Hotrodding the UCD modules

Status
Not open for further replies.
Jump to Latest
#121
SX881663

"To get back on track I am going to be replacing the 470uf caps on my modules with much larger ones. I will be using regulated supplies so am opting for 63 volt units that will be at 59-60 volts. These are about the lowest ESR units I could find (around 7 millohm). Do you think there will be a problem with ringing? If so would an RC snubber suffice, thinking of something like .22uf and .2 ohm?"

Which make and value did you use to replace the 470uf caps?


Thanks
 
#122
Stevenacnj said:
SX881663
Which make and value did you use to replace the 470uf caps?


Thanks
Click to expand...

I just received my new modules and haven't done anything yet. The ESR spec I quoted was for a large bulk storage type cap, sorry. Nothing that good is available. Surprised no one picked up on that.
The ones I am trying I bought off of eBay,

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=36336&item=7526019566

Another I will try comes from,

http://www.illinoiscapacitor.com/pdf2/Series/AFD.pdf

There will be several others I will try that look promising including offerings from NIC, Nichicon, Panasonic and Sanyo.
I will report my findings.
Roger
 
#123
Re: Sonicap

sx881663 said:
Julien,
Teflon is a wonderful material for an insulator. It is so good that a static charge on it can last about forever. The problem with it is it is soft and don’t have a great deal of physical strength. This means that caps made with it can't be wound as tight and will have some mechanical give to them. The military versions get around this problem with metal jackets to shield and protect them. Unfortunately this totally screws up their sonics. I haven’t tested any commercial versions myself but would suspect things like a high level of microphonics as well as value change with applied voltage.
A cap is nothing more than 2 conductive surfaces separated by an insulator. Its value is directly related to the separation of these conductive plates. Any motion of the plates will change the value of the cap. With a fixed amount of charge on it this means the voltage must change as well. This is how capacitive sensors and microphones work.
The point is Teflon may not be that good of choice for this particular application where there is no DC and a lot of AC across the cap. The full audio output is across it as well as a bit of the switching frequency. Teflon really shines in tube applications where they have a large fixed voltage across them and may be exposed to high temperatures. This bias voltage creates a great deal of electrostatic force helping to keep the cap rigid and less microphonic. Extremely low leakage and long life make them the best choice here.
I know lots of words that don’t apply here but I would think useful knowledge to help make informed decisions for the proper application of these wonderful and expensive parts. If you do use them and can do an a/b comparison with a metalized poly type this would be useful information. Maybe all my worry is unfounded and these parts are even better than I think. Let us know.
Roger
Click to expand...

Thanks for the info, Roger.
I've ordered them allready, so I guess I'll have to try them. I might get some Auricaps in the future and compare. Worst case scenario, the Sonicaps will be good for PS bypass in another component...
 
#124
JP or Bruno

UCD 180AD Module question:

1.What is the largest diameter in MM that will fit (normal top mounting) to replace the two big 470uf 63v caps?

2. Do you see any problem if these replacement caps are mounted on the bottom side of the board?

Thanks
 
#125
Stevenacnj said:
1.What is the largest diameter in MM that will fit (normal top mounting) to replace the two big 470uf 63v caps?
2. Do you see any problem if these replacement caps are mounted on the bottom side of the board?
Click to expand...

1. 12.5mm
2. No. Especially if you're doing listening tests on tons of different caps, mounting them on the bottom side saves a lot of time.
 
#126
Bruno

Questions for UCD180AD module - These should be my last questions 😀

While looking at the top side of the module with the heatsink "T" at the top:

1. On the left side of the "T" heatsink, close to the heatsink there is a cap, which you called the bootstrap cap and the value is 220uf - 16v. Is this correct?

2. On the right side of the "T" heatsink, close to the heatsink, between the heatsink & coil, there appears to be another cap. What is the purpose and value of this cap?


Also due to your comments regarding C35 & C36 caps, I am not going to replace those.


Thanks
 
#127
Stevenacnj said:
1. On the left side of the "T" heatsink, close to the heatsink there is a cap, which you called the bootstrap cap and the value is 220uf - 16v. Is this correct?

2. On the right side of the "T" heatsink, close to the heatsink, between the heatsink & coil, there appears to be another cap. What is the purpose and value of this cap?
Click to expand...


1. Yes.
2. 680n, 63V. This is the C in the output LC filter. This is a nice tweak spot. Do not use caps that are too large, physically, because that ruins EMC performance and produces funny distortions at high output levels. Basically, parts with a pin pitch over 10mm are to be avoided.
 
#128
Bruno,

"680n, 63V. This is the C in the output LC filter. This is a nice tweak spot. Do not use caps that are too large, physically, because that ruins EMC performance and produces funny distortions at high output levels. Basically, parts with a pin pitch over 10mm are to be avoided."


