UcD180ad Vs 41Hz Audio AMP5 (Tripath TA2022)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Here are the two amps I built. I'd like to share the results of my today listening to both amps for comparation.
UcD_Amp5.jpg

First a few informations about the sound chain and the music played. Without these the report would be too subjective. Music is played from an Hard Disk in wav format, through an EMU1212M professional sound card, to an external Wolfson 8740 DAC (117dB SNR), with a toslink cable (to avoid ground loop and noise from pc). The DAC feeds directly the amps, while the pc acts as pre (just for volume control). Speakers are Fostex full range FE206E (96dB SPL) in a back loaded horn enclosures that go down to 40hz. No crossovers nor other filters.

Music played were only classical and the tests were done together with a professional violinist (my brother, owner of a Klimo valve sound system) to judge the amps with respect to live orchestral sound. In particular, string quartets (Rossini, Beethoven, Tchaikovsky), piano solo (Listz, Chopin) and wide orchestral pieces with "explosions" (Listz Rhapsodies and Mahler's 2nd symphony).

Results:

Both amps are VERY good. Both have been able to reproduce "natural" instruments and showed an impressive stage and dynamic. In the Mahler's 2nd symphony, for example, is needed e very wide dynamic range to reproduce the light flute coming from the back of the orchestra at normal volume, then suddenly a fortissimo with a lot of horns, trumpets and timpani: usually you must stand up and fastly turn down the volume. These amps, even with very loud "fortissimo", don't shout, but you can still perceive distinctly the various sections of the orchestra.

Which is better?

Well, from an objective point of view, UcD180 has better performances. Noise is lower and the instruments timber is more close to reality with UcD just as with valves. The sounds of strings, in quartets, and still more with piano solo, are more precise, have more fidelity to the original live instruments. The orchestral stage with UcD is too more "stable" and coherent: each instrument has his own precise location and correct volume. To judge the quality of high, mid and low ranges, too, UcD is better, the whole range is better controlled and you never hear a single sound that is not like it should be.

That doesn't mean that AMP5 is bad, we are at very high levels of fidelity, simply those things are done better by UcD. But there is something else that AMP5 does better. It is difficult to say but both me and my brother agreed: it is like as from AMP5 come out more information. It is not a matter of timber, dynamic, stage, bandwidth and so on. In a naive way I should say that AMP5 produces more sounds, even in a single instrument. In other words the sensation is that every sound is more rich and the whole stage is more "live".

Which one to hold?

My brother would prefer UcD, because of timber and fidelity that are priorities. I shall hold both, but I think that Amp5 has wide margins for ameliorations (for instance, dual mono configuration and better caps), while UcD is a "mature" project. Besides, richness of sound with tripath is so extraordinary that it is difficult to renounce to it.

Other experiences?

Regards

Thomas
 
classd4sure said:
Hello,

An interesting read. I'm just curious why you'd say the amp 5 can be further improved yet consider the UCD to be a "mature" project. I'd think it stands to reason you could improve on either one with better components and layout?

Would you not agree?

Regards,
Chris

Everything can be improved, I agree. But Amp5 is a diy project made by Jan as hobby (audio engineering is not his job) starting from a Tripath's reference board schematic, while Ucd is produced by professionals and already sold to manufacturers (in this sense it is "mature").

What would happen to Amp5 if somebody like Bruno Putzeys or you would improve this project?

Besides I'm sure there is much space for improvements because Audio-research and Belcanto-design tripath based amps sound much better.

Ciao

Thomas
 
Thomas, thanks for the interesting comparison.

One thing worth mentioning is that the UcD modules are sold to varied markets, not just high-end audio fans. Therefore price of the components is important and some parts such as the input caps can be replaced by better ones on the UcD180 and UcD400 as indicated by Hypex people themselves (Jan-Peter and Bruno). Maybe that would unveil some of the information you are missing?

I am finishing the first of two UcD triple mono amp (for a tri-amplified speaker). In the fall I will build 2 or 3 stereo 1000VA Plitron +/- 80V DC rails Tripath TA0104A amps (as I don't have on hand the two I previously built) and will be able to compare high-end Tripath with high-end UcD.

I am still wondering if with good amps I will be OK with anything less than my ~70' ribbon tweeters (which are overseas, I loved the TA0104A sound with them). With the passing years maybe my ears won't be able to tell the difference anymore... I think we need more comparison of amps and tweeters by people with young ears.
Guy
 
guyv said:
Thomas, thanks for the interesting comparison.

One thing worth mentioning is that the UcD modules are sold to varied markets, not just high-end audio fans. Therefore price of the components is important and some parts such as the input caps can be replaced by better ones on the UcD180 and UcD400 as indicated by Hypex people themselves (Jan-Peter and Bruno). Maybe that would unveil some of the information you are missing?

AMP5 costs €78 by 41Hz.... ;) for a new comparison I should replace better input caps for both: for UcD there is a big thread about mods, but no information about Tripath. Have you any hint? I'd be very grateful.

guyv said:
I am finishing the first of two UcD triple mono amp (for a tri-amplified speaker). In the fall I will build 2 or 3 stereo 1000VA Plitron +/- 80V DC rails Tripath TA0104A amps (as I don't have on hand the two I previously built) and will be able to compare high-end Tripath with high-end UcD.

