a CRC power supply for the F5 or the Aleph J? I've seen PS schematics with 4 .47 ohms resistors in parallel for an effective value of .117 ohms and in another 4 1 ohms in parallel for a value of .25 ohms. So what's the acceptable range of R value that can be used in a F5 or Aleph J power supply? Thanks,
David
Click to expand...
one way to help determine the resistor value is to look at the frequency response of the filter that is created.
The R + C form a low pass filter that attenuates the higher frequencies.

If you have adopted +-68mF as the second stage capacitance and use 1r0//1r0//1r0//1r0 then the filter F-3dB = 1/2/Pi/0.068/0.25=9.3Hz.
The filter will reduce 9Hz by ~3dB relative to the DC that will pass straight through.
at 1 octave above 9Hz i.e. 18Hz the response will be ~9dB down and a further octave up, i.e. 36Hz it will be ~15dB down. @ 50Hz (mains frequency) it will be ~18dB down (nearly 1/10 or one tenth) lower than if the resistor was not there. A really thick wire connection before the 68mF capacitor will not filter as well as a thin wire with some significant resiatance.

You may have other criteria for choosing a resistance value.
 
#4,934
Anyone compared CRC vs. CLC soundwise?

Yes. As an experiment, I used the primarys of a 330va transformer (I am dual mono, one of these for each amp) I had laying around. CLC common mode, 1 set windings per rail. Rail ripple went from 400mv (4x.47R CRC) to 20 mv. At 2.4 ohm DCR, the rail voltages went to about 17.5. The sound that came out of this amp was not to be believed (by me), I was knocked out. I am in the process of trying the same thing (with much lower values) isolating the jfet source.
I don't know the actual L values I am using, sorry. I just ordered a meter yesterday
 
#4,937
Would some of those inductors pulled from PC power supplies or similar work for that application?


This kind of thing:

FIT50-7.jpg


or

An externally hosted image should be here but it was not working when we last tested it.




I take it these are too low inductance?

Fran
 
You must log in or register to reply here.
Top Bottom