I have used Saft VRE-D5500's (they tend to charge upto 7500 mah each with my graupner ultramat compu-charger) they are 1 year old now, bought them new. I have six 10-cell packs (6x12V) soldered with 5%S-95%Sn.
I connected the six packs in two series-tree parallel (resulting in 24V-22.5Ah) on dynamic speakers in the beginning.
After a while I connected my ESL-57 treble panels with the same "subs" on all 6 packs in parallel (12V-45Ah). I've also used two gel carbatteries (12V 65A each), but without audible difference to me.
The Saft "6-pack" (12V-45Ah) cost me roughly 400 USD, the car batteries were a lot cheaper...
Just like in my car I use vaseline on the poles, they're clean...
Cables are fine quality (a silver plated (for durability) twisted pair), I have done quite extensive A/B testing througout my life with a lot of cable qualities. I never "heard" dramatic differences between power cables. Speaker cables I think are most critical, but above 5 EUR/M I don't hear any difference either (I have "Rhapsody" (1500 EUR/pair!) speaker cables, very good looking, but that's about it...)
Or maybe I wasn't blessed with good enough ears...😀
My thinking is the "supply" should be able to deliver-and restore any drawn current from the load at least as fast as it is drawn. Any kind of supply should be able to "sound" equal given they comply to these demands..
Can you really keep different power cables apart in their sound blindfolded Soundcheck?
Still I can distinguish the difference between the SMPS-sound and fully-charged battery.
Or would it be the 3V difference, after all the SMPS is 15V and the NI-CAD packs are only 16.5V for the first couple of minutes., sink to 12.5V after this for hours... (this is normal, read NI-CAD specs...model car-racers profit largely from the "start-boost" only a NI-CAD can supply).
P.S. I've once read a very profound capacitor info page saying using a "bypass cap" in the supply-buffer won't make any difference in sound quality. It would rather make things worse...
I have to little experience and knowledge on this subject to judge for myself though...
I connected the six packs in two series-tree parallel (resulting in 24V-22.5Ah) on dynamic speakers in the beginning.
After a while I connected my ESL-57 treble panels with the same "subs" on all 6 packs in parallel (12V-45Ah). I've also used two gel carbatteries (12V 65A each), but without audible difference to me.
The Saft "6-pack" (12V-45Ah) cost me roughly 400 USD, the car batteries were a lot cheaper...
Just like in my car I use vaseline on the poles, they're clean...
Cables are fine quality (a silver plated (for durability) twisted pair), I have done quite extensive A/B testing througout my life with a lot of cable qualities. I never "heard" dramatic differences between power cables. Speaker cables I think are most critical, but above 5 EUR/M I don't hear any difference either (I have "Rhapsody" (1500 EUR/pair!) speaker cables, very good looking, but that's about it...)
Or maybe I wasn't blessed with good enough ears...😀
My thinking is the "supply" should be able to deliver-and restore any drawn current from the load at least as fast as it is drawn. Any kind of supply should be able to "sound" equal given they comply to these demands..
Can you really keep different power cables apart in their sound blindfolded Soundcheck?
Still I can distinguish the difference between the SMPS-sound and fully-charged battery.
Or would it be the 3V difference, after all the SMPS is 15V and the NI-CAD packs are only 16.5V for the first couple of minutes., sink to 12.5V after this for hours... (this is normal, read NI-CAD specs...model car-racers profit largely from the "start-boost" only a NI-CAD can supply).
P.S. I've once read a very profound capacitor info page saying using a "bypass cap" in the supply-buffer won't make any difference in sound quality. It would rather make things worse...
I have to little experience and knowledge on this subject to judge for myself though...
Attachments
Just ordered an AMP9 at 41hz, very curious to the sound... I'll post some pictures if nobody minds....
I also still have another MC4x100 that needs a case, want to make it as flat as possible so I'm going to make one myself again..
In the style of the little AMP32 case I recently made...
