reducing the resistances to zero ohms, has the effect of increasing the Q of every capacitor and every inductor.How about using superconductors for the interconnects?
That leads to a multiplicity of perfect oscillators.
It's the damping effect of the resistances that damps the Qs of all the components and that in turn damps or even eliminates the ringing when step changes in current come along to kick the circuit into responding.
Each link? Did you leave one or more out or do you mean links within the thread you linked?
Gents
I am building the power supply for my mono F5s using Civiller power supply boards version 1.0-2000-12-14,and eight Kendeil K05 105 c, 10,000uf, 40v capacitors.
I also have available some BHC ALS30A473NP100, 47,000uf, 100v, Ripple currrent 25200mA.
Are the BHC's considered a useful addition in the above F5 power supply. If so, in what location, and would this require a change of value to R9 and R10.
Within. The link actually goes to a single post and there is a set of links at the end of that post, in an attempt to save you from a lot of extraneous material.
It's worth reading.
James,
Electro caps are a bit like small planes (or cars, for that matter) - each one has a different characteristic and reliability and is built to do a different job, and so forth - the electrolytic caps are similar and they all have a different sound characteristic and the specs don't show any of this at all, unfortunately.
Now, the "sound" of the Kendeils K05s are quite different to the BHCs, regardless of the different sizes - the Kendeils will produce a noticeable heavier bass and mellow mids where the BHCs will give you a much more forward mids/tops that'll sound a bit "thin" in the bass, but are a lot "faster" (please excuse the terms) - a totally different sound my opinion here - perhaps using 94mF/rail of the 100volt BHCs will alter this, but not by much [the Rifa peh 169/200 caps are perhaps the "fastest", most dynamic of all the production electro caps, but can be tricky to "get right"].
So, as you have both types, simply allow room to try both types out - I'd suggest the K05s next to the diodes, and the BHCs as the power caps to begin with - a mixture.
I'd also suggest you dispense with the pcb and wire the caps up directly via top quality, high current solid wire, not pcb tracks, even tho they're quite thick - it makes a quite significant difference.
This amp can be altered/optomised to a surprising degree!
Electro caps are a bit like small planes (or cars, for that matter) - each one has a different characteristic and reliability and is built to do a different job, and so forth - the electrolytic caps are similar and they all have a different sound characteristic and the specs don't show any of this at all, unfortunately.
Now, the "sound" of the Kendeils K05s are quite different to the BHCs, regardless of the different sizes - the Kendeils will produce a noticeable heavier bass and mellow mids where the BHCs will give you a much more forward mids/tops that'll sound a bit "thin" in the bass, but are a lot "faster" (please excuse the terms) - a totally different sound my opinion here - perhaps using 94mF/rail of the 100volt BHCs will alter this, but not by much [the Rifa peh 169/200 caps are perhaps the "fastest", most dynamic of all the production electro caps, but can be tricky to "get right"].
So, as you have both types, simply allow room to try both types out - I'd suggest the K05s next to the diodes, and the BHCs as the power caps to begin with - a mixture.
I'd also suggest you dispense with the pcb and wire the caps up directly via top quality, high current solid wire, not pcb tracks, even tho they're quite thick - it makes a quite significant difference.
This amp can be altered/optomised to a surprising degree!
Can you elaborate on the tweaking tuning possibilities?
Beyond increasing rail voltages, what else is out there to optimize the f5? (I am limited to the 25v mundorf caps in my psu, so can't further increase there)
Ah, with an equiptment lineup like you have, it's clear that you're into detail without fatigue in a big way! The 1541A-S2 in NOS?
Okay, lets start at the supply first - assume you have a good line filter like a Schaffner, etc and a "reasonable quality" power cord.
The transformers - good quality torroids (or Ei) like 300VA are the first item, fed thru a dc trap after a CL-60, as usual. [Single primary but dual secondaries/channel]
Diodes on the bridge [MSR1560G or BYV32 soft recovery Shottky]
The Mundorf caps are a fixture but can add a low drop out Cmultiplier like the one that Patrick is using in the F5X project - cost you a couple of volts but either upgrade transformer or try it as it is - very worth while addition.
UPOCC copper wire (Ohno crystal copper) - short leads to amp.
Add snubbers after the diode bridge (R+C//C "Hagerman", etc network) - plenty of discussions about this here.
An old idea, forgotten these days, but it's called "tuning the transformer winding" and you'll have to chase this one up yourself at "Promethius Audio". I think "jackinj" can add more here, but this 'tweak' can have startling changes, but is a real PIA to get right....
Okay, that's enough for the power supply for now.
Amp - okay, first is the power fets - the IRPFs, Fairchilds, Toshiba k1530/j201, or the triple k2013/j313 of Juma's. - the Toshibas would be your best bet for your system.
