UcD400 Q & A

Originally posted by Bob Lewis;
You emphasize in the description of the UCD700 the low ESR of the bypass electrolytic caps, lowered even further by there being several in parallel, while at the same time warning against using too low a value in the 180/400 modules. What is different about the 700 modules which allows the use of low ESR components?
In the UcD700 Bruno has created a stripline configuration with all those capacitors in parallel. In the UcD400 there is only one capacitor and when you use a low ESR version the circuit will resonate. So there is mount a small capacitor of 10uF to damp this resonation.

Also could the DC offset adjustment circuitry of the 700 be adapted for use on the 180/400 modules?

Of course this is possible. No plans yet, but it would be a logical upgrade….;)

P.S. Excellent service by the way, the 4x 400AD modules purchased recently arrived only two days after ordering!
;) ;)

Jan-Peter
 
Jan-Peter said:

In the UcD700 Bruno has created a stripline configuration with all those capacitors in parallel. In the UcD400 there is only one capacitor and when you use a low ESR version the circuit will resonate. So there is mount a small capacitor of 10uF to damp this resonation.



Of course this is possible. No plans yet, but it would be a logical upgrade….;)


;) ;)

Jan-Peter



Just to confirm. I think the newer UcD400 modules have the BC136 capacitor (470uF) as decoupling cap. That is a low ESR cap but because it is followed by a small resistor (5mOhm) and then there is a 10uF cap which is not a low ESR type, there should be no issues with resonances, correct? So I think those big UcD400 caps could be safely replaced by low ESR types.

At least I hope so as I just bought quite a stock of 100V 680uF Panasonic FC caps, which are low ESR caps as well (had trouble finding a source for BC136, got my panasonics via digikey).

Anyway, I was planning to check with a scope if there would be any increase/reduction in ringing when caps are replaced. If I have some results, I will post them here (may take some time).

Best regards

Gertjan
 
Thanks, Jan Peter, for the overcurrent clarification in the new Ucd700. Does that mean that the Ucd400/180 shut down definitely when an overcurrent of enough duration is detected or that they just shut down for a while (2-3 secs)?

About the last discussion on low ESR capacitors. When you speak about ringing due to resonance, where are you measuring it, at the rails or at the speaker output?

Best regards,
Pierre
 
Gertjan:
At least I hope so as I just bought quite a stock of 100V 680uF Panasonic FC caps, which are low ESR caps as well (had trouble finding a source for BC136, got my panasonics via digikey).
Anyway, I was planning to check with a scope if there would be any increase/reduction in ringing when caps are replaced. If I have some results, I will post them here (may take some time).

Yes, please post your findings :)
I've been thinking of this mod but I'm not convinced about FC's. Everytime that I've used them on PS I've found unnatural, flat midrange and sterile (hard) HF.
I wish Rubycon's were 100V!
Will look for other brands also.

Are we talking about audible ringing or scope's mesurable ringing?

Good luck.
Mauricio
 
On a sidenote...

Just read about about THX specs for power amplifiers:

THX Ultra 2 specs demand the following (probably more but these is the most important parts)

* 3.2 ohms channel loading capability
* 1, 4 or 5 channel simultaneous testing (if multichannel amplifier)
* 18A peak current capability.

So, the UcD400 in a well designed 5 channel enclosure (or monoblocks) could possibly get THX Ultra2 approved :D

And if it doesn't... the 700 surely would!
 
Hi,

While I do find that interesting or maybe entertaining, even enough that I'd like to see the full specs for THX 2 cert. for power amps, I have to say the over driven amp sound from cert. theatres has ruined any semblance at hi fi for thx for me at least, it's just become another marketing tag and a very expensive one.

So while said modules may qualify let's hope we don't end up paying for such a stamp.

Regards,
Chris
 
Chris,

Well, here in Belgium we are very lucky with our "Kinepolis" group theatres - they also are proud of their THX label, but that's only one thing, they try to have the best quality possible - before this group, all independent theatres were going out of business in belgium, because of easy video rentals. DVD almost killed off theatres all together, but the quality they try to keep up worked, people started to return to see movies on the big screen...

Of course it is a marketing tag, but it was nice to see what the different specs actually mean (THX, select, ultra, ...) and that our beloved Hypex DIY amps could easily handle the highest home specs.

THX has sold out a bit though... the "thx" label for computer speakers must be ridiculous. The computer shop I work for has these l*g*t*ch "thx approved" sets... I'm always laughing when I hear them hum when idling from a few meters away... wow, dude, thx speakers! whoaa!