Pics in here: (just scroll down to see the latest...)
http://www.diyaudio.com/forums/showthread.php?s=&threadid=89549&perpage=10&pagenumber=4
I also still have another MC4x100 that needs a case, want to make it as flat as possible so I'm going to make one myself again..
In the style of the little AMP32 case I recently made...
Pics in here: (just scroll down to see the latest...)
http://www.diyaudio.com/forums/showthread.php?s=&threadid=89549&perpage=10&pagenumber=4
Whats the difference?
This is my second TAA4100 from Audiodigit, this time built in to a 'full size case for bi amping. The first one used a cheap SMP and I was impressed. So now I have added a 'soft start' board (so I can power up from the pre amp) and a balanced input module (with its own supply). I’m currently running it using the onboard power supply plus a good quality 10.000uF 50v off board capacitor and a cheap encapsulated transformer. I think the additional cap gives some more bass weight but I find it sounds a little ‘hard’ at times. The question is……...how much better would it sound with a high quality 27v dc linear power supply (it uses a 300va custom made transformer)?
This is my second TAA4100 from Audiodigit, this time built in to a 'full size case for bi amping. The first one used a cheap SMP and I was impressed. So now I have added a 'soft start' board (so I can power up from the pre amp) and a balanced input module (with its own supply). I’m currently running it using the onboard power supply plus a good quality 10.000uF 50v off board capacitor and a cheap encapsulated transformer. I think the additional cap gives some more bass weight but I find it sounds a little ‘hard’ at times. The question is……...how much better would it sound with a high quality 27v dc linear power supply (it uses a 300va custom made transformer)?
Attachments
I think it sounds a lot better with the SMPS they offer at http://www.autocostruire.com for 50EUR.
SMPS:
http://www.autocostruire.com/catalog/product_info.php?products_id=324
A thing you could try is making the leads shorter to the offboard cap.
Or get rid of the 2200uF onboard caps and replace them with three 15000uF caps like AMP9 from http://www.41hz.com
Did you notice a lot of EMI, is that why you added such extensive shielding 'round the amp? I think the amp might be better off with toroidal coils like the ones on AMP9.
The review at TNT is out by the way...
http://www.tnt-audio.com/ampli/autocostruire_mc4x100_e.html
Quite profound review, I hope AMP9 will be reviewed someday.....I think it sounds a bit better on all fronts...
SMPS:
http://www.autocostruire.com/catalog/product_info.php?products_id=324
A thing you could try is making the leads shorter to the offboard cap.
Or get rid of the 2200uF onboard caps and replace them with three 15000uF caps like AMP9 from http://www.41hz.com
Did you notice a lot of EMI, is that why you added such extensive shielding 'round the amp? I think the amp might be better off with toroidal coils like the ones on AMP9.
The review at TNT is out by the way...
http://www.tnt-audio.com/ampli/autocostruire_mc4x100_e.html
Quite profound review, I hope AMP9 will be reviewed someday.....I think it sounds a bit better on all fronts...
Thanks V-bro, I think my problem is that I’m a bit old fashioned when it comes to power supplies. I still find it hard to believe that a €48 SMPS can sound better than a 6A linier power supply that has virtually no ripple. But that’s the problem I just don’t have any experience with SMPS in HiFi.
Take your point on the caps, now you mention it the leads don’t need to be that long.
As for the screening this is an area that I believe should receive a lot more attention than it does. I have formed this view over the last few years and have seen the benefits of effective RFI and EMI screening. (The company I work for has an extensive Rad Haz test facility) Unfortunately it is an area that is not well understood (I am no expert, I rely on the guys at work) but if done properly, you shouldn’t be able to here it! The effects are subtle but a well designed and screened system has more ‘space’ and lower noise.
Take your point on the caps, now you mention it the leads don’t need to be that long.
As for the screening this is an area that I believe should receive a lot more attention than it does. I have formed this view over the last few years and have seen the benefits of effective RFI and EMI screening. (The company I work for has an extensive Rad Haz test facility) Unfortunately it is an area that is not well understood (I am no expert, I rely on the guys at work) but if done properly, you shouldn’t be able to here it! The effects are subtle but a well designed and screened system has more ‘space’ and lower noise.