Now, the trickie bits - the resistors, a suggested lineup, or mixture - unfortunately not cheap.
R3,4 with Rhopoint 8G16 'econister'
R13,14 (the gate stoppers) Neohms (Zfoils, an alternative)
R11,12 (0.47R, 3/10W Isabellemnhutte PBH source resistors
R5,6 (feedback) Caddock MP930 (or possible VPR 220, power Zfoil from Group buy here)
R1,2 (jfet 10R) TRW or Caddock MP915
Extras like Mundorf Silver/gold wire (in cotton, if you prefer!), silver phono plugs, dc protection on o/p, some exotic o/p terminals, silver wire replacement for copper tracks on pcb, and so on.
You can pick an' match to your hearts content, and this is without changing your Mundorf power supply caps!
There was a recent upgrade for the Aikido by the now "banned" Bob Pragnell (Beau 2314?) via John Broskie - well recommended. Like the Tempo 2SE's.
Perhaps this might be of interest....
Okay, lets start at the supply first - assume you have a good line filter like a Schaffner, etc and a "reasonable quality" power cord.
The transformers - good quality torroids (or Ei) like 300VA are the first item, fed thru a dc trap after a CL-60, as usual. [Single primary but dual secondaries/channel]
Diodes on the bridge [MSR1560G or BYV32 soft recovery Shottky]
The Mundorf caps are a fixture but can add a low drop out Cmultiplier like the one that Patrick is using in the F5X project - cost you a couple of volts but either upgrade transformer or try it as it is - very worth while addition.
UPOCC copper wire (Ohno crystal copper) - short leads to amp.
Add snubbers after the diode bridge (R+C//C "Hagerman", etc network) - plenty of discussions about this here.
An old idea, forgotten these days, but it's called "tuning the transformer winding" and you'll have to chase this one up yourself at "Promethius Audio". I think "jackinj" can add more here, but this 'tweak' can have startling changes, but is a real PIA to get right....
Okay, that's enough for the power supply for now.
Amp - okay, first is the power fets - the IRPFs, Fairchilds, Toshiba k1530/j201, or the triple k2013/j313 of Juma's. - the Toshibas would be your best bet for your system.
Now, the trickie bits - the resistors, a suggested lineup, or mixture - unfortunately not cheap.
R3,4 with Rhopoint 8G16 'econister'
R13,14 (the gate stoppers) Neohms (Zfoils, an alternative)
R11,12 (0.47R, 3/10W Isabellemnhutte PBH source resistors
R5,6 (feedback) Caddock MP930 (or possible VPR 220, power Zfoil from Group buy here)
R1,2 (jfet 10R) TRW or Caddock MP915
Extras like Mundorf Silver/gold wire (in cotton, if you prefer!), silver phono plugs, dc protection on o/p, some exotic o/p terminals, silver wire replacement for copper tracks on pcb, and so on.
You can pick an' match to your hearts content, and this is without changing your Mundorf power supply caps!
There was a recent upgrade for the Aikido by the now "banned" Bob Pragnell (Beau 2314?) via John Broskie - well recommended. Like the Tempo 2SE's.
Perhaps this might be of interest....
thanks james. Indeed 1541aS2 in Nos mode...
i did build in a schaffner filter, but left it unconnected so far. Did you find an improvement using it?
got it... Single transformer though... No dual mono...
[quoteDiodes on the bridge [MSR1560G or BYV32 soft recovery Shottky] [/quote]
yes, think i am using mur1560's indeed
The Mundorf caps are a fixture but can add a low drop out
thanks! Good idea
i don't use snubbers... may be a good addition. Do you hear significant differences?
will have to look that one up...
mmmm i am on 'stock' irfp's. Where can i find info on the different sonic attributes of the alternative options?
wow yeah.... That's a full recipe... I am using a mix of prp, dale and vishay naked resistors now, could look into some of the options you present...
definately interesting! Thanks!
I also need to look into that aikido tweak....
Aah so much still to explore!
Ah, as with 'most tweaking', some of these things also go backwards and using line filters can sometimes be "off", so it's got a bipass switch - the amp 'goes travelling' a bit and other people's mains power isn't always clean.
I have an old industrial isolation transformer in the system mains here with it's secondary centre tap to earth (so-called balanced power?) and this is quite useful for noise reduction - not much of a problem here anyway, but steadily getting worse.
I'm not convinced that separate transformers are always better, but using 2 transformers doubles the amp's capacity, halves the .... etc and this will produce changes but with all things audio, sometimes not what you wanted at all.
I have a 400va EI transformer built here (huge thing, many $s!) and it's got the extra taps on the separate sec windings at 18 & 20Vac (for using the Cmultiplier) and this is clearly a superior supply - had to bolt it to front plate - added screens, etc - monster block of iron! Completely 'over the top' but just going to extra distance to see what is possible ...