Just to verify - 680n = .68uf?

Thanks
 
#131
Bruno Putzeys said:



1. Yes.
2. 680n, 63V. This is the C in the output LC filter. This is a nice tweak spot. Do not use caps that are too large, physically, because that ruins EMC performance and produces funny distortions at high output levels. Basically, parts with a pin pitch over 10mm are to be avoided.
Click to expand...


Bruno,

I was thinking about replacing the C cap on my UcD400AD modules with Sonicaps (Teflon cap). These are cylindrical and much larger (0.71"x1.30") than the one in place now. I was thinking about installing the caps under the board at the speaker output connection. Do you think that could be an issue? What kind of sonic distortions should I expect to encounter?
Thanks.
 
#132
Bruno,

Please look at the attched picture of a UCD 180 module.

I have marked the caps with values as I understand them.

Questions:

1. Are the values correct where indicated?

2. There are two electrolytic caps, on either side of heatsink, with question marks. What is the purpose of these caps and what value (uf & v) are they?


Thanks
 

Attachments

#134
Julien_M said:
I was thinking about replacing the C cap on my UcD400AD modules with Sonicaps (Teflon cap). These are cylindrical and much larger (0.71"x1.30") than the one in place now. I was thinking about installing the caps under the board at the speaker output connection. Do you think that could be an issue? What kind of sonic distortions should I expect to encounter?
Thanks.
Click to expand...
Such a large capacitor has a large series inductance, simply because of its size. Expect something on the order of 20nH for the size you quote. This allows quite a lot of RF "hash" to make its way to the speaker terminals. This manifests itself as a sudden increase of the switching frequency at higher modulation indexes. It sounds like a bad case of quantisation distortion ("tzzz"). It's quite possible that you won't notice at nominal listening levels. Usually such problems start only at several 10s of watts. The increase in RFI often shows up as whistling (difference frequencies between 2 amplifier channels).
I do have a trick to circumvent the self-impedance of large caps but it would be extremely unwieldy on the tiny 180's.
I can't stop you from trying it, but at least I've warned you 😀

fyi. in my book, teflon caps produce drop-dead-gorgeous holography in the stereo imaging, but also very serious colouration due to the low winding tension.
 
#135
Stevenacnj said:
1. Are the values correct where indicated?
2. There are two electrolytic caps, on either side of heatsink, with question marks. What is the purpose of these caps and what value (uf & v) are they?
Click to expand...
The caps marked "?" on your drawing are in fact C35/C36, and are 22uF/63V. The caps marked C35/C36 on your drawing are no longer there in the current product. The rest is correct.
 
#136
Bruno Putzeys said:

Such a large capacitor has a large series inductance, simply because of its size. Expect something on the order of 20nH for the size you quote. This allows quite a lot of RF "hash" to make its way to the speaker terminals. This manifests itself as a sudden increase of the switching frequency at higher modulation indexes. It sounds like a bad case of quantisation distortion ("tzzz"). It's quite possible that you won't notice at nominal listening levels. Usually such problems start only at several 10s of watts. The increase in RFI often shows up as whistling (difference frequencies between 2 amplifier channels).
I do have a trick to circumvent the self-impedance of large caps but it would be extremely unwieldy on the tiny 180's.
I can't stop you from trying it, but at least I've warned you 😀

fyi. in my book, teflon caps produce drop-dead-gorgeous holography in the stereo imaging, but also very serious colouration due to the low winding tension.
Click to expand...

Thanks for all the info, Bruno.
I guess I'll have to use these Teflon caps somewhere else then...
Is there any cap that you might recommend for this application that you feel is an improvement of what's already in place?
Thanks.
 
#137
Julien_M said:


Thanks for all the info, Bruno.
I guess I'll have to use these Teflon caps somewhere else then...
Is there any cap that you might recommend for this application that you feel is an improvement of what's already in place?
Thanks.
Click to expand...

Something that's on the edge of being too large but still doesn't produce too much adverse effects (and sounds great) is BC components MKP379, 0.68u, 160V. You will need to mount it on the bottom side of the board.
 
#138
Bruno

Thanks for all the cap clarification assistance. I have read the 180 Q&A and this thread and the info presented was different than pictures of the 180 module on your website. But I am clear now.

In regard to replacing the .68uf cap with a larger cylindrical cap. You wrote:

"This allows quite a lot of RF "hash" to make its way to the speaker terminals. This manifests itself as a sudden increase of the switching frequency at higher modulation indexes."

Would it help if we wrapped the larger cylindrical cap with ERS cloth to disapate the RF contamination?


Thanks
 
Status
Not open for further replies.
Top Bottom