Comparison gainclone with UCD

[2000franz]

Hello everybody! I do own a pair of modified UCD400 mono amps. For my active speaker project i would like to listen to other amps (possibly lowercost) as well. I would like to compare my UCDs side by side against the famous gainclone, for example. Someone interested? Alternatively: Are there any readily built (possibly lower cost) gainclones from webshops?

[ashok]

Hi Franz,
You might have missed out on th efact that there are MANY flavours of the gain clone . To start with they have the LM3886 and LM3875 based units apart from other TDA chips. Then they have non inverting and inverting designs. Direct input and opamp buffered input and tube buffered input . Standard parts and exotic parts versions. Large capacity power supplies and skimpy power supply caps.
So which one would you you want to check out ?

[2000franz]

Hi ashok,
do they really sound soooo much different? From reading this forum, i would like to focus on the LM3875. Furthermore i'm interested in the 'pure' circuit, no input buffer, as my experience in audio often was "less is better". Is there is a circuit using the LM3875 in a differential input configuration? At least in theory, this would help getting the maximum performance out of my differential output DACs. And again: How does the result sound, when comparing to the UCD400?

[Dxvideo]

I always think the "less part is the best" like you.
However some DIYers say "buffered gainclones generates better results" especially for JFET buffered gainclones!
I havent tried a buffered one but I am happy with my LM4780 basic gainclone. Its as basic as the factory circuit.

[ashok]

Hi Franz,
I have some 3875 chips but have not yet built an amp with them. However I have tried the 3886 and like it very much. I also do have a pair of UCD400's but have not yet set them up. Will do so soon and will get back to you. I do have some Tripath amps from 41Hz. The sound is very different from the 3886 . Image seemed to be very precise in the Tripath amp but I felt it's a bit 'colder(?). That's not a good way of describing the sound but the best way I can put it now.
I haven't really tried them together and maybe they sound more similar than I think. I'll do a comparison after I rig up the UcD's.

The LM 3875 is generally regarded as better sounding than the LM3886 but I cannot say if both implementations used similar power supplies. The implementation and selection of parts could make a big difference . So making comparisons is not an easy thing to do and say chip A is better than Chip B.
All things being equal it might be possible that both chips sound very similar!

You will notice that some compared th eLM3886 to amps costing several thousand dollars and prefered the LM sound. Others don't subscribe to this opinion. So you should build the amp you want and make comparisons with your implementation .

About buffer amps and preamps . I also subscribe to less is better. But this has to be taken with a pinch of salt. In some cases a buffer or pre amp can sound better than a direct connection from the source to the power amp ( via a passive pot ).
I am not sure why this happens and it is not true in all cases. However when it does sound better I have not been able to find any difference in measurements to support the reason why it sounds better ( when it does).
Cheers.

[maxlorenz]

Hi 2000franz,

Do you mean that you will use your UCD400 plus other amps, together on your active speakers, or will you use other amps (same brand) on your actives?
Be aware that mixing amps on the same active speaker could atempt against homogenous sound. It is better IMHO to use only one type of amp. For example, all UCD180.

But if you want an excellent sounding amp you should consider:

http://www.twistedpearaudio.com/
Look for MyRef RevC.

Good luck,
M

[t.]

I've tried and heard quite a few chipamps, Valve buffered Inverted GC, Peter Daniels simple none inverted using fancy components, Pedja's J-fet buffered discrete regulated, Carlos LM3886 with snubbered supply and the Mauro MyrefC.
All sound very good, they have their strengths and weaknesses, which one is best, hard to say tbh, I'd say its more down to personel taste.
I did very much like Pedja's version

I have a few other amps too including highly modified UCD's, in my own system I prefer the modified UCD's, they sound the most neutral,detailed with a wider soundstage and more depth.
BTW all of the amps drive my speakers ok and have more than enough volume for my needs

[ashok]

Hi t ,
What mods did you do to the UcD modules ?

Franz,
What are 'your' mods for the UcD modules.

Thanks,
Ashok.

[t.]

psu's dual transformers, dual rectifiers with discrete MBR20200, T-network capacitors wired up as recommended by classD for sure.

modules, LM4562's with CRD's added to rail with worth psrr,coupling caps linked out, feedback/gain setting altered to better suit LM4562 using mini-melf resistors, onboard reg modified CRD's fitted biasing the zener, better caps etc, main 470uf caps replaced with BC136's,output 680nf filter caps upgraded, other bits and pieces I can't remember

[2000franz]

ashok,

my mod on the UCD400 is as follows: The OP amp differential input amplifier is completely removed and bypassed. This was possible, since my DACs have enough output level to drive the modulator directly.


maxlorenz,

thanks for Your comment! Yes, i am aware, that different amps for mid/high can lead to 'inhomogenious' sound, as can different speaker technologies. But i found, that a mix of amps between bass and mid/high is not critical, as long as the bass is not used for midrange. You find this mix in almost every satellite/subwoofer concept. So my idea is 1 UCD400 for bass and 2 gainclones for mid/high.