OTOH, the THX specs for theatres is something very different. They don't just look at amps & speakers, but at sound isolation in the theatre, they set a max. ambient noise, set a contrast range you should respect, a minimum/maximum image brightness, and even seating quality...
 
Yves Smolders said:
Chris,

Well, here in Belgium we are very lucky with our "Kinepolis" group theatres - they also are proud of their THX label, but that's only one thing, they try to have the best quality possible - before this group, all independent theatres were going out of business in belgium, because of easy video rentals. DVD almost killed off theatres all together, but the quality they try to keep up worked, people started to return to see movies on the big screen...

Of course it is a marketing tag, but it was nice to see what the different specs actually mean (THX, select, ultra, ...) and that our beloved Hypex DIY amps could easily handle the highest home specs.

THX has sold out a bit though... the "thx" label for computer speakers must be ridiculous. The computer shop I work for has these l*g*t*ch "thx approved" sets... I'm always laughing when I hear them hum when idling from a few meters away... wow, dude, thx speakers! whoaa!

OTOH, the THX specs for theatres is something very different. They don't just look at amps & speakers, but at sound isolation in the theatre, they set a max. ambient noise, set a contrast range you should respect, a minimum/maximum image brightness, and even seating quality...


Maybe there it is different, I've always found cheap sounding underpowered amps being heavily overdriven, common complaint that seems to reverberate is "too loud", and I'm more comformtable watching tv at home, sitting on my floor. To me THX has come to mean something you must endure.
 
Power Supply

Hello everybody...
Very long thread, I have tried to read through it, but I just can`t seem to find all the answers to my questions.. So, if this is has been answered before, I would be glad to see a link :)

What is a good PS for 2 ucd400? I don`t mean the best, because my budget won`t cut it, but an ok starter.. I have a 800VA 2x40 volt toroid. And I have 2x10000uF 63V caps.
I have some electronic skills, but not nearly enough:)() so values of components would be great appreciated.. ::D

1. Should I use 2 bridge rectifiers? (think i saw Bruno said one is ok)
2. Should the/those bridge(s) be bypassed? What values of the caps?..
3. "Solid state power amplifier supply guide" uses an RC network and and RF filter. Is this recommended, or commonly used by you guys?

Im going unbalanced (Denon 3803 as a source) Not sure what the manual means by: "In case of an un-balanced source: connect IN- to signal-ground at the signal-source". Anyone care to elaborate?

Is there any way to stop dc going in to the amp? Like a cap in series with the in signal? Or does that degrade the signal too much?

Hopefully someone will care to answer my questions..

Tor Christian
 
Re: Power Supply

tokki said:
Hello everybody...
Very long thread, I have tried to read through it, but I just can`t seem to find all the answers to my questions.. So, if this is has been answered before, I would be glad to see a link :)

What is a good PS for 2 ucd400? I don`t mean the best, because my budget won`t cut it, but an ok starter.. I have a 800VA 2x40 volt toroid. And I have 2x10000uF 63V caps.
I have some electronic skills, but not nearly enough:)() so values of components would be great appreciated.. ::D

1. Should I use 2 bridge rectifiers? (think i saw Bruno said one is ok)
2. Should the/those bridge(s) be bypassed? What values of the caps?..
3. "Solid state power amplifier supply guide" uses an RC network and and RF filter. Is this recommended, or commonly used by you guys?

Im going unbalanced (Denon 3803 as a source) Not sure what the manual means by: "In case of an un-balanced source: connect IN- to signal-ground at the signal-source". Anyone care to elaborate?

Is there any way to stop dc going in to the amp? Like a cap in series with the in signal? Or does that degrade the signal too much?

Hopefully someone will care to answer my questions..

Tor Christian

i had pretty much all these questions and more... there were all answered in this thread:

http://www.diyaudio.com/forums/showthread.php?s=&threadid=66069

its 18 pages, but there is a lot of good info there that answers everything.
 
Hi Tokki :D

I have some electronic skills, but not nearly enough()

I don't have any and I'm building my third amp (actually two monoblocks) :cool: I build the simplest of PS, the one that appears at Hypex' site and they sound excellent to me.

1. Should I use 2 bridge rectifiers? (think i saw Bruno said one is ok)

Start with one and enjoy your amps. When you think you deserve an upgrade you can try the other option. Mail me if you find any improvement :D


2. Should the/those bridge(s) be bypassed? What values of the caps?..

Yes, if they are common type. Bruno recommends 47nF.

Is there any way to stop dc going in to the amp? Like a cap in series with the in signal? Or does that degrade the signal too much?

