|
|
Post by szoze on Apr 1, 2012 14:15:09 GMT
Hi
I just saw on Mikes photos that he has attached larger heatsinks to the output transistors of the X-Can v3. Anybody knows the part number of these heatsinks on RS?
Cheers
Szoze
|
|
|
|
Post by PinkFloyd on Apr 1, 2012 14:25:17 GMT
Actually best not to fit them... the transitors are thermally coupled and heatsinks are not required. When you split them in two to slide the heatsink in the pads sometimes lift... they really are not necessary on the V3.
|
|
|
|
Post by szoze on Apr 1, 2012 18:14:32 GMT
Actually best not to fit them... the transitors are thermally coupled and heatsinks are not required. When you split them in two to slide the heatsink in the pads sometimes lift... they really are not necessary on the V3. Thanks Mike! I was looking at these: www.ebay.co.uk/itm/130660046515?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649 to possibly use in the output stage. Do you think they're good enough? I beleive that you used Suntan capacitors there. Shame I can't find them on e-bay.  |
|
|
|
Post by PinkFloyd on Apr 1, 2012 18:42:45 GMT
|
|
|
|
Post by szoze on Apr 1, 2012 20:05:44 GMT
Thanks. I think I will try those. What will the main difference be with larger capacitors in the output stage?
|
|
|
|
Post by PinkFloyd on Apr 1, 2012 20:59:01 GMT
Thanks. I think I will try those. What will the main difference be with larger capacitors in the output stage? Basically, one less capacitor per channel so already "less is more"... MF put in two x 220uF non polars per channel (in parallel) which equates to 440uF per side. Much better than the 220uF they use in the V2. Corner frequency isn't as important with low Z 'phones as it is with the 220uF they use in the V2... 440uF (in the V3) gives a cut off frequency of 11.3Hz with 32 ohm 'phones and 22.6Hz with 16 ohm 'phones so, really, one solitary 470uF coupling cap per side would fit the bill..... 470uF with 300 ohm 'phones gives a cut off of 1.13Hz. 1000uF (which I use) gives a cut off of 0.53Hz (300ohm) and 4.97Hz (32 ohms)..... Lets put it another way.... if you were to fit a 47uF output coupling cap it would cut the bass off at 105Hz in a 32 ohm load.... if you fitted 100uF it would be rolled off at 49Hz (32 ohm load). If you were ONLY ever going to use 300 ohm 'phones then you could get away with 100uF caps as the cut off would be 5.31Hz but 1000uF is a good "catchall" value for most common headphone impedances 32 to 600 ohm. The V2 employed 220uF per channel which equates to 45Hz cut off (16 ohms) and 22.61Hz (32 ohms) a bit too close to the bone (in my opinion).... even though we can't "hear" 7Hz through headphones (according to some) some 32 ohm headphones are specced at 7Hz to 35kHz so if the amp is cutting off the bass at 22.61 ohms then we are not hearing the full potential of the headphone in the lower registers. 1000uF also "sounds" better to my ears with high impedance 'phones... technically, 100uF would be fine but in my listening tests I found 1000uF to be the best for both low and high Z 'phones.... I certainly wouldn't supply more expensive 1000uF type caps with my kits if I thought 470uF sounded as good. Maybe Frans can expand / extrapolate / totally agree / disagree... he loves his measurements  I'll tell you WHY MF use two 220uF in parallel per channel too.... it's nothing to do with sound quality... they noticed my criticisms of the 220uF in the V2 (at the time) with regards to low impedance headphones and decided to fit 440uF into the V3... they had massive stock of 220uF non polars so instead of binning that stock they simply employed two of them in parallel (per channel) into the V3... I even commented on that when the V3 was launched, I couldn't BELIEVE their penny pinching was THAT bad  The V8 FINALLY saw the 1000uF per channel I had been going on about all these years (to be precise it was two x 470 in parallel per channel)... an inside source told me they were planning on upping the V8 to 470uF (one per channel) and bought a huge stock of 470uF, at the last minute they decided to go with "two" 470uF, in parallel, per channel giving a total of 940uF. So funny but I'm glad they eventually concurred that 1000uF was the best catchall value. It "could be argued that fitting two in parallel is a belt and braces approach... ie: if one of them failed your headphones would still be protected... maybe MF can use that as an explanation as to "why" the V3 and V8 had two per channel..... they'd have a hard time explaining the one 220uF per side the V2 had though Why not 2 x 110uF? or for belt, braces AND sellotape... 4 x 50uF? Hell.. the input coupling caps could fail so why not replace the one solitary 10uF with TEN 1uF in parallel... well I think we know the answer to that less is more when it comes down to sound quality. The ELNA or Nichicon are good quality caps, the likeliehood of failure in this position is nil (IMO) I have experience of MANY MANY V3's and you only need ONE coupling cap. Hope this has been of help. Mike. |
|
XTRProf
Fully Modded
Pssst ! Got any spare capacitors ?
