tg
100+
YAGOF
Posts: 154 
|
Post by tg on May 9, 2009 0:42:36 GMT
Does anyone have a suggestion for an available substitute for a Black Gate N type 4.7uf 50V output coupling capacitor ?
Looking for near equivalent SQ/performance.
|
|
robertkd
Been here a while! 
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on May 9, 2009 0:49:36 GMT
Does anyone have a suggestion for an available substitute for a Black Gate N type 4.7uf 50V output coupling capacitor ? Looking for near equivalent SQ/performance. for me I would prefer 4.7uF metalised polyester over electrolytic's or better metalised polypropylene (not that i have been able to pick the difference between Poly and PP types) Robert |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on May 9, 2009 4:14:22 GMT
Does anyone have a suggestion for an available substitute for a Black Gate N type 4.7uf 50V output coupling capacitor ? Looking for near equivalent SQ/performance. Tony If you can make room, a couple of reasonably priced 4.7uf polyprops from Jaycar,(as seen in some SC HA photos) or if you have worked a bit of O/T, a couple of 4.7uFAuricaps from Soundlabs. Normal Wima MKS2 4.7uF sound crappy in comparison. Just ask the resident Chief DIY Bodger ! ;D Alex P.S. I presume you are talking about valve output stages, and thus no output capacitor isn't an option ? |
|
tg
100+
YAGOF
Posts: 154
|
Post by tg on May 9, 2009 5:59:57 GMT
Alex,
no, not tube output, DAC I/V gainstage. Space is a premium, Auricaps are $$$ and damn big compared to the space on the PCB (and under the lid) - I would be concerned about the effects of lead inductance as well (without understanding fully what that might be) though putting the input end close to the PCB might solve that one.
27mm by 29mm w axial leads v about 5mm by 11mm with 4mm pitch radial and $62 a pair plus postage.
The object is to rob a pair from one DAC to soup up the other one without compromising the one deprived too much (or costing too much).
I have not seen anywhere where someone has used this DAC without output caps.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on May 9, 2009 6:05:42 GMT
Tony
Seeing space is at a premium, do you have a pair of those Blackgates, or a pair of higher capacitance Blackgates? If no Blackgates, you could try a couple of 4.7uF Wima MKS2 , which would still be better than ordinary electros.
Alex
P.S.
Have you been able to measure the voltage to earth at the DAC end of the capacitors ?
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on May 9, 2009 6:46:55 GMT
can you get creative and fit 30*10 axial 4.7uF 100V poly?
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on May 9, 2009 7:10:58 GMT
Seriously lead inductance, I take it one side is driven by an opamp <50R and the other side presumably feeds a high impedance input >10KR ?? I modded the TDA1543 * 8 Stacker by AC coupling the input to the OPA604's with these ERO 1813 4.7uF 100V and removing the 10uf nichicon muse BP's on the output replacing those with 10R resistors overall much cleaner.  here  |
|
tg
100+
YAGOF
Posts: 154
|
Post by tg on May 9, 2009 8:40:00 GMT
Robert, opamp not - this is the circuit diyparadise.com/web/index.php?option=com_content&task=view&id=95&Itemid=5 Alex, looks like the most likely BG to be found is 47uf @ 50v - quite bigger (physically) but doable - what would be the likely effect of this big an increase in capacitance ? Or possibly 33uf @ 16v - was that the point of your question about measuring the output voltage ? The trimpot is used to adjust the output voltage to around half the input voltage, in this case it is about 7.5. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on May 9, 2009 8:54:34 GMT
Tony
I was trying to find out if it used a single supply rail, or +-supply rails. Probably O.K. with the 47uF, but why not try the 4.7uF Wima MKS2 first to see how that goes ?
Alex
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on May 9, 2009 9:00:43 GMT
Robert, opamp not - this is the circuit diyparadise.com/web/index.php?option=com_content&task=view&id=95&Itemid=5 Alex, looks like the most likely BG to be found is 47uf @ 50v - quite bigger (physically) but doable - what would be the likely effect of this big an increase in capacitance ? Or possibly 33uf @ 16v - was that the point of your question about measuring the output voltage ? The trimpot is used to adjust the output voltage to around half the input voltage, in this case it is about 7.5. meh still low impedance source so i'm thinking lead inductance isn't going to be a big contributor to anything in this situation. Ok if you have a bias voltage e.g. half supply then electros may not be so bad (yuck electros in signal path  )  mmm but I see a TDA1543  Ok I have these 2 muse 10uF 50V BP there 8mm dia * 12mm * 3.55mm ish lead spacing that I popped out of the TDA1543 *8 DAC PM me you postal and I'll bung them in the mail ;D Robert |
|
leo
Been here a while!
Team wtf is it?
Posts: 3,640
|
Post by leo on May 9, 2009 16:33:03 GMT
Those MKS2 would be ok there Tony, with low DC bias across them (mV's) they sound crap but with say 3v DC as in that discrete I/V they'll be ok |
|
leo
Been here a while!
