OPA2604AU replaceable with a TL072?

[Techknowman]

I have and old Crest C9 that im repairing one channel on.
The unit has 2 OPA2604AU and I was just going to buy a few two have around but mouser is 6.44 each (WOW) :*
This just a club amp and the extra, extra clean price is not nessisary can't it be replaced with a TL072?
Both pin the same, The TL will handel higher voltage.
the only difference that I see is .0003% distortion versus .003%. in a transister based brute amp like the C9 you would never hear it much less be able to measure it.
Am I way of here?

[J M Fahey]

If I were designing a new amp, would go straight for the TL072, being a minimalist who loves using *good* available-anywhere parts, *but* repairing a commercial amp, if at all possible would stick to what they specify.
Not because of distortion but of stability, phase shifts, open loop gain, that sort of things.
The amp might be compensated using that specific Op Amp, considering its particular quirks.
Any time wasted re-tweaking it would not compensate the $6 difference.
*Now*, having said that, try a TL072 there and check whether itīs that critical or not.
Maybe it works perfectly.
Only experience will tell.
Good luck.

[soundguruman]

They seem very similar, but I think the 72 has a higher output voltage maximum rating, anyhow do it, I think it would work.
There is 2 zeener diodes located near the output heat sink, replace those. I don't have the schematic in front of me but I remember that those go bad. I think they are 1.2 volt, or some similar voltage...
anyhow they go bad and the amp stops working.
Overall a pretty good amp and for sure worth fixing. I like these amps, aside from those diodes going out. Nice beefy design...just use 5 watt diodes...

[Techknowman]

There is 2 zeener diodes located near the output heat sink, replace those. I don't have the schematic in front of me but I remember that those go bad. I think they are 1.2 volt, or some similar voltage... anyhow they go bad and the amp stops working. Overall a pretty good amp and for sure worth fixing. I like these amps, aside from those diodes going out. Nice beefy design...just use 5 watt diodes...

The heatsinks on this thing is long. Hard to know which ones you are talking about.. But i'll look, thanks.

[Steve Conner]

I agree that the 2604's distortion spec doesn't matter, but it might also have a higher output current capability. IIRC, it's rated to drive a 600 ohm load but the TL072 isn't.

That could be important if it was being used as some kind of driver.

[soundguruman]

Sorry I don't have the schematic, but pretty sure they are both 1.2volt
They get a tiny bit of reverse leakage and the protection circuit won't turn on.

[km6xz]

Another good example of why the internet is dangerous. A claim that an unseen, untested amp needs diodes replaced when there is no evidence of a problem in this one amp or a general circuit design defect affecting all amps, no service bulletins from Crest or rash of defective diodes in C9s is irresponsible and intentionally misleading.

I was a Crest warranty station for years and never saw a problem with those diodes. Replace them if they are not meeting spec but there is only a remote chance that they are defective.

The IC the amp was designed for is suitable for a number of desirable characteristics, analyze the circuit to determine whether those feature are needed. But that said, the OP2604 has twice the slew rate, twice the gain bandwidth product, lower input bias, lower output Z/higher current, and other reasons it was selected. Your statement that you can never near the difference in .03 and .0003 is probably true but in the amp, you CAN hear when an amp is slew limited or when the TL072 is called on to drive lower Z or reactive loads, which comes into play when the feedback is generated by a speaker. If it was my amp, I would prefer that the tech maintained the performance I originally paid for.

[soundguruman]

The diodes go bad and the amp stops working. I have seen it happen several times. I was one of the first Crest service centers, more than 20 years ago. So What? I want to replace them to ensure reliability.
Nothing dangerous, just preventative maintenance, to prevent a known problem from occurring. Accurate and quite conscientious, I would say.
Put in a TLO 72, replace the diodes, So What? "irresponsible and intentionally misleading" Bull. Pure Sensationalism I would say.
You can sit around all day measuring slew rate and buying $6 Ics, we would rather just USE our amplifiers.

AS A MATTER FACT, as I recall-- it was the Crest service adviser, at the Crest factory, who first pointed out that these zeeners were failing, and advised me to replace them routinely, to ensure reliability. And so, I invest $2 in preventative maintenance, SO WHAT???
And I point this out to other Crest owners and repair persons...SO WHAT???

You can stick in the TLO 72 and the thing will work for YEARS. I am still waiting for the "dangerous" and "irresponsible" to occur. I'm still waiting...

[km6xz]

Yes, it is a continuation of a long series of misleading and just wrong claims you make on the forum. The diodes were never a problem more than random statistical failings, like every other part on the board. So calling those out for preventative maintenance is misleading and irresponsible.

Regarding slew rate limiting, what you you say that impacts stability, overload recovery and other very obvious performance characteristics of the total amp? At low levels, into non-reactive loads not so much of a problem. High rates, at high voltage swings, that the 2406 is used for in that circuit, the effect can be heard and measured in normal application. A circuit that has more post diff amp voltage gain can get by with the TL072 without a problem and perform well but there is little voltage gain in that circuit before the current amps. Amps that are often driven to clipping, recovery is important and the front end of the C9 was very good for that and could handle fast rise time square waves, like are found in high level pa systems often.
Slew rate limiting is most heard as the breaking glass sound in crash cymbals and vocals. It is irritating to listen to.

[soundguruman]

Note to Crest owners: Send all broken Crest amps to St. Petersburg, Russia.
Our American repair shops are just too irresponsible and dangerous to fix them. The sound of breaking glass is too irritating, without $6 op amps installed.

[Jazz P Bass]

Back to the subject.
Here is the schematic.
Page 5. Input board.

[km6xz]

Don't you think that is pretty immature to be upset because someone made serious suggestion that does impact sound in real world situations? I am all for using the cheapest parts that will provide the needed performance including the TL072 if it did meets the requirements of the circuit. In this case it does not. It has nothing to do with cost, but all to do with operating characteristics.

There are lots of places were TL072 are fine, used within their design spec that are one of the most useful general purpose devices out there. The company did not want to splurge but the 2406 provided the characteristics that met the design topography and goal and was used for good reason.

There is no need to send the amp to me 7k miles away, lots of bench techs back home in the US know how to get the amp back to meeting advertised spec. I know several there in California whom I trained.

[J M Fahey]

Note to Crest owners: Send all broken Crest amps to St. Petersburg, Russia.
Our American repair shops are just too irresponsible and dangerous to fix them.

Weelll, I would not generalize.
Maybe there is*just one*

irresponsible and dangerous

service shop to be avoided

Another doubt: you claim direct contact with

the Crest service adviser, at the Crest factory

,
yet they still Factory used OP2604 instead of TL072.
*Maybe* they had a reason to ignore you after all ?

[Techknowman]

I was in touch with crest on this amp at the beginning.
The CA9 i have is something of a hybrid. the amp(s) are like most CA9's but acording to Crest, and the serial number it should have had an older preamp board that was not surface mount however this one does. So i had to get a different schematic from them.
and one channel works fine. the other was damaged when a liquid or conductive element bridged across two of the pins between the preamp and amp. It burned the board and one 5.1ohm 3w resister. I repaired the PCB and replaced the parts. the amp sections both work when pluged into the "good" preamp channel so just fixing that.
All i got from Crest was schematics, no service manual or info on biasing , test etc.

[diagrammatiks]

if they designed it around an opa2604 you should use a 2604.

if they could have gotten away with using a tl072 they would have.

I'd just replace both with TLE2072s and call it a day.