Very interesting so a powerful Tripath amp. Audio Research 300.2 has such power http://www.audioresearch.com/300.2.html but I don't know what chip they use.

Have you a pcb and schematics of your TA0104A amps? This chip is stated as obsolete by Tripath, but on the market are still TA0105A and TDA2500. I don't know any diy hi end project based on best tripath chips, yours would be very interesting.

Regerds

Thomas
 
Basically I changed a resistor to adjust the gain to 26dB and replaced the input caps and resistor to make sure the highpass filter stays low enough FP = 1/((2PI x CIN)(RIN + 5000)).

Changing the input caps is similar on UcD and Tripath and you could use the same input cap to be fair to both amps. Just use the total resistance to ground instead of RIN+5000 above to ensure xover frequency is at or below 10Hz. Usually 2uF is a good value for the input cap. I had used digikey part P3394-ND .39uF polypropylene caps but will now use 2uF Auricaps from partsconnexion.

I am using eval boards for the TA0104A with mostly the above mods. I had bought them 4 years ago and never had a stable environment to build them. Some schematics are in the eval board manual at http://www.tripath.com/downloads/EB-TA0104A.pdf. The current eval boards are priced out of reach. I assume the Audio Research might be using the TD2500. The TD2350 seems to be their only recommended high power chip currently.

Regards,
Guy
 
thomaseliot said:

Very interesting so a powerful Tripath amp. Audio Research 300.2 has such power http://www.audioresearch.com/300.2.html but I don't know what chip they use.

The handles for the 300.2 cost $80 by themselves. ::shudder:: I didn't happen to see a price on it on the link you posted, but I'm sure it's there somewhere.

(ARC created a class T? I really must be living in a cave!)
 
Hi Guy,

thanks for the answer. I'm going to try two pairs of 2.2uf auricaps for both amps: I'll let you know the results.

TK2350 evaluation board is no more avalaible. But 41Hz's AMP2 use the same chips (but you have to solder yourself the very tiny legged two chips and other smd components). Now best Tripath chip is TDA1400 (max 400W 4ohm; 0,1 THD+N at 180W 8ohm) mono. Reference boards cost $600 and you need two.

Ciao

Thomas
 
eLarson said:


The handles for the 300.2 cost $80 by themselves. ::shudder:: I didn't happen to see a price on it on the link you posted, but I'm sure it's there somewhere.

(ARC created a class T? I really must be living in a cave!)


Hi eLarson

Audioresearch 150.2 costs $2900 while 300.2 $3900. But these are not based on Tripath evaluation boards, like Belcanto design Evo series, but on own designs with high value components. Look at this (amp 150.2):

An externally hosted image should be here but it was not working when we last tested it.


I think here is a lot to learn about PS and components (if I only could identify everything :bawling: )

Reviews of these amps are surprising: are valued as at the top on an absolute scale (not only in their niche).

If you are curious about best companies that have built Tripath based amps (all with best reviews) look at this link:

http://www.tripathpower.com/page3.html

Regards

Thomas
 
Thomas,
FYI I have gathered some diyaudio advice about using the auricap.

I am unable to attribute each quote to the exact author.

True, However if the cap has an identified outside foil lead, this should connect to the input jack for lower noise. Connecting the outside foil to the lower impedance can make a significant difference in noise pick up. In bypassing this would be the closest ground.
---
Originally posted by matjans
Q: How do i see which side of the film cap is the outside foil lead? Is there a sign for this?

I am only sure of the Auricap. Its outside foil is marked by a black lead. Some manufactures post this info on there websites. On stacked film types it generally doesn’t matter as the exposed surface is quite small for the value. I would use the stacked film type for coupling only if there were no options other than electrolytic.
Roger
---
If using input caps from the RCA’s put a 100k ohm metal film resistor across the RCA jack to insure any stored charge is dissipated.

Guy
 
I find it incredible you and your brother rated the UCd so high with the stock electrolytics in place.

I also note the small difference of a cover on your transformer for the amp5. Is it a fully encased ultra expensive audio grade unit??

What you're using for the UCD is obviously your most basic power transformer correct? As you said for the amp5 the UCD's may also be built full mono and with higher grade components.

Your comparison by no means definative.

However, since it is not capacitors for which you aim to compare, but amplifiers, my vote would indeed be to DC couple the UCD, which I don't think you can actually do with the Tripath based amp?

I think you'd find the difference nothing short of amazing, but maybe that would be a bit too unfair?
 