MSRs are slightly different to the MURs and so on - wouldn't get too concerned as small difference - just being a bit 'anal' here!
Not the same with the snubbers - there IS a noticeable difference but you do have to get them right, if you're wiling to go to the trouble - it's just another small thing but adding these small things up is what "tweaking" is all aboiut IMO.
And, you teach yourself to "listen better" and will cary this over into other sections like room acoustical improvements, cables, etc - an attitude thing that unfortunately, can become an obsession! Ah, I think you have caught it as well!
Just out of curiosity, try adding a Nichicon ES, KG or KZ (bipolar or superthru) across the the power lines near to the Fets on your amp board, and run separate 0volt wires back to your centre earth - BIG surprise if you haven't tried this one, particularly with the Mundorf's!
If there isn't room, try it across the power caps, the closest cap to the amp.
The differences beteen the different Fets is buried in the main thread awhile back - basically (IMO, again here) the IRFs are the basic ones, the Fairchilds have a slightly 'lighter' sound, and the Toshibas have more detail but slightly lower gain - Patrick did quite a bit of work in sorting out the way you alter the jfets to cater for the differences (or similarities actually) in transconductances in these power Fets, so it's not just a parts change, unfortunately. John walton is also pretty clued up on this, and a number of other guys have done a lot of work on this - details, hmmm!
The triple k2013/j313 of Juma's is the best of the lot, IMO, but I did this on 'veroboard' with that fancy silver wire for connections and not copper tracks, so could be different too. And biased high at 0.5A each - very nice indeed!
I see you have some Zfoils, possibly for R3,4? This is where the Rhopoints are most noticeable, and at those 10R ones at the jfets - then the Fet gate stoppers, and the the feedback resistors - that's nearly all the resistors in the circuit!
I would try the TRW/IRC replacement for the Dales - 1/2 watt ones
Oh yes, with your benign load of the Tempo's, you could possibly reduce your feedback as per the Turbo design.
There are a couple of us here that're playing with the old Phillips dac chip - with Ian's latest Fifo buffer, Patrick's CEN o/p stage, aikido with Lightspeed, etc - and so on! (my compromise to going back to vinyl!)
I have an old industrial isolation transformer in the system mains here with it's secondary centre tap to earth (so-called balanced power?) and this is quite useful for noise reduction - not much of a problem here anyway, but steadily getting worse.
I'm not convinced that separate transformers are always better, but using 2 transformers doubles the amp's capacity, halves the .... etc and this will produce changes but with all things audio, sometimes not what you wanted at all.
I have a 400va EI transformer built here (huge thing, many $s!) and it's got the extra taps on the separate sec windings at 18 & 20Vac (for using the Cmultiplier) and this is clearly a superior supply - had to bolt it to front plate - added screens, etc - monster block of iron! Completely 'over the top' but just going to extra distance to see what is possible ...
MSRs are slightly different to the MURs and so on - wouldn't get too concerned as small difference - just being a bit 'anal' here!
Not the same with the snubbers - there IS a noticeable difference but you do have to get them right, if you're wiling to go to the trouble - it's just another small thing but adding these small things up is what "tweaking" is all aboiut IMO.
And, you teach yourself to "listen better" and will cary this over into other sections like room acoustical improvements, cables, etc - an attitude thing that unfortunately, can become an obsession! Ah, I think you have caught it as well!
Just out of curiosity, try adding a Nichicon ES, KG or KZ (bipolar or superthru) across the the power lines near to the Fets on your amp board, and run separate 0volt wires back to your centre earth - BIG surprise if you haven't tried this one, particularly with the Mundorf's!
If there isn't room, try it across the power caps, the closest cap to the amp.
The differences beteen the different Fets is buried in the main thread awhile back - basically (IMO, again here) the IRFs are the basic ones, the Fairchilds have a slightly 'lighter' sound, and the Toshibas have more detail but slightly lower gain - Patrick did quite a bit of work in sorting out the way you alter the jfets to cater for the differences (or similarities actually) in transconductances in these power Fets, so it's not just a parts change, unfortunately. John walton is also pretty clued up on this, and a number of other guys have done a lot of work on this - details, hmmm!
The triple k2013/j313 of Juma's is the best of the lot, IMO, but I did this on 'veroboard' with that fancy silver wire for connections and not copper tracks, so could be different too. And biased high at 0.5A each - very nice indeed!
I see you have some Zfoils, possibly for R3,4? This is where the Rhopoints are most noticeable, and at those 10R ones at the jfets - then the Fet gate stoppers, and the the feedback resistors - that's nearly all the resistors in the circuit!
I would try the TRW/IRC replacement for the Dales - 1/2 watt ones
Oh yes, with your benign load of the Tempo's, you could possibly reduce your feedback as per the Turbo design.
