• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

OP Amp upgrades - OPA2604AP (obsolete)

Hyperion

Hyperion

Roobarb & Custard
Subscriber
Quite a while ago, and on EW's recommendation I upgraded the OP amps in my AU-D11 II from the original M5219L SIL-8 package OP amps to OPA2604 PDIP-8 OP amps. This was only possible because of the excellent 'Brown Dog' PCB adapters made by Cimarron Technology ( https://www.cimarrontechnology.com/ ) (no afil'). Using them meant you could easily replace the above for a very worthwhile improvement in SQ and offset stability.

Move forward a couple of years and I discover the OPA2604 PDIP-8 package OP amp has been discontinued and is NLA. After a bit of research I found out that the reason it had been discontinued was that its performance couldn't be guaranteed with the supply voltage close to its published limit of +/-24V because of heat output and package dissipation concerns (these OP amps in the AU-D11 II do get warm). I couldn't have my beloved AU-D11 II's future existence in doubt because of running hot OP amps that are obsolete. So I began to research a solution that would resolve this problem.

The OPA2604 was chosen because it has FET inputs and was one of the very few OP amps that would operate with up to a +/-24V supply, most others being +/-18V or lower. The supply provided in the AU-D11 II being ~ +/-21.5V meant they were even being under-run a little. So, I hear you say, why not tweak that supply and bring it down to +/-18V to expand the available OP amp choices? The problem is that the supply is not just for the OP amps, but it also supplies the associated bipolar transistor circuitry as well. There were also concerns that the SQ from the OP amp might be affected by a lower supply voltage. So as far as I could see the effect of doing this would be significant, and not something I would like to even try.

I discovered that the SOIC-8 package OPA2604 is still available in very limited quantities, but also the single channel version of the same OP Amp the OPA604 is freely available in SOIC-8 or PDIP-8 - result! - To the rescue comes Cimarron again - with their dual SOIC-8/PDIP-8 single channel OP amp to SIL-8 converter - and we have problem solved. The single versions of the OP amp can shed enough heat for there to be no performance issues. Thus replacement is a simple job, I already have the OP amps and the Brown Dog adapters to make the conversion, and have made up one of the Brown Dog adapters with 2 (single channel) OPA604AP's in PDIP-8. :king:

Update 2022
The PDIP-8 OPA604AP is becoming obsolete, so need to use the SOIC-8 OPA604AU with appropriate Brown Dog adapter. (https://www.cimarrontechnology.com/product/browndog-040701-sil-single-to-dual-op-amp-adapter) ;)

I'll update here after I have changed them. ;)
 
Last edited:
Register to hide this ad
I've been using the OPA2604AP Dual op amps on BD adapters in my AUD-11II's for many years now with no issues. In another application I have some surface mount (SOIC-8) AD825's (I think that's what they are) soldered, one to each side of a BD adapter in my CD-63. It's a ticklish solder job, requires a very fine tip. But works well. Perhaps I'll follow your lead. What sort of temperatures are you seeing in the DOIP op amp?
 
Last edited:
Hi John,
Too bad they discontinued the opa2604. I agree that reducing the rails is not the optimum solution. You probably already descarded the lme49860 which is only +/-22V V. I am curious why a JFET input op amp would be needed here. Are you concerned about loading the input diffetential pair?
 
There was somebody using an OPA275 instead, but it is rated to 22mv, spec for 11II rail is 21.2 mv. That seems like asking for trouble..
 
LWB said:
I've been using the OPA2604AP Dual op amps on BD adapters in my AUD-11II's
Click to expand...
So have I - but I don't want to be caught out just supposing one fails. ;)
JoseHH said:
You probably already descarded the lme49860 which is only +/-22V V.
Click to expand...
Yes, and I knew the one EW recommended performs very well and sounds great. ;)
JoseHH said:
I am curious why a JFET input op amp would be needed here. Are you concerned about loading the input diffetential pair?
Click to expand...
No loading concerns, I thought the M5219L was a JFET input OP Amp - a check today reveals it appears to be bipolar. ;)
 
Last edited:
Just a n00b question: Is this potential for sound improvement on the phono stage with either the Cimmaron+OPA2604 PDIP-8 or Cimmaron+2*OPA604?
What would be the exact code for the Cimmaron board? Their webpage shows quite a lot of Op Amp Adapter boards and the Mitsubishi datasheet only says 8P5 outline, but no mech drawing of the package in the dataheet of the part.

