A nice and solid star GND in the amp is certainly good to avoid hum / ground loops within the amp. But I don't see how it would break the loop that is formed by the external wires and their GND connections on both sides (at the source and at the amp):
Source GND ---> left channel cable GND ---> power amp GND ---> right channel cable GND ---> Source GND
A common approach to break this loop is to insert resistors between the GND connections of the left and the right cables. See here (section "The Cross Channel Ground Loop Remedy").
It's probably better to arrange as CRCC, to keep losses to a minimum. This is a power amp after all...
(6) 0.47 3W resistors in parallel (as in the original PSU) is roughly an 18W 0.1ohm resistor.
You could search for something like that, or just parallel a bunch of the 3W resistors, which is a lot easier.
(6) 0.47 3W resistors in parallel (as in the original PSU) is roughly an 18W 0.1ohm resistor.
You could search for something like that, or just parallel a bunch of the 3W resistors, which is a lot easier.
make R of the rCRC small.
Try 5 off 1r0 in parallel for an effective 0r2.
If you use the standard 600mW resistors then you have a continuous current rating of 3.87A
That's a lot of bias for any ClassA amplifier.
Make the final C big enough to supply all the LF transients.
The first r is the sum of all the resistances in the charging circuit, (transformer+rectifier+all the wiring).
Since you are proposing 3 sets of capacitors you have a choice of using a three stage filter rCRCRC or what I would recommend is using a two stage rCRCC where the doubled CC supplies all the LF transients demanded by the speaker.
But do check you have adequate ripple capability in the first C, otherwise you could overheat it and severely shorten it's life.
Use PSUD2 to check the ripple on that first capacitor.
A cheap way to buy extra ripple capability is to install many paralleled capacitors in that first location, eg. 5off 2m2F instead of 10mF or 10off 4m7F instead of 47mF, if that's what PSUD2 suggests.
Try 5 off 1r0 in parallel for an effective 0r2.
If you use the standard 600mW resistors then you have a continuous current rating of 3.87A
That's a lot of bias for any ClassA amplifier.
Make the final C big enough to supply all the LF transients.
The first r is the sum of all the resistances in the charging circuit, (transformer+rectifier+all the wiring).
Since you are proposing 3 sets of capacitors you have a choice of using a three stage filter rCRCRC or what I would recommend is using a two stage rCRCC where the doubled CC supplies all the LF transients demanded by the speaker.
But do check you have adequate ripple capability in the first C, otherwise you could overheat it and severely shorten it's life.
Use PSUD2 to check the ripple on that first capacitor.
A cheap way to buy extra ripple capability is to install many paralleled capacitors in that first location, eg. 5off 2m2F instead of 10mF or 10off 4m7F instead of 47mF, if that's what PSUD2 suggests.
Last edited:
A few moments of searching came up with this data sheet... http://www.chemi-con.com/upload/files/8/1/1061909384505b66a161392.pdf
...and it shows that your caps are rated for a breathtaking 27A rms ripple capacity.
Build happily away! Those will work beautifully for this PSU.
...and it shows that your caps are rated for a breathtaking 27A rms ripple capacity.
Build happily away! Those will work beautifully for this PSU.
good morning to all! have a question about heat radiator for F5 (not F5T). Have an idea to use pair of these per channel.
182x45x70 mm S>=1300 cm2
would be enough?
182x45x70 mm S>=1300 cm2
would be enough?
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Thank you. So maybe four of these per channel (8x per amp) must be good for F5?