Hi Guys,
I'm going to fiddle with the value of R16 to bring the bias current down just a little. Last night (which was a little warmer than the previous few days) the heatsinks got up to 55 deg. C, which is starting to be a little too hot. I wouldn't be too concerned, of it weren't for the fact that this isn't the hottest (or driest) part of the year, and I lost speaker drivers to a blown mosfet eaflier in the year... So I'm going to see if increasing R16 from 0R46 I have there now to 0R5 or 0R56 will bring the temperature down to about 49-50 deg. C.
While I have everything open, I will probably follow juma's advice and take off the folded-wire heatsinks from Q4, assuming current measures as it should, and probably try to tidy up the wiring a little. Any other suggestions for tweaks? How about R7 -presently 0R46 - any point in altering it also? (Would be simple to put 0R56 there, maybe 0R5. Any benefit for switching from the wirewound I have to a metal oxide? )
Cheers
Nigel
I'm going to fiddle with the value of R16 to bring the bias current down just a little. Last night (which was a little warmer than the previous few days) the heatsinks got up to 55 deg. C, which is starting to be a little too hot. I wouldn't be too concerned, of it weren't for the fact that this isn't the hottest (or driest) part of the year, and I lost speaker drivers to a blown mosfet eaflier in the year... So I'm going to see if increasing R16 from 0R46 I have there now to 0R5 or 0R56 will bring the temperature down to about 49-50 deg. C.
While I have everything open, I will probably follow juma's advice and take off the folded-wire heatsinks from Q4, assuming current measures as it should, and probably try to tidy up the wiring a little. Any other suggestions for tweaks? How about R7 -presently 0R46 - any point in altering it also? (Would be simple to put 0R56 there, maybe 0R5. Any benefit for switching from the wirewound I have to a metal oxide? )
Cheers
Nigel
Actually, I have both in my drawer, so I have to choose... I suspect the choice makes very little diference, especially for R16. Any opinion on the value of R7? Worth changing from what it already is? (My guess is no, but I'd like to hear contrary opinions.)
Cheers
Nigel
That's the problem with Aleph type of amp - if you want it to sound really good you got to step on the bias current. It's your task to find compromising value that will still sound good and yet not dissipate too much heat - don't forget that it's an USA product, and you know what they say: There's no replacement for displacement (meaning - step on the bias and get bigger heatsinks)
Using metal oxide resistor should be advantageous (theoretically), but I doubt that anyone can hear the difference...
That's the problem with Aleph type of amp - if you want it to sound really good you got to step on the bias current. It's your task to find compromising value that will still sound good and yet not dissipate too much heat - don't forget that it's an USA product, and you know what they say: There's no replacement for displacement (meaning - step on the bias and get bigger heatsinks)
Yes, I understand this... That's why I spend so much time looking for big, cheap (preferably free) heatsinks
In this case I'm not looking to reduce the bias current much - I'll be happy just to bring 55 deg. C down to about 50 deg.C, as I ws originally hoping for.OK, then MO it is, if the values are right. I'm not willing to pay money for an improvement no-one can hear, but since both choices are free it make no sense to use the (theoretically) poorer option, either...
While I'm in there changing R16, should I bother with R7? Or leave it alone at 0R46? (I'm quite tempted to change it out for 0R5 also, wich is what you originally specified on the schematic... but again, will probably make little difference.)
I'll let you know how it turns out.
Cheers
Nigel
Changing R7 (in such a small range) won't change anything important (except DC offset, but you have a pot to adjust it). It's easy enough to change it and see if you get any difference...
So I changed resistors in and out for R16 a few times last night. After checking the resistors on hand I found that what I thought was metal oxide isn't, although they are Dale resistors bought from Mouser, which should be of more reliable quality than the no-brand stuff I can buy around here.