Deckman said:
I do frankly prefer the simplicity of a linear supply, but believe me (as I've converted my belief) that class D sounds really brilliant on a 'switcher'...
Deckman said:
I've read the lead length is very important with class D...(believe I read it on the 41hz site...)
Deckman said:
I removed the rectifier from the board and placed the amp in a really tight-fitting aluminum cast case, the power supply has it's own case...so I think I 'accidentally' shielded the circuits quite well...😀
One thing I must add is that no matter the SMPS or way I shielded it the amp seems to be deadsilent at idle... Never heard any noticeable noise whatsoever from my speakers....other than the right noise
Ok v-bro I will take the plunge and get a SMPS. I’ll probably go for the same one as datawolf, which is a 150w job with variable output so you can get 26.5v http://www.powersolve.co.uk/jkcm/default.aspx (ESP150)
With regard to shielding, I do it as a matter of course now weather the circuit requires it or not! The problem is that there is so much EMI rubbish in the air now and its not easy to tell how susceptible a piece of kit will be, so its ‘belt and braces’ The thing to remember with screening boxes is to get the ones with deep labyrinth seals. Some of the cheaper boxes make things worse because they just act as a slot aerial! Holes are generally ok as long as the depth is more than 3x the Dia. Long thin ‘gaps’ are bad news!
(Your box looks fine)
I’m glad you mentioned the rectifier. I was thinking what’s the point in having a nice clean supply and then putting it through some ‘less than perfect’ components on the board. Is it a simple case to remove it, or are there links that need to be made? And on the subject of onboard components, can I get better quality capacitors on the board? What would you suggest?
With regard to shielding, I do it as a matter of course now weather the circuit requires it or not! The problem is that there is so much EMI rubbish in the air now and its not easy to tell how susceptible a piece of kit will be, so its ‘belt and braces’ The thing to remember with screening boxes is to get the ones with deep labyrinth seals. Some of the cheaper boxes make things worse because they just act as a slot aerial! Holes are generally ok as long as the depth is more than 3x the Dia. Long thin ‘gaps’ are bad news!
(Your box looks fine)
I’m glad you mentioned the rectifier. I was thinking what’s the point in having a nice clean supply and then putting it through some ‘less than perfect’ components on the board. Is it a simple case to remove it, or are there links that need to be made? And on the subject of onboard components, can I get better quality capacitors on the board? What would you suggest?
Oh yes, that SMPS looks fine! 🙂
I used desolder tape (copper stuff, flint like solution...suck all access solder away..) to remove the rectifier bridge. Works very well..
I'm not a capacitor expert and think any fast recovery cap will do. Many people find 'black gate' electrolytic caps great, or some 'Panasonic' types. Some prefer 'Mallory' or 'Rubicon'...I was more referring to the capacity of the caps 41hz.com use in AMP9 (3x15000uF!) compared to the Audiodigit MC4x100 (6x2200uF)
I don't know what the advantage should be of more smaller caps, I would figure in this comparison the three large caps in AMP9 are the winner...
I used desolder tape (copper stuff, flint like solution...suck all access solder away..) to remove the rectifier bridge. Works very well..
I'm not a capacitor expert and think any fast recovery cap will do. Many people find 'black gate' electrolytic caps great, or some 'Panasonic' types. Some prefer 'Mallory' or 'Rubicon'...I was more referring to the capacity of the caps 41hz.com use in AMP9 (3x15000uF!) compared to the Audiodigit MC4x100 (6x2200uF)
I don't know what the advantage should be of more smaller caps, I would figure in this comparison the three large caps in AMP9 are the winner...
Deckman said:
v-bro said:
That didn't quite answer your question...the answer is a simple 'no'.
...as long as you use the DC input of the board....I did manage to use the PCB holes from the rectifier to add an on-board fuse on the AMP9 PCB....allready had an external fuseholder for the Audiodigit so I never tried to place one on the board.....
v-bro said:
You will have lower total ESR and ESL from many paralell caps.
