Hotrodding the UCD modules

Status
Not open for further replies.
Jump to Latest
#61
sx881663 said:


The original opamps were the 5532's and had enough DC offset to be a problem. Using caps after them took care of this. Later when the opamps were upgraded offset was not an issue and the output caps were no longer needed. Input DC protection is still necessary so the input coupling cap requirement still exists.
Click to expand...


So, am i understand right that we can only bypass input cap if DC offset in speaker output (input dc protected with external cap) does not exeed "safety" voltage limit (10mV?)?
Is it same situation in UCD180 std v.2.0?
Maybe it is anyway better to use AD version, that has lower dc-offset.

Lot of guestions..
Thank you very much for your answers to newbie.
 
#62
Julien_M said:



Greg,
Try these:
http://www.my3c.net/
http://diyzone.net/index.php
You might want to contact Ken anyway. I'm sure he can arrange shipping to the UK directly from Taiwan.
Hope that helps.
Click to expand...


Hello all,

Thanks for all the help on the chassis. I knew all the links that everyone posted. Just the link posted by Julien was new to me. And I like those chassis since they have more room inside than most other chassis. I find 19inch chassis too big, but this one is 39cmx36cm inside so plenty of space.

I e-mailed Ken yesterday and got a reply in a few hours. There will soon be a E100 chassis (10cm high) specially for a 2-channel power amp application with all connectors already pre-installed at the back. I will go for that one. The size of the case is such that about any power supply and upto 4 or so UcD modules should fit in without a problem. I would have to make an additional connector set since I need 3 channels. Standard front panel is 4mm thick while for additional $30 you can get a 10mm thick front panel. I think I'll go for 4mm, that will be tough enough to drill holes for 3mm LEDs. The standard one will cost $120 and they can ship directly from Taiwan to reduce shipping costs if you are close to Taiwan.

Thanks and best regards

Gertjan
 
#65
ackcheng said:
Dear Gertjan

If you are in Japan, you can also order it from here

http://www.my3c.net/
http://www.my3c.net/product/default.asp?gid=116

This is the original site for these products. They are made in Taiwan, I have frequent order from them.... Quality is quite good. But language can be a problem


All the best!
Click to expand...


Hi Arthur,

Thanks for this site, I had seen it but had given up since it is all Chinese, I can read a tiny bit of it since I can read some (limited) Japanese. One of my colleagues is from Taiwan, so I'll ask him at work.

Best regards

Gertjan
 
#66
Pasi P said:



So, am i understand right that we can only bypass input cap if DC offset in speaker output (input dc protected with external cap) does not exeed "safety" voltage limit (10mV?)?
Is it same situation in UCD180 std v.2.0?
Maybe it is anyway better to use AD version, that has lower dc-offset.

Lot of guestions..
Thank you very much for your answers to newbie.
Click to expand...


Pasi P,
If you have off board caps to block the DC, jumpering out the electrolytics is the correct thing to do. With older versions that have the 5532 op amp you do need to check for output offset. I would like to see less than 30 MV, but up to 100 MV should not be a problem. Jan-peter or Bruno should be asked to verify this.
Yes, the AD version is superior in several respects including sound.
Roger
 
#69
pasi p, if your question was "can you replace the 5532 with an ad8620 without other modifications" the answer is no.
On a standard ucd180 the voltage used for the ne5532 opamp is +/- 18V (? or was it 15?); the ad8620 has a max voltage of +/- 13V so you have to replace some voltage regulators or zeners. I think it were zeners...

Yes, the AD version is superior in several respects including sound.
Click to expand...

Maybe, but the AD8620 has clear sonic signature, more than the 5532... I think they only put it on the market because it's a popular opamp with audiophiles...
I personally happen to like the sig of this opamp.
 
#70
matjans,

Thank you. I ask this because 5532 can have too much dc-offset that input caps can be shorted.

Anyway, sonically i am currently very pleased with std. UCD so maybe i will not change these to AD versions very soon.
(I have just received AD versions to my Orions treble amp, i am using std. versions in midrange amp.)


PS.
I am very imressed how much improvement i get to sound when i change Rotel RB971 to UCD180 in Orion midrange.
I have also tried UCD in treble replacing Creek 5350SE and i feel improvement is clear also. Soon i get both mid an treble to play with UCD, almost can not wait results..

UCD rules (and Linkwitz Orions also) 🙂
 
#71
IVX said:
Guys, what about Amidon core using for output choke, instead ferrite coil of the original UcD? JohnW and Bruno mentioned about Amidone as very linear core.
Click to expand...

What I don't like about toroids is that the switching signal is "visible" from the outside. This can (will) capacitively couple into the surrounding casework and cause EMI trouble. There's little impetus for me to move to toroid coils because these cores can only approach, but never exceed the linearity of gapped ferrite. Furthermore, there are tens of companies that make useful gapped ferrite cores, compared to only one supplier of good toroids (Amidon). So, I'm happy to stick with ferrite.
 
