As some of you have read I am in the middle of repairing a Adcom GFP 555 pre in THIS thread. I am at a loss when it comes to op amps. I have a good understanding of what they do but when looking at the specs I get a little confused. I would like to use the Adcom as reference just so I have something tangible to work with and look at. The single channel op amps are labeled LT 8713 the dual channel I don't know off the top of my head.
Questions and assumptions?
Looking at mouser what is the difference between Audio IC and Amplifier IC?
I have read the Adcom GFP 555 has a 22 volt power supply so I am guessing the op amps need to be able to run 22 volts. But some of the recommendations I have read here like THIS one has a rating of 15 volts. To me that doesn't sound right. Then there is THIS one which has a rating of 22 volts but the some of the other specs are different.
It would make sense that over the years the op amps specs have changed because technology has gotten better but which specs do you look at and which ones if any do you ignore?
Questions and assumptions?
Looking at mouser what is the difference between Audio IC and Amplifier IC?
I have read the Adcom GFP 555 has a 22 volt power supply so I am guessing the op amps need to be able to run 22 volts. But some of the recommendations I have read here like THIS one has a rating of 15 volts. To me that doesn't sound right. Then there is THIS one which has a rating of 22 volts but the some of the other specs are different.
It would make sense that over the years the op amps specs have changed because technology has gotten better but which specs do you look at and which ones if any do you ignore?
15V vs +/- 15V
22V vs +/- 22V
What is the supply that is actually used?
To troubleshoot effectively you need to understand how the part is supposed to work.
Read the data sheets for the op-amps used and look at suggested circuits. This will help you learn.
It really helps if you have a schematic of the circuit you are working on. You may even have to draw it out. 🙂
Audio IC / Amplifier IC...all amplify and most (almost all) can be used for audio. "Audio IC" are usually designed for audio performance and have audio use information in the data sheet.
Op-amp specs:
ABSOLUTE MAXIMUM RATINGS are kinda important.
Pin-out and footprint if you want a direct physical drop-in part.
If you are only going to buy one or two buy the best spec one.
Drive capability, distortion, input bias current, output voltage swing (vs frequency) are just a few of the features to look at.
Have fun learning. 🙂
22V vs +/- 22V
What is the supply that is actually used?
To troubleshoot effectively you need to understand how the part is supposed to work.
Read the data sheets for the op-amps used and look at suggested circuits. This will help you learn.
It really helps if you have a schematic of the circuit you are working on. You may even have to draw it out. 🙂
Audio IC / Amplifier IC...all amplify and most (almost all) can be used for audio. "Audio IC" are usually designed for audio performance and have audio use information in the data sheet.
Op-amp specs:
ABSOLUTE MAXIMUM RATINGS are kinda important.
Pin-out and footprint if you want a direct physical drop-in part.
If you are only going to buy one or two buy the best spec one.
Drive capability, distortion, input bias current, output voltage swing (vs frequency) are just a few of the features to look at.
Have fun learning. 🙂
Last edited:
It sounds like you need to start at the beginning, here is a good site for starters: DIY Audio Articles - Scroll down about a half-page to the "Beginners' Luck" section: Designing With Opamps I - What they are, what they do, and how they do it
Mike
Last edited:
What you can do first is measure the DC voltage feeding the present opamp in your preamp. Multimeter probes on pins 4 and 8 of the opamp will tell you if you can use the opamp you have in mind and is the best way to know the actual DC voltage feeding the stock opamp.
Last edited:
You always need to be aware that things don't always work as expected when swapping opamps.
http://www.diyaudio.com/forums/anal...u-have-checked-see-its-stable-havent-you.html
Very few audio oriented opamps are specced for operation at more than -/+18v as an absolute maximum. There are some notable exceptions though. As others have said... we need to see a circuit.
http://www.diyaudio.com/forums/anal...u-have-checked-see-its-stable-havent-you.html
Very few audio oriented opamps are specced for operation at more than -/+18v as an absolute maximum. There are some notable exceptions though. As others have said... we need to see a circuit.
I have attached a copy of the service manual that has a circuit lay out in it.
I have tried looking up the specs on what is currently in there but I am not having any luck. I read some where else that said the original were op37's and I am able to pull up specs on those.
I am at work till 11p tonight so I appreciate the reading material!