The modules already have input caps. Actually this is one of the prefered target mods: upgrade for better sounding caps or take off if you have a DC free preamp. ;)

have a 800VA 2x40 volt toroid.

That will be OK. Only +/- 56VDC, less than maximum output, I supose, but good margin for your 63VDC caps.

2x10000uF 63V caps.

I would try to get two more : 4700 or 10000uF, Hi quality caps.

These are only advises from a well intentioned, non-expert, forum mate (who lives happilly with a pair of UCD amps ;) )

Good luck!
Mauricio
 
Power supply thinking

Hi all,

I'm researching power supplies for the UCD400. One of the characteristics I like about Class D amps is the speed ("attack") and dynamic nature they bring. So I'm looking to build a power supply using similar ideas to Hugh Dean's recommended ASKA power supply (very low ESR, dont use too large a transformer or too big capacitors). Interestingly this is a similar theory to the power supply for the "gainclone" trends.

My current thinking is for each UCD400 monoblock driving a Dunlavy SC-IVA (5 ohms, smooth impedence vs freq) for home theater and critical stereo listening:

- Use a slightly smaller transformer, 300VA per module. I dont need full 300 odd watts that the UCD will delivery with 5 ohm speakers as they are 91db sensitive in a mid sized room.

- Keep power supply noise down by using a toriod with an electrostatic shield between the windings, schottky diodes in the power supply (without snubber caps across the diodes), separate diode bridges per rail.

- Dont go over the top with the caps, and use good quality audio grade caps, with quite a few in parallel to really allow fast current dumping into the UCD during transients. I figure 10,000uF to 12,000uF should be sufficient, like the transformer dont go over the top in total uF size.

- Caps must be very low ESR to achieve a dynamic and fast sound, and doing a comparison of audio grade caps, the Rubycon ZL series (1500uF 63v) comes way out in front. Using these caps in parallel will give a very low ESR supply.

- Use a snubber on the output of the power supply, value to be determined while testing under load. I am undecided if the power supply module should have a snubber on the output, but I can experiment.

- Replace the UCD on board rail decoupling caps with the same brand/series as the main power caps.


My questions:
- I can't find a source to supply the Rubycon ZL at 1500uF 63v. Any ideas (I've tried farnell and rs-components, they dont have this size)?

- I recall a discussion about not changing one of the UCD on board caps due to the need for it to be a high ESR type snubber to avoid ringing. Doing a search its hard to work out if it is possible to swap caps on the UCD400 module (most of the discussion is for the UCD180). What caps are safe to swap out?

- Roger (sx881663) posted previously about a special run auricap to replace the output cap. Any news on that?

- Any feedback about this approach for a "dynamic" power supply by not using oversized toroid & parallel low ESR caps? Has anyone compared this type of supply with the typical DIY supply of few, large caps?

Regards,
Dean
 
Re: Power supply thinking

deandob said:
Hi all,

I'm researching power supplies for the UCD400. One of the characteristics I like about Class D amps is the speed ("attack") and dynamic nature they bring. So I'm looking to build a power supply using similar ideas to Hugh Dean's recommended ASKA power supply (very low ESR, dont use too large a transformer or too big capacitors). Interestingly this is a similar theory to the power supply for the "gainclone" trends.

My current thinking is for each UCD400 monoblock driving a Dunlavy SC-IVA (5 ohms, smooth impedence vs freq) for home theater and critical stereo listening:

- Use a slightly smaller transformer, 300VA per module. I dont need full 300 odd watts that the UCD will delivery with 5 ohm speakers as they are 91db sensitive in a mid sized room.

- Keep power supply noise down by using a toriod with an electrostatic shield between the windings, schottky diodes in the power supply (without snubber caps across the diodes), separate diode bridges per rail.

- Dont go over the top with the caps, and use good quality audio grade caps, with quite a few in parallel to really allow fast current dumping into the UCD during transients. I figure 10,000uF to 12,000uF should be sufficient, like the transformer dont go over the top in total uF size.

- Caps must be very low ESR to achieve a dynamic and fast sound, and doing a comparison of audio grade caps, the Rubycon ZL series (1500uF 63v) comes way out in front. Using these caps in parallel will give a very low ESR supply.

- Use a snubber on the output of the power supply, value to be determined while testing under load. I am undecided if the power supply module should have a snubber on the output, but I can experiment.

- Replace the UCD on board rail decoupling caps with the same brand/series as the main power caps.