This needs to be very low at the switching frequency of the amp, 120Hz value should not be used.
Of course you can add a much larger cap outside of the board to
beef up the total capacitance.
/Eric
At least it didn't go bang!
Over the weekend I wired up the new SMPS. While I was there I made the leads to the external cap even shorter (1/2”) and removed the rectifier. It’s easy to see where the DC +/- go so I soldered the wires directly to the board. And then in great anticipation I flipped the switch and… nothing!
Generally when things don’t work it’s because I have done something stupid, but no, none of the wires were shorting, and everything was in its place. I suspect the power supply because when you turn on you can here the normal ‘start up’ noise from the speakers and a green LED in the PS comes on, but then goes off? I think that maybe the power supply is going in to protection mode and shutting down.
I’ll check it out later, when I’m in a better mood!
Oh the joys of DIY.🙁
Over the weekend I wired up the new SMPS. While I was there I made the leads to the external cap even shorter (1/2”) and removed the rectifier. It’s easy to see where the DC +/- go so I soldered the wires directly to the board. And then in great anticipation I flipped the switch and… nothing!
Generally when things don’t work it’s because I have done something stupid, but no, none of the wires were shorting, and everything was in its place. I suspect the power supply because when you turn on you can here the normal ‘start up’ noise from the speakers and a green LED in the PS comes on, but then goes off? I think that maybe the power supply is going in to protection mode and shutting down.
I’ll check it out later, when I’m in a better mood!
Oh the joys of DIY.🙁
I'm very sorry to read this Deckman, I wish I could help out a bit...🙄
Check whether you made any short by desoldering the rectifier...
Hold the pcb against a strong lamp so you can see through it, this can make it much easier to spot any bridges you might have made with desoldering...
Or it's merely the start-up taking too much power, or the SMPS can't hold up...
Oh the perils of DIY indeed...sigh...
Check whether you made any short by desoldering the rectifier...
Hold the pcb against a strong lamp so you can see through it, this can make it much easier to spot any bridges you might have made with desoldering...
Or it's merely the start-up taking too much power, or the SMPS can't hold up...
Oh the perils of DIY indeed...sigh...
Thank V-bro, the good news is that the amp is OK. I connected up a cheap 12v transformer PS and everything worked OK. (In fact it sounded embarrassingly good!) So we think the problem may lie with the large (10.00uf) external cap. I know some SMPS don’t like big capacitive loads, so that’s the task for tonight! If that isn’t it then, dear Watson, by process of elimination, it must be the PS. (either that or I haven’t used enough Hi Fi quality snake oil)🙂
It may very well be the problem, SMPS work on high frequency, so the caps don't even have to be that large...
They are also quite heavily protected most of the times...
And they work on a fine balance in which each component has to be within quite narrow tolerances, many -especially cheaper units- come DOA....
They are also quite heavily protected most of the times...
And they work on a fine balance in which each component has to be within quite narrow tolerances, many -especially cheaper units- come DOA....
🙂 We have lift off!
Everything is now working. The problem was eventually traced to the PS which has a +/_10% adjustment on the output voltage (24v) I thought that I’d left it in its nom position but it was just a bit too high. At 25.8v its fine and it shuts down at 26v! And the difference between it working and going to bed early is 0.00001of a bloody degree on the adjustment pot. Still, got there in the end. It sounded a bit ‘thin’ to start with so I re connected the external cap [10,000uf] which gave a worthwhile improvement to the bass. (It’s probably on the limit of what the PS will tolerate) The treble still sounds a little obvious, (more so than with the linier PS) but I’ll give it a few days to run in. I think a change to the input caps may be on the cards?