Posts: 5,689
|
Post by XTRProf on Apr 1, 2012 23:56:16 GMT
Sorry, wrong thread thru the rush hour.  Had been transferred to the correct thread already. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 2, 2012 0:10:35 GMT
There is no way in the world that you can do a fair comparison here, as you have been raving on about your preference for the artificially warm sound of the vast majority of valve amplifiers since you became a member.  Have you heard a valve anplifier using grounded grid topology ? That is the only type of valve amplifier that has the neutrality and wide bandwidth of a good SS design. Any other comparisons are just Apples with Oranges. Alex |
|
XTRProf
Fully Modded
Pssst ! Got any spare capacitors ?
Posts: 5,689
|
Post by XTRProf on Apr 2, 2012 3:34:59 GMT
There is no way in the world that you can do a fair comparison here, as you have been raving on about your preference for the artificially warm sound of the vast majority of valve amplifiers since you became a member. Have you heard a valve anplifier using grounded grid topology ? That is the only type of valve amplifier that has the neutrality and wide bandwidth of a good SS design. Any other comparisons are just Apples with Oranges. Alex Had been transferred to the correct thread. Sorry .......... Hmm, let me put in the parts, beefed the PSU and the action will start after BI.  Btw, Ian and Miguel had been raving the V2 and not me yet and is one of the amps that they will keep. I don't buy just because it's a hybrid. I bought because it's a somewhat powerhouse hybrid. That's why I bought one to try as well as I don't believe in reviews. Else, I will have try the Sunrise. Frans' datas definitely put me on the road on Audiofool Day to find one. Quite fortunate I found one in good condition physically except for those problems mentioned that can be resolved quite easily like Pinkie had said. I somewhat concur with Ian and Miguel hook with it until I can honestly hear it properly after solving the current problems with it.  So keep your finger uncross as it's definitely coming. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 2, 2012 5:10:25 GMT
Mike
I think you may have meant to say here here that if one electro went o/c the headphones would still function.
If one went s/c or quite leaky, you would end up with DC to the headphones.
Kind Regards
Alex
|
|
|
|
Post by szoze on Apr 16, 2012 11:58:36 GMT
Maybe a strange question but is there any difference between bi-polar and non-polar electrolytic capacitors? The bi-polar electrolytics that I intend to use in the output stage (ELNA and Nichicon) have leads of different length (like polarized capacitors). Why is that important if the capacitors are bi-polar?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 16, 2012 12:39:40 GMT
Bipolar is 'similar' to 2 polar caps (double the value) with the - connected together (or the +'s). They are electrolytics (wet) that are not sensitive to polarity because of their construction. The shorter leg is probably a production thing (maybe they get assembled at similar production lines) This is just a guess though. Nonpolar are NOT electrolytic capacitors (dry). The nonpolars have a better longevity because they are dry. Distortion of AC signals depends on manufacturer/type/age. The 'sound' of capacitors is not 'my thing'. Sunrise (II) and Horizon all have 2200uF in the output stage. It's a low frequency extension thing of mine.  Fitting electrolytic polar caps on a position where there is no DC on it (output stage X-Cans may not be the wisest decision one can make. |
|
XTRProf
Fully Modded
Pssst ! Got any spare capacitors ?
Posts: 5,689
|
Post by XTRProf on Apr 16, 2012 12:55:26 GMT
Non polar and bipolar are actually terms used interchangeably. The longer and shorter leads may not be important to BP but according to BG they are important when connected in a super E-cap configuration.
Fyi.
|
|
|
|
Post by szoze on Apr 16, 2012 13:10:33 GMT
|
|
|
|
Post by szoze on Apr 16, 2012 13:12:22 GMT
Non polar and bipolar are actually terms used interchangeably. The longer and shorter leads may not be important to BP but according to BG they are important when connected in a super E-cap configuration. Fyi. Thank you. I suspected it was something like that. |
|
|
|
Post by szoze on Apr 16, 2012 18:54:57 GMT
I have now replaced the output capacitors with 4 Elna 220 uF bipolars. The amp has been on for 3 hours now. I have my friend's non-modded V3 and can make direct comparison. No blindfolds this time. The difference in sound is obvious, more defined bass, more musical, smoother treble and mids. The soundstage is bigger (especially noticable through AKG k701). k701 really sing now and I understand why peaople say these are very special headphones. I agree they really are.