EDIT: found the drawing of the 8P5 package here: http://bitsavers.informatik.uni-stuttgart.de/components/mitsubishi/_dataBooks/1987_General_Purpose_ICs.pdf
=> So the Cimmaron board is this one? https://www.cimarrontechnology.com/product/browndog-130801-single-to-dual-op-amp-adapter


The phono stage is the only thing I am a bit disliking on my 907-DR. When cranking up the volume on old records (that are low in volume but lots of dynamics so you crank up the volume and these records sound so great. Far better than overcompressed stuff that is just a wall of sound) there is quite some white noise and the general sound is a bit different to my old Technics amplifie that I would prefer. When the 907-DR will get its service, I bring back my AU-919 for sound comparison in the meantime.

EDIT2: the M5219L is speced to only +/-22.5V working conditons (+/-25V abs. max rating), while the OPA says +/-24V (with +/-25V abs. max rating, hui... that's close with max vs. abs max. :))
 
Last edited:
IMO, those DIP8 parts run way too hot. I thought about epoxying a small heat sink on them. Was there really a noticeable difference in sound or measurement with the "upgrade?"
 
ConradH said:
IMO, those DIP8 parts run way too hot. I thought about epoxying a small heat sink on them. Was there really a noticeable difference in sound or measurement with the "upgrade?"
Click to expand...
A big improvement in DC offset stability, and also in sound quality, the stock OP amps sounded 'artificial' compared to the new replacements - which sound marvellous. (there are DC offset settings for the 'Phono amp' or EQ and for the Control Amp, both of which contain these OP amps).

Regarding the heat, ever since I did the upgrade I have been concerned about how hot those OP amps get. Hence I have been looking for a solution for some time, but I was a bit stumped with the supply voltage issue.

The replacements (OPA604's) are still going to get hot, but being split into 2 packages reduces the thermal load considerably, 'considerably enough' so that the manufacturers are keeping them as a current product in PDIP-8 - at least for the time being. :rolleyes:

Thank you for commenting on this Conrad your input is always welcome. ;)
 
Last edited:
I know I mounted some SOIC-8 op-amps onto DIP-8 adaptors for @nj pheonix but I can't remember what kind they were. It sucks that more and more through hole parts are becoming obsolete.

Lee.
 
leesonic said:
It sucks that more and more through hole parts are becoming obsolete.
Click to expand...
Absolutely, the small crumb of comfort here was the discontinuance of the OP amp was for a technical reason, not my perception of the the usual reason, 'we don't sell enough of these to make a big profit, so we'll stop making them' reason. ;)
 
Yes, I can confirm that the OPA2604 is a distinct sonic improvement. If they run hot, I might try to mitigate that by following Hyperion's lead. But I have had these in my amp for many years now.
 
leesonic said:
I know I mounted some SOIC-8 op-amps onto DIP-8 adaptors for @nj pheonix but I can't remember what kind they were. It sucks that more and more through hole parts are becoming obsolete.

Lee.
Click to expand...

They were in fact OPA1611.
I later found a vendor that sells them installed on adapters already.
Thank you again Lee

Rereading Hyperion's comments.
It was Cimmaron that sold adapters and an option to buy op amps already installed.

I think member Wyn Palmer actually designed op amps ( retired).
He'd probably be a good source of info in finding possible current production op amps for certain applications.
 
Last edited:
There are very few operational amplifiers that can handle supply voltages beyond +/-22V. I tried to compare some and put them in a table (see attached), but some are not suitable for audio...
 

Attachments

So, I changed one of the previously fitted OPA2604AP's (the one in the 'Control Amp') with a pair of OPA604AP's tonight - it's playing now, it hasn't lost any of its sparkle - sounds just the same. The replacement OP amps run quite a lot cooler, no longer worryingly hot as it was with the one package dual OP amps. ;)
I am very pleased. :banana:
 
Last edited:
Right behind you, John. I ordered a couple from Cimarron, they haven't arrived yet. I took my amp's temperature on the top of the face plate and I could tell that there was a warmer spot above where the DIP OPA2604 sits beneath/behind the face plate.

Question: apologies that this is not in keeping with the thread topic, when you replaced your AUD-11II power supply diodes, what did you replace them with?
 
You must log in or register to reply here.
Back
Top Bottom