Hotrodding the UCD modules

Status
Not open for further replies.
Jump to Latest
#162
originally posted by Bruno Putzeys

Capacitors rated for "high frequency" are actually optimised for handling large currents. The currents here are not very large (few 100mA rms), so you can try the "general purpose" type as well.
Click to expand...

Hi,

I think that a word of caution is in place here. If UcD modules are specified as in good old times "400W into 4Ohm fom 20Hz to 20Khz with no more than 0.1%THD", then testing at full power at 20kHz puts considerable stress on output capacitor. 40V rms at 20kHz will produce 3.41A rms circulating current through .68uF output capacitor. I think this is beyond most polyester caps and calls for polyproplyene.

I have found a little wierd polyproplyene caps that might be suitable. They are 4 leg EPCOS B32656 series . They claim very low inductance, and despite very large dimensions (lead spacing 37.5mm, voltage from 850 to 2000V) 270nF part exhibits only 2 ohm impedance at 70MHz, yielding calculated inductance of only 4.5nH.

But I would agree that for normal music such large caps are unnecessary.

Best regards,

Jaka Racman
 
#163
Jaka Racman said:
I think that a word of caution is in place here. If UcD modules are specified as in good old times "400W into 4Ohm fom 20Hz to 20Khz with no more than 0.1%THD", then testing at full power at 20kHz puts considerable stress on output capacitor. 40V rms at 20kHz will produce 3.41A rms circulating current through .68uF output capacitor. I think this is beyond most polyester caps and calls for polyproplyene.
Click to expand...
It's correct that one should not run the amplifier continuously at 20kHz full power. Polyester caps will overheat if run for minutes in these conditions.
Long-term power-testing an amplifier at 20kHz is not normal practice. IHF testing calls for preheating an amplifier for one hour at pr/3 (1kHz), followed by Pr (1kHz) for 5min, then for verification of rated power at the frequency extremes.

Competing amplifiers tend to have a far greater frequency/power limitation built in under the form of a zobel network.
 
#164
output LC filter cap

Yesterday I got my polypropylene caps for the output LC filter on my UcD400AD modules.
They were all somehow compact, but much larger than the original 0.68uF/63V. I opted to install the smallest of them all, the Vishay/BCcomponents MKP 416 (63V). I was also very careful in keeping the cap leads as short as possible. While I had my modules out, I also replaced the bootstrap cap with Panasonic FM 220uF/25V that I happen to have. As usual, one must be careful, patient, and use a fine point soldering iron while modifying these modules.
After about 12 hours of music playing I can say that there's a definite improvement. The midrange has much more presence and the sound is more "together", more "round". There's also a noticeable improvement in handling musical crescendos. The system doesn't appear to be struggling as it could sometimes happen prior to this mod at high volumes.
Surprisingly, I found the sound of these caps very similar to Auricaps. I guess that's the sound of polypropylene film caps.
I think the caps still need about a week to be properly broken-in. There's still that typical unrelaxed sound when components are new, but overall the improvements are obvious and worth the trouble IMO.
 

Attachments

  • hotroded_modules_2_sml.jpg
    hotroded_modules_2_sml.jpg
    77.2 KB · Views: 1,293
#165
Hi,

another technically possible output filter capacitor would be metallized paper RIFA PME260. Company I work for has been using them for almost 30 years with excellent results regarding reliability in pulsed applications.

How do they sound? I have no idea, and I am afraid there is not much information on the net either. Maybe they are comparable to Jensen paper in oil caps.

Best regards,

Jaka Racman
 
#166
Jaka Racman said:
another technically possible output filter capacitor would be metallized paper RIFA PME260. Company I work for has been using them for almost 30 years with excellent results regarding reliability in pulsed applications.

How do they sound? I have no idea, and I am afraid there is not much information on the net either. Maybe they are comparable to Jensen paper in oil caps.
Click to expand...

Try 'em out I'd say, and tell us what you find. The last time I tried these was in my tube days, as coupling capacitor. They were overly bright (=quite the opposite from paper in oil), but otherwise quite appealing. They used to make a series called PHE430, and these were absolutely stunning (again in tube circuit, not tried as a class D output cap), but no longer made.
 
#169
Bruno,

Cap replacement .68uf - 63v

I was reading on Munorf's website and ran across this tidbit with respect to inductance with their Supreme series of caps:

"The Supreme capacitors all use a special induction-free winding technology. Two capacitor windings are interleaved so that their inductances effectively cancel each other out. These two windings are connected in series. This means that it takes two 2mfd windings to make a single 1mfd capacitor. This technology results in a more spacious and detailed sounding capacitor. This type of winding is used in both the Supreme and the Supreme Silver/Oil capacitors. "

What do you think?

