Hotrodding the UCD modules

Status
Not open for further replies.
Jump to Latest
#761
maxlorenz said:
Very interesting. Go on please. 😀



Yeah, it's a shame. Are there other classes of Rubycon that go up and still sound good?



Have you noticed this science fiction thread before?

http://www.diyaudio.com/forums/showthread.php?threadid=64748&highlight=

Look at post nº17 and at all Jneutron's replies.
I have tried it and it worked for me.

Good luck with your mods :smash:
Mauricio
Click to expand...

Hi Mauricio,

Are there any other good Rubycons, how about Blackgates😀 actually Rubycon YXF are not too bad, not sure about any more, the ZA and ZL's are still one of my favorites, shame they have magnetic legs

I'll let you know how the 470uf FC sound

Will check that number 17 post now
 
#762
In reading the posts about replacing the .68mf "output" cap on the UCD180 I was left with the impression that everyone was trying to mount the new replacement into the original holes but on the underside of the board. In looking at the board tonight I realized that this cap is actually placed right across the outputs as the last element of the circuit. Is there any reason one could not just use the large solder joints of the output connectors to hang the cap from? It seems much more secure than trying to hang it off the original solder pads nearby.

Thanks a lot,

Paul
 
#763
Thought I would throw my 2¢ in on the 470 ufd supply cap. So far I have tried 4 different caps’ there, The Yageo, Panasonic 680 ufd FC, Elna Cerafine and the Panasonic M-series. Of these I like the M-series the best by far. The Elna was a distant second. I noticed Panasonic has a 1000 ufd in this cap that will fit. I may try one to see if the extrs ufd's will give the improved bass of the FC's. I used the Elnas And the M-series with the Wima .68ufd caps in place. The Wima's have still been about the best tweak I have done with the caps. Also tried replacing the cap's in the low voltage section with Panasonic FC' but I did not care for them there either. I should have the new .68 ufd Auricap's in this week so I will be able to compare them to the Wima's.

Mourip, You can place the caps on the backside of the board as you mention. But if I remember Bruno correctly you need to keep the leads as short as possible to avoid RF problems. You can look at my post on page 68 of this thread to see how I install mine on the topside. Doing this keeps things closer to the ground plane.

Hope this helps,
Mike
 
#764
Mike2 said:
Thought I would throw my 2¢ in on the 470 ufd supply cap. So far I have tried 4 different caps’ there, The Yageo, Panasonic 680 ufd FC, Elna Cerafine and the Panasonic M-series. Of these I like the M-series the best by far. The Elna was a distant second. I noticed Panasonic has a 1000 ufd in this cap that will fit. I may try one to see if the extrs ufd's will give the improved bass of the FC's. I used the Elnas And the M-series with the Wima .68ufd caps in place. The Wima's have still been about the best tweak I have done with the caps. Also tried replacing the cap's in the low voltage section with Panasonic FC' but I did not care for them there either. I should have the new .68 ufd Auricap's in this week so I will be able to compare them to the Wima's.

Mourip, You can place the caps on the backside of the board as you mention. But if I remember Bruno correctly you need to keep the leads as short as possible to avoid RF problems. You can look at my post on page 68 of this thread to see how I install mine on the topside. Doing this keeps things closer to the ground plane.

Hope this helps,
Mike
Click to expand...

Great, thanks Mike🙂
I'll stick the Panasonic M-series on the list too if I can find them
 
#765
maxlorenz said:
Very interesting. Go on please. 😀



Yeah, it's a shame. Are there other classes of Rubycon that go up and still sound good?



Have you noticed this science fiction thread before?

http://www.diyaudio.com/forums/showthread.php?threadid=64748&highlight=

Look at post nº17 and at all Jneutron's replies.
I have tried it and it worked for me.

Good luck with your mods :smash:
Mauricio
Click to expand...


No I hadn't seen it but it is very interesting. Thanks bringing it up.

Cheers
 
#766
No I hadn't seen it but it is very interesting. Thanks bringing it up.
Click to expand...

Your welcome 😀
You can count on me to bring strange ideas 😉

Here's one pick of one of my humble monoblocks. I think the copper foil is visible on PS decoupling and "small decoupling" caps. No, I haven't made an AB test here due to lack of time 🙁


An externally hosted image should be here but it was not working when we last tested it.
[/URL][/IMG]

I wanted to share another ( off topic) point, concerning signal input wiring. I ran balanced source->TVC->UCD400. As shown on pic. I made a "conventional" wiring, as recommended on Hypex' notes. Well, I was suffering from an annoying HF problem (also with my previous amp) wich I mistakenly atributed to the DAC.
Today I changed to Rane's corp. recommended wiring and I can say I have almost no HF troubles (really? ).
Speakers are dead quiet now at idle.

http://www.rane.com/note110.html

Basically, it says hard ground Balanced connector's pin 1 to chassis and forget about attaching to "signal ground".

I decided to post this here as the original UCD thread is sleeping but I could post it also there.
Sorry if this was commented before. I can't remember everything, can I? 😡

Cheers :cheers:

Mauricio
 
#767
Maxlorenz,

Thanks for your post.

One question and one comment.

What value and type of cap are your bypassing your bridge with?


I can't tell from your pictures where chassey ground comes from.
It should come from the cap ground that is going to the amp.

My UcD400 is monoblock also, but single ended, for now. The system is dead quiet, and I tied by chassy ground to the cap ground, and the input sytem ground to the -Vin. If it were balanced, I would expect tying this to xlr pin 1 would be best, as it would minimize common mode issues.