Team wtf is it?
Posts: 3,640
|
Post by leo on May 9, 2009 16:36:33 GMT
BTW I have an M3 here too
|
|
tg
100+
YAGOF
Posts: 154
|
Post by tg on May 9, 2009 23:04:28 GMT
Thanks for the thoughts all,
Alex - it is single rail
Robert - thank you for the offer - PM sent
Leo - as always and as our resident cap "Connoisseur" - your input is much appreciated.
How do you find the M3 ? - how does it compare to the latest Beresford ?
I can't afford an M3 ATM but want to try the latest gainstage mods on my M2 - I have the board and most of the bits but could not source an extra pair of small BG N to do the "Super E Cap" output. Since I have both the M2 I built originally and a USB M2 I want to rob the output pair from the USB one and replace them without degrading the SQ too much.
quoting Yeo on the output caps
Final Q on this topic, since it appears I could obtain 47uf N type - will that much of an increase in capacitance have undesirable effects ? (given that although Yeo uses 4.7uf the original circuit used 100uf in that position).
Also note here that the input voltage in this case is 15 and the "bias" is 7.5 app.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on May 10, 2009 0:12:48 GMT
Tony
The lower value will give a small amount of LF rolloff, which is unlikely to be noticed with anything but low input impedance amplifiers. The higher quality of a decent non electrolytic 4.7uF, in comparison to a typical 100uF electrolytic capacitor, should more than make up for that. Perhaps that is why Yeo did that ?
Alex
|
|
tg
100+
YAGOF
Posts: 154
|
Post by tg on May 10, 2009 1:10:25 GMT
Alex,
note his comment re the transparency of the N type vis a vis film types he tried.
My question is will a 10x increase in capacitance have undue effect compared with his original spec of 4.7 ?
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on May 10, 2009 1:32:01 GMT
Alex, note his comment re the transparency of the N type vis a vis film types he tried. My question is will a 10x increase in capacitance have undue effect compared with his original spec of 4.7 ? the effects would be to reduce the lower frequency response, and given that the capacitor is in deed an electrolytic generally will increase the leakage current which is dependent on the capacitance. I also observed that the original value was 100uF polarised IIRC so honestly I doubt it will have any undue effect. The fact that the capacitors have some bias voltage across them also helps. you could also try out a super capacitor configuration I believe with 2 back to back which will give you 1/2 the total capacitance, e.g. 2 * 47uF reduces to 23.5uF! It will however give you greater turn on transients if your main amplifier is on and you turn this unit on, watch those speaker cones move  For me I would still prefer quality polys or PP or PIO over electros Robert |
|
leo
Been here a while!
Team wtf is it?
Posts: 3,640
|
Post by leo on May 10, 2009 1:46:26 GMT
Thanks for the thoughts all, Alex - it is single rail Robert - thank you for the offer - PM sent Leo - as always and as our resident cap "Connoisseur" - your input is much appreciated. How do you find the M3 ? - how does it compare to the latest Beresford ? I can't afford an M3 ATM but want to try the latest gainstage mods on my M2 - I have the board and most of the bits but could not source an extra pair of small BG N to do the "Super E Cap" output. Since I have both the M2 I built originally and a USB M2 I want to rob the output pair from the USB one and replace them without degrading the SQ too much. quoting Yeo on the output caps Final Q on this topic, since it appears I could obtain 47uf N type - will that much of an increase in capacitance have undesirable effects ? (given that although Yeo uses 4.7uf the original circuit used 100uf in that position). Also note here that the input voltage in this case is 15 and the "bias" is 7.5 app. I've not heard latest Beresford yet only the older one. Did you mean Buffalo? I find M3 very smooth, the HF roll off is quite noticeable but its pleasant , bass is a little soft I'm still experimenting with it The M3 uses pairs of 4.7uf's in parallel so you get around 10uf anyway. 47uf shouldn't be a problem, just remember BG caps have quite a heavy signature, the NX grades tend to be a lot smoother sounding than N, the key is not to over do it or the signature will just dominate |
|
leo
Been here a while!
Team wtf is it?
Posts: 3,640
|
Post by leo on May 10, 2009 1:50:35 GMT
BTW a 47uf N is 50v I think and would be quite expensive, I'm not sure if 10uf N is still available
I've just had a look and it seems these are gone too
If your determined to use BG N series I'd go for 33uf instead of those 47uf, the 47uf is also physically larger
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on May 10, 2009 2:59:18 GMT
Alex, note his comment re the transparency of the N type vis a vis film types he tried. My question is will a 10x increase in capacitance have undue effect compared with his original spec of 4.7 ? Tony Probably not. I doubt that too many here could give you a definite yes or no with that question. Is the expense of hunting all around the world for a no longer manufactured BG, worth the cost and effort in anything but a really top notch DAC ? In other words would you really be able to appreciate any differences due to it? Added to that, is that BGs aren't everybody's cup of tea. Just like 701s aren't everybody's cup of tea ? |
|
) 