Now something I don't understand happened during the changes I tried. The first thing I did was to put in a pair of the Dale 1R resistors in parallel, to make 0R5. This worked OK, at least on a brief check, and brought the temp down to about 52/53 instead of 55. I then put a 1R2 standard 3W resistor (metal film? metal oxide? carbon something? No idea....) in parallel with one of the Dale 1R, to make about 0R55, and hooked it up to check temp. To my surprise I had a strange distortion; when listening with the right channel only I could hear sound from the left channel, a sort of static/noise/distortion, quite quiet but easily audible, in time to the music. Clearly something in the left channel was picking up the signal in the right channel. Same thing (although quieter) when listening to the left channel - same interference audible in the right speaker. My first guess was I had a bad solder joint or something (although I have no idea why a bad joint would cause this), so I resoldered all the joints to be sure, but the same thing happened. So I then took out both the resistors, and put in a 3W 0R56 wirewound, and the problem disappeared. This gives 51 degrees C right behind the mosfets compared to an ambient temp. of 28. So that's right about where I would want it, I think. It gives marginally more than 1A bias current, which is perhaps a little lower than one might hope for, but (IMO) this is probably all these heatsinks will take.
So, anyone have any idea what might have caused the distortion? It's gone now, so the problem is solved, but I'd still like to understand what caused it. The smart money has to be on some mistake I made, but is it possibly due to the resistors themselves? Or to some problem using different resistors in parallel?
Cheers
Nigel
P.S.
No 100k pot on hand, I'm afraid, or I would have tried this.
Now something I don't understand happened during the changes I tried. The first thing I did was to put in a pair of the Dale 1R resistors in parallel, to make 0R5. This worked OK, at least on a brief check, and brought the temp down to about 52/53 instead of 55. I then put a 1R2 standard 3W resistor (metal film? metal oxide? carbon something? No idea....) in parallel with one of the Dale 1R, to make about 0R55, and hooked it up to check temp. To my surprise I had a strange distortion; when listening with the right channel only I could hear sound from the left channel, a sort of static/noise/distortion, quite quiet but easily audible, in time to the music. Clearly something in the left channel was picking up the signal in the right channel. Same thing (although quieter) when listening to the left channel - same interference audible in the right speaker. My first guess was I had a bad solder joint or something (although I have no idea why a bad joint would cause this), so I resoldered all the joints to be sure, but the same thing happened. So I then took out both the resistors, and put in a 3W 0R56 wirewound, and the problem disappeared. This gives 51 degrees C right behind the mosfets compared to an ambient temp. of 28. So that's right about where I would want it, I think. It gives marginally more than 1A bias current, which is perhaps a little lower than one might hope for, but (IMO) this is probably all these heatsinks will take.
So, anyone have any idea what might have caused the distortion? It's gone now, so the problem is solved, but I'd still like to understand what caused it. The smart money has to be on some mistake I made, but is it possibly due to the resistors themselves? Or to some problem using different resistors in parallel?
Cheers
Nigel
P.S.
No 100k pot on hand, I'm afraid, or I would have tried this.
There's no way to tell which of the two in parallel would be the "suspicious" one, but in any event they all measure approx. where they should on the DMM. Should be borne in mind that the lowest scale on the meter is 200R, so I can't measure their resistance very accurately, but at least there is nothing grossly wrong. I can rig up some sort of circuit to measure resistance more closely, if there is any reason to do so.
In any event, if the resistance was very wrong then this would affect the bias current, but I can't see why this would produce the odd behaviour I saw. (Although maybe you're just suggesting I be thorough in checking things out, which is good advice, I guess...)
Cheers
Nigel
nothing else than note that sometimes happens that parts are miss-labeled .
probably not so often in DIY praxis , but relatively often in repair biz.
I'm not talkin' about fakes , that's another issue , usually reserved for active parts ( just a matter of overall quantity of used parts, nothing else ).
I had few times various resistors mislabeled as sub-ohmic ones , even if they're well in tolerance range of real value (meaning good one 470R labeled as 0R47)
probably not so often in DIY praxis , but relatively often in repair biz.