Thanks
 
#170
Stevenacnj said:
I was reading on Munorf's website and ran across this tidbit with respect to inductance with their Supreme series of caps:

"The Supreme capacitors all use a special induction-free winding technology. Two capacitor windings are interleaved so that their inductances effectively cancel each other out. These two windings are connected in series. This means that it takes two 2mfd windings to make a single 1mfd capacitor. This technology results in a more spacious and detailed sounding capacitor. This type of winding is used in both the Supreme and the Supreme Silver/Oil capacitors. "

What do you think?
Click to expand...
When I wrote this:
http://www.diyaudio.com/forums/showthread.php?postid=672919#post672919

I was explicitly thinking of this story by Mundorf. This capacitor is not less inductive than a solid copper cylinder of the same size. Other film capacitors are not more inductive than that either.
You can't outwit the laws of physics.
 
#171
Bruno,

In regard to the .68uf - 63V cap, couple of questions:

1. What is the lead spacing of the current cap? (UCD 180AD module if differnent from the UCD 400)

2. Given the small package size requirement, cap selection for the .68uf value is very limited. Is there any flexibility to use a cap of greater or lesser UF value?

Thanks
 
#172
Stevenacnj said:
1. What is the lead spacing of the current cap? (UCD 180AD module if differnent from the UCD 400)

2. Given the small package size requirement, cap selection for the .68uf value is very limited. Is there any flexibility to use a cap of greater or lesser UF value?
Click to expand...

On the UcD480, lead pitch is 5mm. On the 400, a dual footprint is used to fit both 5mm and 7.5mm parts.

Performance-wise the circuit is not very fussy about capacitor value, but for various practical reasons I strongly suggest to stay within the 470n to 680n range.
 
#174
I think my output LC filter caps are finally broken in now.
IMO, they lost any big coloration they had. The main sonic difference is that the amp seems to be able to handle crescendos more dynamically. It doesn't seem to compress at high volumes. Bass presence seems improved as well as transparency.
Overall, after all the "mods" I've done in my modules, the tonality of the amp hasn't change much (a good thing) but transparency, dynamics, presence, and neutrality were greatly improved.
Lately, I've been revising some of my other components power supplies. I have 2 digital sources I've been working on, a CD-PRO2 kit and a Sony ES DVD/SACD 5 disc player.
I've been using Panasonic EB, FC and FM caps bypassed by Panasonic polypropylene film caps (usually 0.1uF), and the results are quite impressive. I think these are one of the best electrolytics out there. I haven't compare them with other high quality industrial caps such as Vishay mainly because I can't find these parts for retail. But compared to Nichiccon Muse, Elna, and BlackGates, I find them more transparent and neutral sounding. They take a while to break in (about one week). The sound is soft, unfocused and bottom heavy at first, but once they're broken in they seem to disappear sonically.
So for now I'm not touching my UcD400AD modules. I'm pretty happy with the way they're sounding.
 
#175
update

I am still waiting for the last of my parts to come in. I will be building 2 sets of identical monoblocks so I can make changes in one and do an A/B listening comparison. I plan on trying a lot of things as well as bench testing before listening.
I also like the Panasonic FC but it is getting a bit long of tooth. These are some new products from NIC, Nichicon and Rubicon that look very promising.
Roger
 
#177
Stevenacnj said:
sx881663

Care to share any of your planned mods (if different from what youve stated already)?

Thanks
Click to expand...



This is very much a work in progress. I haven’t finalized all I will try as I have a lot to learn. I will defiantly be replacing the 470uf caps, the input caps with off board 2.2uf Auricaps and the output cap as well. Not sure what the final part will be for that but am looking into several possibilities. Don’t want to screw up Bruno’s excellent EMI performance, in fact hope to improve on it. I am still anxiously waiting for the final parts to arrive so I can get on with it.
Another area I will be working on will be an off board +/- 12 volt supply for the front end. This will probably be a small SMPS with external buffer caps. Although small this supply will be as large as the amp! A lot of this is due to spacing requirements when dealing with line voltage.
All this is just one of many projects I am currently working on including some interesting stuff with 41Hz amp modules.
Roger
 
#179
Julien_M:

I'm reading your opinions very interested because we will end with a comparable system: I'm waiting for parts for a pair of UCD400 monoblocks and my pre is a DIY TVC with Stevens&Billington's MKIII Transformers, which will be joined in balanced operation.

I plan to get rid of input coupling caps and for "ground-decoupling " I'm tempted to use BG's in "superE-cap" configuration (if it wasn't for the money), because I recently tried them as coupling caps in my DDDAC 1543 and find them to be very clean and transparent. I hope someone try them first 😉

Now, for the complete out of topic part... :angel:
Not only we will share pre and power but also you have done two of my "to do" projects: CD-PRO2 and moded Samsung HD841.
Can I abuse of your kindness and ask for an opinion about these projects on a price/performance and enjoyment points of view.
Please, mail me if you can. :angel:

Thanks.
Mauricio
 
Status
Not open for further replies.
Top Bottom