Thomas is using the Hypex Tr27 transformer.
http://www.diyaudio.com/forums/showthread.php?postid=940574#post940574

I recall another AMP5/UcD comparison from cotdt who was running a Tripath amp5 DC-coupled and UcD400 with stock electrolytic caps http://www.diyaudio.com/forums/showthread.php?postid=847523#post847523 . He eventually moved to UcD http://www.diyaudio.com/forums/showthread.php?postid=849256#post849256 ; he did not say what UcD mods he did after the first comparison. He also said in the above post that he got tired of repairing the AMP5 after too many 3 feet blue flames http://www.diyaudio.com/forums/showthread.php?postid=845831#post845831 . He did not seem to draw a correlation between running DC coupled and the blue flames... I will not try to run my own Tripath amps DC coupled.
 
classd4sure said:

I find it incredible you and your brother rated the UCd so high with the stock electrolytics in place.

I have a very good silent PC, DAC and efficient speakers: I built them myself :D

classd4sure said:

I also note the small difference of a cover on your transformer for the amp5. Is it a fully encased ultra expensive audio grade unit??

It is a normal Nuvotem encapsulated transformer (€36 RS-Components) 225VA. It would be enough 120VA, but I project to use it with dual mono AMP5 like UcD. :rolleyes:

classd4sure said:


What you're using for the UCD is obviously your most basic power transformer correct? As you said for the amp5 the UCD's may also be built full mono and with higher grade components.

Transformer is from Hypex designed for UcD180. PS is too from Hypex, HG PS (€200). Including the 2 UcD the Hypex set tested costs €470. For full mono I need €230 more and a new chassis. AMP5 costs €78 including dual PS on board (5V, +/-26V), Tripath chip and other components. But you have to solder the kit yourself. In any case is a very economic kit compared to UcD.

classd4sure said:


Your comparison by no means definative.

Of course. The first mod to UcD is to feed the opamp directly with +/-12V (Hypex transformer has another secondary for that and SupplyHG has those voltage ready. But I've not yet the courage to desolder 2 transistors on each UcD module. Then take the input caps off.

On the other side I have then to bridge two AMP5 to have same configuration and about same power as Ucd180. AMP5 is now 2X40W 8Ohm. Then Auricaps as suggested by guyv.

classd4sure said:


However, since it is not capacitors for which you aim to compare, but amplifiers, my vote would indeed be to DC couple the UCD, which I don't think you can actually do with the Tripath based amp?

My aim is not to compare but to enjoy music. UcD has something Tripath has not and viceversa. I'd just like to make them sound better to discover new things in listening. All answers in this thread are useful for this aim.

Thanks

Thomas
 
Yep:) Really not trying to be critical..... just observing out loud and hoping maybe it'll help you somehow.

I would DC couple before implementing an auxiliary supply, just because you're currently listening to an electrolytic cap with no DC bias across it, it makes for kind of a sloppy/mushy effect, which I think is probably worse than anything brought on by the onboard regulators, and will remain with the auxiliary supply in place. Also if you're nervous about removing the transistors, you'll be able to enjoy it for awhile DC coupled with the onboard regulators :)

I honestly am not 100% as to whether or not you are able to DC couple the amp5, at least I know with some Tripath's you can't, but I'm sure you can find out easily enough. 3' blue flames would sure have me looking for that answer.

guyv, photographic memory, have you? wow!
 
maxlorenz, thanks for the link. I didn't know so much to learn about caps.

guyv said:

I recall another AMP5/UcD comparison from cotdt who was running a Tripath amp5 DC-coupled and UcD400 with stock electrolytic caps http://www.diyaudio.com/forums/showthread.php?postid=847523#post847523 . He eventually moved to UcD http://www.diyaudio.com/forums/showthread.php?postid=849256#post849256 ; he did not say what UcD mods he did after the first comparison. He also said in the above post that he got tired of repairing the AMP5 after too many 3 feet blue flames http://www.diyaudio.com/forums/showthread.php?postid=845831#post845831 . He did not seem to draw a correlation between running DC coupled and the blue flames... I will not try to run my own Tripath amps DC coupled.

Thanks for the links. Here too something to learn (or to avoid? :D ). Interesting the copper foil inductors. Is there a pair of these in the 150.2?

It's a pity, comparing amps, forget to say what source, speakers and what music is listened. This way there is nothing to learn (is the AMP5 bridged to let a better channel separation like UcD or he listen to loud disco-music that could be played in mono with few differences?).

What do you think about the output caps in codt's AMP5? Really you can take these off and what would you use?

Ciao

Thomas
 
classd4sure said:
Yep:) Really not trying to be critical..... just observing out loud and hoping maybe it'll help you somehow.

Excuse me :).

May I do a stupid question?

Both UcD and Amp5 seem to sound too low for the power they have. So I should control and maybe fix the amp gain. The problem is how to calculate the input stage gain, as I must know the source signal level. My source is a DAC: where can I see its signal level?

Thanks

Thomas
 
I don't know. You said in your first post I think you're using the DAC to drive the modules directly. The 1212M will already have decent DAC's within.

I'm uncertain you're using it as a volume control in the digital domain, without already doing a conversion internally..

It seems the output of your DAC can vary somewhat by it's supply voltage.. and I have no idea how you're making use of it exactly.

What I'd do though, is ditch the external DAC and use the differential outputs of your audiodock, in fact, that is what I do, and it works great.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.