I'm not talkin' about fakes , that's another issue , usually reserved for active parts ( just a matter of overall quantity of used parts, nothing else ).
I had few times various resistors mislabeled as sub-ohmic ones , even if they're well in tolerance range of real value (meaning good one 470R labeled as 0R47)
That makes sense to me (and maybe it's good to check parts before putting them in), but in this case they measure OK, so that in itself can't explain the problem. Even if one of them was mislabelled (say labelled as 1R but really 1k) then that would give R16 of 1R instead of 0R5 (since they were in parallel) and a lower bias current, but how could that cause this odd distortion in the channel not playing?
Could a bad solder joint have caused this?
Cheers
Nigel
Hi ZM,
Well now that I've changed the resistors out and the problem has gone away it won't be possible to measure anything helpful, but can we treat it as a hypothetical? Suppose I measured the AC gain of the Aleph CCS - how could the answer help us? (except in confirming an accurate value of he resistance...)
This is surely the best guess. Do bad solder joints ever buzz? Or have some other faintly microphonic property? Could this happen if one of the resistros has bad inductive properties?
Cheers
Nigel
Well now that I've changed the resistors out and the problem has gone away it won't be possible to measure anything helpful, but can we treat it as a hypothetical? Suppose I measured the AC gain of the Aleph CCS - how could the answer help us? (except in confirming an accurate value of he resistance...)
This is surely the best guess. Do bad solder joints ever buzz? Or have some other faintly microphonic property? Could this happen if one of the resistros has bad inductive properties?
Cheers
Nigel
being in repair biz for 15,8 centuries ( since Obelix days, more precisely 
) I learned how to minimize banging my head ...... when I'm able to thoroughly check circuit after repair ( majority of them are bad joints and methodical check/replace of caps and/or semis ) - I'm never wasting time in thinking what happened ;
many times customers are talkin' deceiving stories ( probably asking for diminutive price ) ...... even if circuit is like open book regarding past history .
it's just tiresome thinking about culprit post hoc ....... even if I must allow possibility of fun factor - when my own gadget is in case .........
so - if you ask me - take a glass of wine and enjoy in music
) I learned how to minimize banging my head ...... when I'm able to thoroughly check circuit after repair ( majority of them are bad joints and methodical check/replace of caps and/or semis ) - I'm never wasting time in thinking what happened ;many times customers are talkin' deceiving stories ( probably asking for diminutive price ) ...... even if circuit is like open book regarding past history .
it's just tiresome thinking about culprit post hoc ....... even if I must allow possibility of fun factor - when my own gadget is in case .........
so - if you ask me - take a glass of wine and enjoy in music
ZM: I think I'm going to follow your advice... In this case a bottle of beer, not wine, but the same principle applies
A little Scriabin on the new amp, and a little thought to making new speakers for it... or maybe that tasty preamp juma designed using the BF862... hhmmmm...Thanks again for all the help
Nigel
be sure to have some - non-audiophool speakers ;
A class amp are tasty and seducing ; you need speakers of same breed
One of the things all "audiophool" pieces of equipment have in common is that a father of small children can't afford them...
Actually, until a blown mosfet and my stupidity destroyed them, I was using diy FE127E fullrangers. Getting replacement drivers down here is expensive, and a new project is always more fun, so I am thinking of doing something new. If possible it would be nice to use Brazilian drivers, but good information is harder to come by....
Out of interest, what amp/speakers do you use?
Cheers
Nigel
being in repair biz for 15,8 centuries ( since Obelix days, more precisely
Was Obelix around at the time of Vercingetorix, omnia gallia?
I'm using RCA LC1B in big ML TQWT , with any funny enough amp stumbled in my workshop at a moment ;
soon to use Sonido SWR200 in MLTQWT in my living room .
for cheap and hilarious good ( for my ears - better than many Lowthers ) - see FR drivers at Sonido akusztika
15.8 centuries is special number ; imagine it as any amount of time - you need at a moment ....
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.