I beleive this was the most obvious performance gain I've heard with this amp. I will also test with nichicon muse 470uF, two per channel.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 16, 2012 19:25:32 GMT
The short leg is connected to the 'casing' of the cap (similar to the short wire always be connected to the - of a polar cap)
|
|
|
|
Post by szoze on Apr 16, 2012 20:15:41 GMT
The short leg is connected to the 'casing' of the cap (similar to the short wire always be connected to the - of a polar cap) OK  But does it matter in which direction the cap is mounted in the circuit? |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 16, 2012 21:22:03 GMT
That depends on the circuit and if something can (accidentally) make contact with the metal housing.
In this case (as an output cap on a non-DC output) no...
|
|
|
|
Post by szoze on Apr 17, 2012 5:08:51 GMT
That depends on the circuit and if something can make (accidentally) make contact with the metal housing. In this case (as an output cap on a non-DC output) no... OK, I see. Than I guess it is the same with the input capacitors (10uF). Those stock input capacitors are 10uF/63V. Do you think there is a good reason for using capacitors rated as high as 63V? |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 17, 2012 5:29:04 GMT
They probably bought it in bulk.
Interestingly there are other caps in the 'audiopath' that also can be omitted to get an' improvement in the subjective area.
Here is a tip.
repeat the first blind test but jumper the input cap or not.
These caps are there to prevent the volpot from getting 'scratchy' due to DC.
|
|
|
|
Post by szoze on Apr 17, 2012 5:54:02 GMT
They probably bought it in bulk. X-Can seems to be a cost adapted design Interestingly there are other caps in the 'audiopath' that also can be omitted to get an' improvement in the subjective area. Here is a tip. repeat the first blind test but jumper the input cap or not. Yes, I intended to do that. I am a little uncertain about the DC from my source (H/K HD990).  I measured 4,8 mV DC from the source. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 17, 2012 7:06:51 GMT
No worries.. Even with the 10uF bypassed the X-cans STILL has an AC coupled (3Hz - 3dB) input so when there is DC on the input this doesn't reach the input of the amplifier. It is only there to make sure the pot won't 'scratch' due to (superimposed) DC on the input. It is not there to protect the amp to receive DC. |
|
|
|
Post by szoze on Apr 17, 2012 8:42:43 GMT
No worries.. Even with the 10uF bypassed the X-cans STILL has an AC coupled (3Hz - 3dB) input so when there is DC on the input this doesn't reach the input of the amplifier. It is only there to make sure the pot won't 'scratch' due to (superimposed) DC on the input. It is not there to protect the amp to receive DC. OK. I didn't know that . You are talking about C102 and C202, right? |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 17, 2012 9:43:49 GMT
C103 and C203 (near the input connector) are the actual input capacitors of the amplifier circuit.
They are primarely there to prevent the vol pot to become 'scratchy' due to grid leakage currents, on the side they also act as a low-cut-off filter (3Hz @-3dB) and thus as DC blockers as well.
C102 and C202 are the 10uF caps that can be bypassed.
They do serve kind of the same purpose as C103 and C203 but not for protecting the volpot from DC FROM the amps input circuit (grid current) but to 'protect' the pot from DC coming from the RCA input connectors.
They are only there to prevent the volpot to become 'scratchy' when there is DC on the input RCA's.
The -3dB point is around 0.3Hz.
C101 and C201 are the input high-cut-off filters, they do absolutely nothing for audio signals and only start to do something way above 20kHz.
They are there to prevent HF garbage which can be residu from DAC's, nearby HF transmitters (think radio amateurs) or oscillations in the audio chain e.t.c. (above 500 kHz) to enter the differential stage.
If everything is 'clean' you can even simply omit them.
I would keep them in.
|
|
Had been transferred to the correct thread already. 
Have you heard a valve anplifier using grounded grid topology ? That is the only type of valve amplifier that has the neutrality and wide bandwidth of a good SS design. Any other comparisons are just Apples with Oranges. 
Btw, Ian and Miguel had been raving the V2 and not me yet and is one of the amps that they will keep. I don't buy just because it's a hybrid. I bought because it's a somewhat powerhouse hybrid. That's why I bought one to try as well as I don't believe in reviews. Else, I will have try the Sunrise. Frans' datas definitely put me on the road on Audiofool Day to find one. Quite fortunate I found one in good condition physically except for those problems mentioned that can be resolved quite easily like Pinkie had said.
So keep your finger uncross as it's definitely coming. 
But does it matter in which direction the cap is mounted in the circuit?
I measured 4,8 mV DC from the source.