opamp upgrade for MOTU HD192

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

matthias

Well-known member
Joined
Jun 4, 2004
Messages
768
Location
germany / frankfurt
hello,

I recently bought a motu hd192.

it features some very good converters;
the a/d is a akm ak5394 the d/a a akm ak4395

but I think the analog stage is not so "high end" and leaves some space for improvements.

the input after the xlr input looks like this:

drawn out schematic

pic of the pcb

as you can see I already changed two of the smd electrolytics to Panasonic FC 22µF / 50V

but what about the opamps... there's an jrc njm2068 (slew rate of only 6V/µs) in front of the akm converter.

njm2068 datasheet

so what type of dual opamp would you suggest ??

maybe burr brown opa2134 or opa2604 ... ?? or are there some better solutions...

do you think it's worth to change them or should I leave it as it is...??


thanks for your help.
matthias
 
I was told that the jrc njm2068 was "Supposed" to be an upgrade from a 5532 (Lower Noise)

BUT

I didn't like it at all when I tried comparing the two.
The 5532 still wins in my book.

It all depends on the application & also what your looking to get out of the "Sound"

To me, the JRC 5532D are very dynamic & alive sounding but can be colored/analog sounding compared to the NEW/FAST IC's (burr brown opa2134)... which are clear, tight but very "FLAT" sounding... not as much dynamic punch.

Once again, It all depends on the application & circuit.
 
I've read a lot on the net (I'm planning to do opamp upgrades for converters soon, too) and come up with the following:

LM6172 (seems to be a proven upgrade)

ths4032 (is regarded to be one of the best-sounding)

ths4012

LMH6655




They might need additional bypass/filtering caps to prevent oscillation etc. and might be pickier about power supply noise.

I haven't done the changes myself yet, but from audio examples I have heard this is definetly worth it.
 
The NJM2068 isn't a bad chip. Definitely better than many others used out there.

Black Lion does mod these as well. They speak of the OpAmps being the biggest weakness. I remember reading which ones they used, but I can't seem to recall it. And they don't have it on their website.

It looks like a simple enough circuit that you would notice an upgrade. BB2132,2134 are favourite replacements. They are pretty stable and are very flat. I think they would be an upgrade. But some don't like the sort of "laid back" sound they have.

Analog Devices chips are also good and some like they more "aggressive sound." Options like the OP275 or the AD8512 are all good. It is too bad they don't make a dual AD797 which is really highly regarded.

The THS4032 and LM6172 I have never used. But they look superfast, so you will probably need some bypass caps to keep them stable.

Anyway lots of options. Unfortunately it is SMD. So you can't really put a socket in and play away.

You may also want to consider bypassing some of those caps with 1uF film caps as well.
 
For some OpAmp comparisons, check this out:

http://tangentsoft.net/audio/opamps.html

With any audio review, it is all up to the reviewer. This isn't double blind or anything, so not scientific. Plus this is in regards to headphone amps, so sort of different applications.

But the opinions on that site, are very similar to the ones I have experienced and many that I have read on various forums.

There are some that are definite upgrades. But at some point, I found you really get into splitting hairs, and the differences are so subtle that opinions get really broad.
 
What sort of improvement are you seeking?

Recently I made some similar modifications to my Echo Gina24. It also came with NJM2068 op amps, but instead uses the AK4393 and AK5383 converters (which are also fairly high quality). I agree that the analog section of many existing recording interfaces can be a significant source of signal degradation. Although I have tried many of the op amps already listed, in my unit I ultimately chose the LM4562 for both analog input stages and the first two output channels. IMO, the LM4562 has a greater degree of transparency than most of the available competitors. I haven't tried the THS and LMH parts mentioned, however.

Along with the upgrading to the LM4562s, I also bypassed the input coupling capacitors with polypropylene film and removed the output coupling capacitors altogether. As I still have six unmodified outputs with the stock NJM2068s, I can easily compare the sound of the two op amps. Switching to the LM4562 immediately gives a clearer, more highly resolved sound - the subtle details of each instrument are more easily recognizable in the mix.

I'm not a fan of the OPA2134 or OPA2604. IMO, they sound kind of hazy in the highs. I've found the OPA2132 a little better in that regard. NE5532s are okay, but surpassed in openness of sound by the LM4562. I found the OP275 to have an odd coloration which I didn't appreciate, sort of a low-level compression/distortion - interesting, but ultimately fatiguing. As for the LM6172, I found it similar to the LT1358 - clear, but not as natural as the LM4562. Another chip worth looking into is the AD8599, which is a dual related to the AD797, but doesn't seem to have the latter's stability issues. IMO, the AD8599 is nicely detailed sounding, but somewhat warmer overall than the LM4562 (which some might describe as somewhat unforgiving).

My Gina24 seems to have 0.1 µF decoupling capacitors from the factory, so I'm not sure that adding more is necessary. Although I've had no instability problems with the LM4562s, I added extra decoupling anyway. BTW, where electrolytics are required, the newer Panasonic FM series are supposed to be an improvement over the older FC.
 
12Bass, that sounds like very good advice!

Could you elaborate some more on the difference between the high-speed opamps (LM6172, LT1358) and the LM4562 - what do you mean by natural? What's the difference regarding high end detail, transient response, perceived proximity, the perceived "immediateness" of bass?

Thanks!
Gregor
 
[quote author="living sounds"]Could you elaborate some more on the difference between the high-speed opamps (LM6172, LT1358) and the LM4562 - what do you mean by natural? What's the difference regarding high end detail, transient response, perceived proximity, the perceived "immediateness" of bass? [/quote]

It's been some time since I last did critical listening with various op amps and I can't remember the subtleties. I'd have to give them another listen in order to give a better description. By natural, I mean it sounds like it isn't adversely affecting the signal passing through it, thus allowing the natural details to come through. In my experience, almost everything in the signal path colors the sound in some way.... so I look for components which color the least.
 
12Bass said:
What sort of improvement are you seeking?  

A valid question which does not appear to be answered...

12Bass said:
I'm not a fan of the OPA2134 or OPA2604.  IMO, they sound kind of hazy in the highs.  I've found the OPA2132 a little better in that regard.  NE5532s are okay, but surpassed in openness of sound by the LM4562.  I found the OP275 to have an odd coloration which I didn't appreciate, sort of a low-level compression/distortion - interesting, but ultimately fatiguing.  As for the LM6172, I found it similar to the LT1358 - clear, but not as natural as the LM4562.  Another chip worth looking into is the AD8599, which is a dual related to the AD797, but doesn't seem to have the latter's stability issues.  IMO, the AD8599 is nicely detailed sounding, but somewhat warmer overall than the LM4562 (which some might describe as somewhat unforgiving).

These kind of comparisons are completely meaningless and are usually quoted out of context.
It depends a lot on the circuit configuration (ie where/how the opamp is used). Certain opamps perform certain tasks better than others in certain topologies.
 
Hi there,

I've changed decoupling caps on a couple inputs with Panasonic FM 22uF 50V.
It sounded like an improvement to me...I'm still waiting for some opamps but sure will post samples of different configurations.

...
 
I also plan on replacing op amp's (LM4562) and capacitors (Elna Silmic II). Has anyone already done the upgrade? Write to experience ...  :)
 
Back
Top