Everything is now working. The problem was eventually traced to the PS which has a +/_10% adjustment on the output voltage (24v) I thought that I’d left it in its nom position but it was just a bit too high. At 25.8v its fine and it shuts down at 26v! And the difference between it working and going to bed early is 0.00001of a bloody degree on the adjustment pot. Still, got there in the end. It sounded a bit ‘thin’ to start with so I re connected the external cap [10,000uf] which gave a worthwhile improvement to the bass. (It’s probably on the limit of what the PS will tolerate) The treble still sounds a little obvious, (more so than with the linier PS) but I’ll give it a few days to run in. I think a change to the input caps may be on the cards?
Glad you got it working! Perhaps you can measure or estimate the values in ohms from the voltage trimmer and replace it for fixed resistors....About the mods, I think a change to the coils might make a lot more difference...Have you 'heard' AMP9? I really like it, it has toroidal coils and a lot more options (and capacitance in the supply rail). It tends to have slightly more 'balls' than the Audiodigit..I think this is much due to the other coils...
And it costs only 49EUR.!!!
And it costs only 49EUR.!!!
And finally
Yes, I have had a look at the 41Hz website and the prices are not bad (better than Audiodigit) I think I may try something bigger next time. However, now the SMPS has had a good run in it does sound a lot better, more rounded, less ‘edgy’ It all seams to have settled down quite nicely. With the SMPS the bass is ‘faster’ with more impact. If your system needs a bit of sparkle, it’s the way to go. You can get a more ‘laid back’ presentation with a linier supply but you must use a good quality transformer. I have got the best results using a 300Kv Transformer and 20,000uf external cap, (but you should use a soft start unit to cope with the inrush current) for now I'm going to stick with the SMPS.
If anyone is interested I have a couple of SMPS (Traco Power ESP150-24S. RS No 396-3061) that are surplus to requirements. £70 each+P&P (There are circa £160 from RS)
Yes, I have had a look at the 41Hz website and the prices are not bad (better than Audiodigit) I think I may try something bigger next time. However, now the SMPS has had a good run in it does sound a lot better, more rounded, less ‘edgy’ It all seams to have settled down quite nicely. With the SMPS the bass is ‘faster’ with more impact. If your system needs a bit of sparkle, it’s the way to go. You can get a more ‘laid back’ presentation with a linier supply but you must use a good quality transformer. I have got the best results using a 300Kv Transformer and 20,000uf external cap, (but you should use a soft start unit to cope with the inrush current) for now I'm going to stick with the SMPS.
If anyone is interested I have a couple of SMPS (Traco Power ESP150-24S. RS No 396-3061) that are surplus to requirements. £70 each+P&P (There are circa £160 from RS)
More power? Check http://www.diyaudio.com/forums/showthread.php?s=&threadid=93149&perpage=10&highlight=&pagenumber=13
Direkt link:
http://www.profusionplc.com
The same chip as Audio research uses...
Direkt link:
http://www.profusionplc.com
The same chip as Audio research uses...
Use a big battery and big stiffening capacitor to deliver the peaks. it will deliver cleaner power than the SMPS. I tried it with my AMP1 with excellent results. Even with a high PSRR, the Tripath chip can really benefit from good quality power.
To get even better sound, you can have a pair of AMP5's and put them in a balanced configuration. One channel each for R+, R-, L+, L-. I have no idea why, but it sounds warmer and cleaner. And obviously it can deliver more output.
To get even better sound, you can have a pair of AMP5's and put them in a balanced configuration. One channel each for R+, R-, L+, L-. I have no idea why, but it sounds warmer and cleaner. And obviously it can deliver more output.
Or use AMP2, AMP7 or AMP8 in balanced mode...
No really, Jan (41hz) had a board ready for the TA0105 chip when Tripath just decided to quit production.... The profusion chip is expensive, but advertised being better than TA0105 which is regarded the most impressive chip Tripath ever produced... (rail voltage up to 200V!!
😉 )
Damn Rohs regulations!
No really, Jan (41hz) had a board ready for the TA0105 chip when Tripath just decided to quit production.... The profusion chip is expensive, but advertised being better than TA0105 which is regarded the most impressive chip Tripath ever produced... (rail voltage up to 200V!!
😉 )Damn Rohs regulations!
- Status
- Not open for further replies.