#72
transformer input?

You may know that Jeff Rowland uses input transformers in his 201 and 501 ICE-based amplifiers, as he obviously does with all or most of his amps. These are Lundahl LL 1545A.
Can it be that besides needing them for his balanced inputs another purpose is to bypass an input cap on the ICEpower module? Will it make sense to try this also with the UCD modules? Might give the same situation as using a TVC pre.
 
#73
Re: transformer input?

wiligruen said:
You may know that Jeff Rowland uses input transformers in his 201 and 501 ICE-based amplifiers, as he obviously does with all or most of his amps. These are Lundahl LL 1545A.
Can it be that besides needing them for his balanced inputs another purpose is to bypass an input cap on the ICEpower module? Will it make sense to try this also with the UCD modules? Might give the same situation as using a TVC pre.
Click to expand...
It's a way to avoid using capacitors, but not a very good one. If you have DC at the input, and hope to get rid of it using a transformer, keep in mind that the DC is biasing the primary, increasing distortion and potentially saturating the core. Thus, a transformer needs a coupling cap in front of it to keep that from happening.

Neither the Icepower, nor the UcD modules need a balanced signal. The input is differential, but in most cases the inverting input is used as a kelvin sense, tied to GND at the sending end of an unbalanced preamp.
 
#74
Re: Re: transformer input?

Bruno Putzeys said:

It's a way to avoid using capacitors, but not a very good one. If you have DC at the input, and hope to get rid of it using a transformer, keep in mind that the DC is biasing the primary, increasing distortion and potentially saturating the core. Thus, a transformer needs a coupling cap in front of it to keep that from happening.

Neither the Icepower, nor the UcD modules need a balanced signal. The input is differential, but in most cases the inverting input is used as a kelvin sense, tied to GND at the sending end of an unbalanced preamp.
Click to expand...


Bruno,
I think you are short changing input transformers. They do solve a lot of problems like ground loop induced noise. I am not sure why but well designed input transformers with a high nickel content core are very benign, unlike output transformers. A well designed transformer can take quite a bit of DC input as their input DC resistance limits current. They would be able to take a lot more than a direct amplifier input could. In any case excessive DC will only cause distortion not blow anything up.
A further advantage is the DC grounded input. The DC resistance of the winding is usually a lot less than the resistor it replaces improving noise and DC offset. Having the input bandwidth restricted, particularly at low freq. is a big +. All this noise signal reduction will give a deeper sound stage and things within it more in focus.
I to, did not want to deal with transformers till I was forced to in solving a nasty ground loop problem. In this case I used a standard Jensen input unit. It was a shock how much things improved without a clear sonic signature from the transformer. Having available direct single ended or balanced input capabilities with no switch or other compromises is also nice.
The only real disadvantage in most cases is physical size and cost not sonics.
Roger
 
#75
Re: Re: Re: transformer input?

sx881663 said:
I to, did not want to deal with transformers till I was forced to in solving a nasty ground loop problem. In this case I used a standard Jensen input unit. It was a shock how much things improved without a clear sonic signature from the transformer. Having available direct single ended or balanced input capabilities with no switch or other compromises is also nice.
The only real disadvantage in most cases is physical size and cost not sonics.
Click to expand...

It's quite clear that where warranted, a transformer can be an excellent problem solver. A transformer is a good idea when you want to equip unbalanced equipment with floating balanced I/O or when you have ground loop problems. However, as a means for removing DC I would still not recommend them.
 
#76
Re: Re: Re: Re: transformer input?

Bruno Putzeys said:


It's quite clear that where warranted, a transformer can be an excellent problem solver. A transformer is a good idea when you want to equip unbalanced equipment with floating balanced I/O or when you have ground loop problems. However, as a means for removing DC I would still not recommend them.
Click to expand...

Totally agree, DC should be removed first and a cap or transformer used only for covering the case where something fails. Even if it isn’t hurting anything I don’t think a DC bias voltage would improve the sound of interconnect cables.
Bruno, thanks for taking the time to contribute to our little thread, it is appreciated.
Roger
 
#80
Jan-peter,

What you are saying is that we actually need four capacitors for coupling modding? Because I understood that there are two of those before the op-amp. If there are two more after the op-amp, changing only the first pair will only afffect 50% of the possible changes in sound quality, maybe less... These two caps after the op-amp have any special value requirement, or should be dimensioned using the same logic applied for the other coupling caps?(22uf electrolytic -> 2,2 - 4,7 polypropilene) Is this the same for a AD8620 module version?

Allan
 
Status
Not open for further replies.
Top Bottom