I have tried looking up the specs on what is currently in there but I am not having any luck. I read some where else that said the original were op37's and I am able to pull up specs on those.
I am at work till 11p tonight so I appreciate the reading material!
The OP37 is one of the few that can run on -/+22 volt rails.
Just an idea... and not as crazy as it seems would be to consider altering the PSU to give slightly lower rails. The reduction in headroom would only be noticeable by a number printed on a page 😉 and it would give you freedom to use (after consideration as to whether they are suitable) of all the modern popular favourites.
Altering the basic PSU should be easy (just a resistor change in the error amplifier). I spotted a secondary rail fed off the -/+22 volts, and so that would need similar treatment to maintain its specced output of -/+15 volts.
Something to think on 🙂
Just an idea... and not as crazy as it seems would be to consider altering the PSU to give slightly lower rails. The reduction in headroom would only be noticeable by a number printed on a page 😉 and it would give you freedom to use (after consideration as to whether they are suitable) of all the modern popular favourites.
Altering the basic PSU should be easy (just a resistor change in the error amplifier). I spotted a secondary rail fed off the -/+22 volts, and so that would need similar treatment to maintain its specced output of -/+15 volts.
Something to think on 🙂
BUT...I was thinking about that because I think their are jumpers going into the op amps that can be removed and resistors placed there spots. The problem is I don't know what resistance value I would need. I can follow ohms law easy enough but choosing values to plug into the equation has me stumped. That I would still like to learn op amps, haha.
This will be excellent reading material for a good while. Thank you! I think I am trying to learn to much to fast. My brain is going to many directions at once.
Adcom had a service bulletin for the GFP555 II, lowering the voltages in the regulator circuit and replacing the opamps. Not sure if the GFP555 had similar problems.
Craig
Craig
If would help if you post the schematic of the amp. From reading the spec, LME49710 can safely run on +/-15V supplies and OP27 can run on +/-18V. Don't read the +/-22V as that's the absolute max rating. You show the schematic and we should be able to tell what supply voltage of the amplifier.
This is just my observation of the two opamp you posted. From scanning through the datasheet, LME49710 should be more suitable for hifi audio amp application as it said so on the first place. You look at all the graphs on page 6 and 7 telling you the THD ( distortion) specifications. OP27 is very specialized for precision and low noise. We used a lot of this in the past for precision instrumentation design. You don't need precision ( low offset voltage and current etc. ) in audio circuit.
Another thing to look at is slew rate. Opamp with higher slew rate will give you faster attack on the sound when it's called for. LME is 15V/uS vs OP27 only 2.8V/uS. LME is definitely superior in this department.
Then look at the output current. LME spec +/-23mA, OP27 did not spec just say it can swing +/-11V driving 600ohm load. that's less than 19mA. I suspect LME has stronger drive. Stronger drive is better to drive long coax( need series resistor for stability).
All in all, LME sounds a lot better for me. I am sure other expert here can point you to more detail things to look for. What I described are just very basic things, People are talking about what opamp sounds better, warmer etc. This is beyond my knowledge. In fact I am looking at this thread because I am looking for a warmer sounding opamp to replace the LM4562 I am using. I have no idea how to look for this from the datasheet.
Last edited:
Attached is the service manua, including schematic,l for the Adcom GFP-555 pre I am working on.
Based on the schematic, you actually have two different types of op amps with two different types of voltages.
The devices running at +/-22V (i.e. 44V) are IC201 and IC202 are the single channel op amps and appear to be custom Adcom chips, with the dual channel IC203 being an NJM2041.
The one op amp IC204 is an NJM4556 and it runs at +/-15V. You can see that there is another voltage regulation circuit that drops the +/-22V down to +/-15V for this one op amp.
Both NJM have a low 3V/uS slew rate, but they are using bipolar inputs.
Like yungman stated, there really aren't any choices for +/-22V power supply in audio op amps. The +/-18V is pretty much the max usable. There is an OPA445 that allows up to 90V (+/-45V), but no idea on audio performance and it's dual channel only.
The OP27 has a 2V/uS slew and will be slightly slow, but would probably be the closest match for sonic signature. You could try the LF411/LF412 op amps, but you'd need to regulate down to +/-18V to really use them. They have FET inputs which will give you a warmth, but have a higher 15V/uS slew rate.
https://www.digikey.com/product-sea...ee=0&rohs=0&quantity=&ptm=0&fid=0&pageSize=25
The devices running at +/-22V (i.e. 44V) are IC201 and IC202 are the single channel op amps and appear to be custom Adcom chips, with the dual channel IC203 being an NJM2041.