If you've tied the center tap to the chassy ground, then to the cap ground, your going to get significant noise. Big currents flow in the center tap wires. You want the "star ground" to be the supply cap common point. Thus the connection to chassy ground is done by a wire with no current flowing thru it.
If you havn't done it that way, I think you'll hear a big differnce in stage when you do, and it will be dead quite too.

Chris is using balanced, so he likely will have some good advice.

Mike
 
#768
Hey Mike:
Thanks for your interest. 🙂

What value and type of cap are your bypassing your bridge with?
Click to expand...

I don't remember for sure. I think it's 47nf , polypropylene MKP, Epcos.
But these are going to the "reserve" because I got big IXIS , slow recovery bridges (68A). See:

http://www.partsconnexion.com/catalog/semiconductors.html



If you've tied the center tap to the chassy ground, then to the cap ground, your going to get significant noise. Big currents flow in the center tap wires. You want the "star ground" to be the supply cap common point. Thus the connection to chassy ground is done by a wire with no current flowing thru it.
Click to expand...

I think I understand what you mean. It's a big mistake from my part. Center tap is done soldering two leads from the TX and then tiding to the star :bigeyes: Now I see. :bawling: Very lazy I am.
Maybe I should cut those TX's leads to the minimum and take another wire from center tap's soldering point to star ground to avoid big currents.

Is that what you recommend??

Ah! every day I learn something new :angel:

My UcD400 is monoblock also, but single ended, for now. The system is dead quiet, and I tied by chassy ground to the cap ground, and the input sytem ground to the -Vin. If it were balanced, I would expect tying this to xlr pin 1 would be best, as it would minimize common mode issues.
Click to expand...

Here I lost you. Rane states incomming shield/ground tied directly to chassis without going to PCB's signal ground. Also optative is XLR connector grounded to chassis.
I think common mode is canceled at input opamp through signal + and - substraction. Am I wrong?

Many thanks for your post!!!
Mauricio
An ethernal beginner 😉
 
#769
maxlorenz said:
Hey Mike:
Thanks for your interest. 🙂


...But these are going to the "reserve" because I got big IXIS , slow recovery bridges (68A). See:

http://www.partsconnexion.com/catalog/semiconductors.html





I think I understand what you mean. It's a big mistake from my part. Center tap is done soldering two leads from the TX and then tiding to the star :bigeyes: Now I see. :bawling: Very lazy I am.
Maybe I should cut those TX's leads to the minimum and take another wire from center tap's soldering point to star ground to avoid big currents.

Is that what you recommend??

Ah! every day I learn something new :angel:



Here I lost you. Rane states incomming shield/ground tied directly to chassis without going to PCB's signal ground. Also optative is XLR connector grounded to chassis.
I think common mode is canceled at input opamp through signal + and - substraction. Am I wrong?

Many thanks for your post!!!
Mauricio
An ethernal beginner 😉
Click to expand...


I think how you tie xlr ground to the system will be almost a non-issue once you fix your chassy ground. Yes, get your transformer ground right to the copper plate. I'd tie it in on the opposite side from the ground that goes to the amp, but that's just my preference. That plate looks like it will be all the same potential, but having done lots of switch mode power supply layout, my experience is it won't be. Its likely minor at 120Hz, but i think it all matters. There will also be some switching frequency hash, and that's likely a bigger reason to do it this way.

If you do your ground the way you have it, then your chassy ground will be pretty much the same as system ground. BUT, the currents in the ground lead from the caps to the amp will have current, and thus not be exactly the same as chassy, that's why I'd use the "system ground" on the input to tie to the XLR shield.
That's what I'd do, then the input stage is referenced to the input ground of your preamp, and it should be best. If you do it to the chassy, the noise in that ground wire to the amps will be a common mode signal that the op amps reject. My guess is it won't manefest itself in audible noise, as the op amps are pretty good, but your staging will be affected and you will get better imaging if you do it as I recommend, but try it both ways and let us know.

Good Luck.

Mike
 
#772
Re: Speaking of FRED's

Portlandmike said:
What are the FRED's people are using and recommending?

I suspect we want soft recovery, correct?

Do you still want to add a cap, or series RC accross them?

Thanks

Mike
Click to expand...


68 amp 600V IXYS bridges, excellent choice. Hard to go wrong for the price of them, I'm a big fan of mine, certainly the caps aren't being starved. Ultra fast and soft recovery are both desirable characteristics. Soft recovery let's you get away without further bypassing, yet it remains optional, best determined by the rest of your system.
 
#774
Re: Re: Speaking of FRED's

classd4sure said:



68 amp 600V IXYS bridges, excellent choice. Hard to go wrong for the price of them, I'm a big fan of mine, certainly the caps aren't being starved. Ultra fast and soft recovery are both desirable characteristics. Soft recovery let's you get away without further bypassing, yet it remains optional, best determined by the rest of your system.
Click to expand...


The Hypex UcD HG is starting to look like a good price at those prices!

Has anyone tried them or the slit foil caps?


By the way Chris, that same company makes a 4 pole too!

http://www.bhc.co.uk/pdf/5444-fourth.pdf

This site sells them:
http://www.dnm.co.uk/capbhc.html



Mike
 
#778
$45 each, wow that is expensive for a small cap. I was going to buy these once they became available but not at that price - 1/2 the price of an entire UCD module for 1 component!

Does anyone have any listening impressions to see if they make that much of a difference?

Regards,
Dean
 
#779
Following with science-fiction contribution, anyone wants to test BG N type in "superE-cap" configuration for input coupling cap?

I use this type of cap arrangement in my DDDAC's output cap and found it very transparent and neutral.


I have DC coupled UCD400 and can't try them. Maybe I will try them on my UCD180ST (if I discover where are the caps to replace)

Not an expensive solution. 😉

Regards
Mauricio
 
Status
Not open for further replies.
Top Bottom