My questions:
- I can't find a source to supply the Rubycon ZL at 1500uF 63v. Any ideas (I've tried farnell and rs-components, they dont have this size)?

- I recall a discussion about not changing one of the UCD on board caps due to the need for it to be a high ESR type snubber to avoid ringing. Doing a search its hard to work out if it is possible to swap caps on the UCD400 module (most of the discussion is for the UCD180). What caps are safe to swap out?

- Roger (sx881663) posted previously about a special run auricap to replace the output cap. Any news on that?

- Any feedback about this approach for a "dynamic" power supply by not using oversized toroid & parallel low ESR caps? Has anyone compared this type of supply with the typical DIY supply of few, large caps?

Regards,
Dean


Hello Dean,

On the UcD400, you can replace the 470uF caps. I think on the current UcDs, they are already low ESR (BCC136 I think). The UcD400 modules that I have have other caps, I think Yageo or something like that. Plan to replace them with Panasonic FC 680uF 100V caps that ate low ESR caps. I tend to agree with you that a bunch of low ESR caps in parallel should give the best power supply (from an engineering point of view). Those 100V FCs have an impedance of 36mOhm (at 100kHz), so 10-15 or so in parallel gives you only a few mOhm, that is very good and not achievable with any other single big cap.

The caps you should not replace are the two smaller caps that are used for the power supply decoupling close to the power transistors, these are the snubbers.

BEst regards and let us know the results

Gertjan
 
Talking about design changes,

I've got a pair of UCD400 about a month ago, and the output cap was very different from the older types - it's no longer a "white block" but a reddish rounded type.

Everything is built in my enclosure at the moment, but i'll try to post a picture soon.

This cap is supposed to have large influence on the amp sound, so I wonder why it has been changed.
 
Must have been a recent change as I got my UCD400 modules 2 months ago and have a pale yellow output cap. It would be good to know what happened to the auricap trail that Roger was organising.

Spent a couple of hours researching power supply caps by reviewing data sheets and listening impressions. It is clear that the Rubycon ZA/ZL caps are in a league of their own, very impressive ESR specs - however I can't find a supplier.

Next best looks to be the Nichicon KG series (but I can't find any data on their impedance), followed by the panasonic FC.

Regards,
Dean
 
Cap question

deandob said:
Must have been a recent change as I got my UCD400 modules 2 months ago and have a pale yellow output cap. It would be good to know what happened to the auricap trail that Roger was organising.

Spent a couple of hours researching power supply caps by reviewing data sheets and listening impressions. It is clear that the Rubycon ZA/ZL caps are in a league of their own, very impressive ESR specs - however I can't find a supplier.

Next best looks to be the Nichicon KG series (but I can't find any data on their impedance), followed by the panasonic FC.

Regards,
Dean

Dean,
The cap in question is stalled at the vendor. I am not sure exactly what the problem is and no further date of availability has been given. I will pass on info as I get it.
Roger
 
Dean:
Caps must be very low ESR to achieve a dynamic and fast sound, and doing a comparison of audio grade caps, the Rubycon ZL series (1500uF 63v) comes way out in front.

I love Rubycon's and wanted to put them as module's PS caps(instead of Yageo) but I only found 50V types. Did you find those 63V caps?

I am comparing now a BHC ALS40 based PS (2*22000uF/63V) to a Sikorel based PS (2*10000uF/63V) from my two UCD 180amps powering my active system: feeding HF/mid and LF alternativelly. Till now I can confirm the findings of Rha61 (Alain), about Sikorels being the fastest, more transparent sounding.
Quoting from another thread:

rather than paralleling PSU caps , i noticed that pi filtering is VERY effiscient ( on my UCD180 ) :
sikorel + aircore inductor 4mH Rs=0.5 + sikorel

Both caps are very good, IMHO, BHC on the darker, warmer side and Sikorel on the transparent, detailed side. But you must have a really good source because they will show you the bads.

I am now ending the enclosures for my UCD400's :)
I'm hesitating: 2 or 4 Sikorels for each PS? :cool:

Regards
Mauricio
 
Mauricio,

You can only get 1500uF 63v in the Rubycon ZL series, however I can't find any suppliers.

For a "fast" amplifier, I figure you have to allow the caps & toroid to dump current as fast as possible into the UCD modules when required, so resistors or inductors in series will "slow" it down. As I'm using schottky diodes in the bridge, there is less need to include high frequency filtering that a pi network will help with, and paralleling caps will significantly reduce the PSU ESR allowing good transient response.

Where did you source the sikorel caps? I'm struggling to find a good source for audiophile caps.

Regards,
Dean