The one op amp IC204 is an NJM4556 and it runs at +/-15V. You can see that there is another voltage regulation circuit that drops the +/-22V down to +/-15V for this one op amp.
Both NJM have a low 3V/uS slew rate, but they are using bipolar inputs.
Like yungman stated, there really aren't any choices for +/-22V power supply in audio op amps. The +/-18V is pretty much the max usable. There is an OPA445 that allows up to 90V (+/-45V), but no idea on audio performance and it's dual channel only.
The OP27 has a 2V/uS slew and will be slightly slow, but would probably be the closest match for sonic signature. You could try the LF411/LF412 op amps, but you'd need to regulate down to +/-18V to really use them. They have FET inputs which will give you a warmth, but have a higher 15V/uS slew rate.
https://www.digikey.com/product-sea...ee=0&rohs=0&quantity=&ptm=0&fid=0&pageSize=25
The OPA604 (dual is OPA2604) is good to 24V on each rail and its specifically targeted at audio. When running at high rails, I recommend only using the '604 as the '2604 gets a lot hotter and much of the advantage of a JFET input is lost at high chip temperatures due to leakage currents at the gates of the FETs.
There is a lot of discussion of how to replace your op amp with something else, and not much on how to determine it is bad.
With no signal, op amps amplifying audio usually have the + input, the - input, and the output, all at the same DC voltage.
I find solder joints, cheapo ceramic bypass caps, and electrolytic caps over 15 years old, are all more likely to fail internally than op amps.
Actual op amps can get blown up in audio amps by DC rail voltage rampaging through the driver boards from the base line of the blown output transistors. These usually read 0 ohms or infinity from the power pins to the inputs, or output.
If you really need a new op amp for +-22 rails, the TI LM675 is readily available for $7. Converting the TO220 package to SIP or DIP is just a soldering problem.
Good hunting your problems.
With no signal, op amps amplifying audio usually have the + input, the - input, and the output, all at the same DC voltage.
I find solder joints, cheapo ceramic bypass caps, and electrolytic caps over 15 years old, are all more likely to fail internally than op amps.
Actual op amps can get blown up in audio amps by DC rail voltage rampaging through the driver boards from the base line of the blown output transistors. These usually read 0 ohms or infinity from the power pins to the inputs, or output.
If you really need a new op amp for +-22 rails, the TI LM675 is readily available for $7. Converting the TO220 package to SIP or DIP is just a soldering problem.
Good hunting your problems.
But if you look at Fig.5 of the datasheet of OPA2604, even at 75deg, it's less than 10nA bias current. Even with 1M input resistance, it's only 10mV. Doesn't seem to be a problem. Bias current affect DC offset in the circuit, but it's not important in audio circuit that very likely to be AC coupled.
75deg for the die would mean (for the SOIC8 package) that the ambient would need to be no higher than 75-54 i.e. 21deg. This is assuming no load being driven at all which wouldn't be a real-world situation.
The offset voltage wouldn't be my primary concern, noise due to input currents is potentially a more pressing issue with high source impedances.
The offset voltage wouldn't be my primary concern, noise due to input currents is potentially a more pressing issue with high source impedances.
10nA is still lower than most of the bipolar opamp. The closest thermal resistance of SOIC8 I found is like 65deg per watt. this IC draws 11mA, say 30V across. That's 0.33W. So the rise of temperature from die to ambient is 0.3X65=21.45deg. So the ambient temp is about 53deg. That's normal.
For current noise, I did the calculation for 10nA at BW=100KHz and resistance of 100K(typical for preamp input impedance.) it's less than 2uVrms. I don't think this is high. But double check my calculation, I am not going to say I am right.
I would run at lower voltage or find a lower power opamp though.
Last edited:
You'd do well to read the thermal resistance from the DS for the OPA2604, that's what I did. Otherwise its just wish-fulfillment.
30V across then no specific reason to be using OPA2604, I introduced it because the context is higher voltage rails.
30V across then no specific reason to be using OPA2604, I introduced it because the context is higher voltage rails.