There are a couple of us here that're playing with the old Phillips dac chip - with Ian's latest Fifo buffer, Patrick's CEN o/p stage, aikido with Lightspeed, etc - and so on! (my compromise to going back to vinyl!)
F5 psu
Hi there! I only got 22.7 vdc from the psu. Amplimo trafo 625va 2 x 18v, psu boards from Peter Daniels. Is the 22.7vdc enough to power the boards? I alaao paralleled the caps with filmcaps, 20uF, 2.2uF and 0.22uF. Could that be the couse of the drop of Vdc? Thanks for the answer, grtz Teake
Hi there! I only got 22.7 vdc from the psu. Amplimo trafo 625va 2 x 18v, psu boards from Peter Daniels. Is the 22.7vdc enough to power the boards? I alaao paralleled the caps with filmcaps, 20uF, 2.2uF and 0.22uF. Could that be the couse of the drop of Vdc? Thanks for the answer, grtz Teake
F5 DC at output.
Hi guys, wonder if you can help me. Finished f5 build, got it biased nicely and working well for half an hour playing music then I started getting 25v dc from speaker output. Burnt out test speaker. Now across resister 12 I get zero volts on both channels and on speaker output I now have 25 vdc on both.
I was able to bias only by adding two additional 2.2k resistors across r4 on both channels. INitially, When I adjusted the amp warm, I got .59 v across r 12 and r 11 on both channels, but when the amp cooled off, this figure went up to .90 v on both channels, but as the amp warmed up, it went down to the .59 v figure. Is this difference in voltage between a warm amp and a cold amp normal?
Currently, I am getting zero v across resistor 12 and 11 on both channels and 25 volts on both speaker outputs.
Any ideas what could have blown. I looked carefully, resoldered all suspected joints, and nothing looks burnt or damaged on the sight.
Where would you suggest I look?
Hi guys, wonder if you can help me. Finished f5 build, got it biased nicely and working well for half an hour playing music then I started getting 25v dc from speaker output. Burnt out test speaker. Now across resister 12 I get zero volts on both channels and on speaker output I now have 25 vdc on both.
I was able to bias only by adding two additional 2.2k resistors across r4 on both channels. INitially, When I adjusted the amp warm, I got .59 v across r 12 and r 11 on both channels, but when the amp cooled off, this figure went up to .90 v on both channels, but as the amp warmed up, it went down to the .59 v figure. Is this difference in voltage between a warm amp and a cold amp normal?
Currently, I am getting zero v across resistor 12 and 11 on both channels and 25 volts on both speaker outputs.
Any ideas what could have blown. I looked carefully, resoldered all suspected joints, and nothing looks burnt or damaged on the sight.
Where would you suggest I look?
Thanks, one power supply for whole amp, grounded to both channels build following Nelson's F5 pdf writeup.
R11 and 12 .47 ohm checked, and r17 and 18 will check to make sure 1k.
If one Fet problematic would this make DC 25 volts at both channels? Thanks and will try to remove and test if I can get them out without ruining.
R11 and 12 .47 ohm checked, and r17 and 18 will check to make sure 1k.
If one Fet problematic would this make DC 25 volts at both channels? Thanks and will try to remove and test if I can get them out without ruining.
make sure your PSU works as it should. i suspect a bad connection on the negativ rail.
Yes, I think my problem was power supply... which may have caused other damage. power supply on positive rail was fine, but on negative rail was zero. 18 v transformer and one rectifier bridge does not appear to be working and was getting very hot. to test I disconnected capicitor bank and then tested for dc from bridge... when I tested negative rectifier bridge on negative rail it measured 36v for a few moments then to zero on the negative rail so looks intermitten and then positive rectifier gave me 36v.
I believe I made an error in the power supply which manifested itself... I mistakenly hooked positive from the bridge rectifier to negative on the capacitor bank. I attach a pdf of power supply diagram I used, just to make sure this is correct, positive from the bridge rectifier goes to positive side of capicitors on both positive as well as negative cap bank as per this attached pdf. Does this pdf appear to be correct?
Will sort this out tommorrow and see if power supply hooked up properly fixes it, or will I have blown out some of the fets....funny however, it worked with this error for about half an hour.
I believe I made an error in the power supply which manifested itself... I mistakenly hooked positive from the bridge rectifier to negative on the capacitor bank. I attach a pdf of power supply diagram I used, just to make sure this is correct, positive from the bridge rectifier goes to positive side of capicitors on both positive as well as negative cap bank as per this attached pdf. Does this pdf appear to be correct?
Will sort this out tommorrow and see if power supply hooked up properly fixes it, or will I have blown out some of the fets....funny however, it worked with this error for about half an hour.