I have this amp that uses 2 NE5532 and 1 TL072 in the output
the 2 NE5532 are signetics, and the TL072 is a TI
Should I change those with newer op amps ?
The amp is a Hafler Pro5000
see schematics
the 2 NE5532 are signetics, and the TL072 is a TI
Should I change those with newer op amps ?
The amp is a Hafler Pro5000
see schematics
I have been looking for ne5532 alternative for quite awhile. I have not found anything meets all the specs I want yet.
NE5532 specs:
1. ne5532 supports up to +-22V power supply. You can easily build a 20W amp with +-22V.
2. 10MHz bandwidth.
3. 38ma output current.
4. It is cheap.
NE5532 specs:
1. ne5532 supports up to +-22V power supply. You can easily build a 20W amp with +-22V.
2. 10MHz bandwidth.
3. 38ma output current.
4. It is cheap.
5532 use more supply current than newer IC's but since the power supply was designed to handle it, no sonic problem.
TL072 are a bit noisy, I have had good results with ST33078 and NJM4580 , but TL072 is fet input and the others are bipolar, so in some circuits that will not work.
Note op amps with faster slew rate can oscillate, so beware. Replacing 4558 with 33078 on a 50x RIAA input, I had a ~ 1 mhz oscillation. I had to add .1 uf ceramics within an inch of the ICs (one .1 for two IC's). and put 33 pf ceramic capacitors around the feed back resistor to kill HF gain. But the hiss was waay down, definitely worth the effort.
TL072 are a bit noisy, I have had good results with ST33078 and NJM4580 , but TL072 is fet input and the others are bipolar, so in some circuits that will not work.
Note op amps with faster slew rate can oscillate, so beware. Replacing 4558 with 33078 on a 50x RIAA input, I had a ~ 1 mhz oscillation. I had to add .1 uf ceramics within an inch of the ICs (one .1 for two IC's). and put 33 pf ceramic capacitors around the feed back resistor to kill HF gain. But the hiss was waay down, definitely worth the effort.
Last edited:
Perhaps the OPA2604 (for the TL072), would seem to be a reasonable fit. They are obsolete but if you look around carefully you can probably find genuine ones.
Last edited:
Someone just sent me a youtube vid of a guy comparing opamps and trying paralleling new stock 5532 with remarkable results.
Having said that, the signetics are completely different animals from the current 5532 so I don't know how that would work in the specific circuit!
Having said that, the signetics are completely different animals from the current 5532 so I don't know how that would work in the specific circuit!
Have a look at the LME49720. Lower noise, better slew rate, lower quiescent current, good drive capability and it sounds great. But typicallry 3~4x the price of a 5532. There are no free lunches but the 5532 is a real value.
Leave them alone.
Op amp rolling is mostly bull poop, for people who percieve problems when there are none.
Or do tests with an external board, joining the main amp at pre-amp in for trials. Some track cutting and soldering may be needed.
Like @indianajo says, you need a scope amd some expertise, people on the web say just change the op amp and voila!
You have to tweak the circuit for each op amp you try, and in the end, the output stage and its limitations will be also involved.
And the result will not be very different, in my opinion you can do better things with your time.
If you do go ahead, try and find JRC parts, they are said to be better than Signetics / TI, now the company has changed name to Nisshinbo or similar. Again hearsay.
And consider what is the headache to fix your Hafler amp if it is damaged during trials, some parts may be difficult to obtain.
Capacitors, switches, pots, grounding issues, those can also add noise, dry joints at connectors also, so I would do that TLC before any circuit modifications, and you may be surprised how much the sound improves...
Op amp rolling is mostly bull poop, for people who percieve problems when there are none.
Or do tests with an external board, joining the main amp at pre-amp in for trials. Some track cutting and soldering may be needed.
Like @indianajo says, you need a scope amd some expertise, people on the web say just change the op amp and voila!
You have to tweak the circuit for each op amp you try, and in the end, the output stage and its limitations will be also involved.
And the result will not be very different, in my opinion you can do better things with your time.
If you do go ahead, try and find JRC parts, they are said to be better than Signetics / TI, now the company has changed name to Nisshinbo or similar. Again hearsay.
And consider what is the headache to fix your Hafler amp if it is damaged during trials, some parts may be difficult to obtain.
Capacitors, switches, pots, grounding issues, those can also add noise, dry joints at connectors also, so I would do that TLC before any circuit modifications, and you may be surprised how much the sound improves...
None of the scopes I bought worked more than an hour or two. My disco mixer was producing 1.5 vac that was not audible through the power amp & speakers. It was detectable on the 2 vac scale of my simpson VOM. The fact that it would pass through a 390 pf capacitor proved it was ultrasound. I then followed the acceptable procedures to eliminate oscillations in op amps. Voila! for $15 at the flea market + $25 parts and hours of work, I had a RIAA "preamp" which sounded as good as my PAS2 tube preamp which had unpurchasable capacitors that had failed. Real RIAA preamps on the market were several hundred dollars, or **** like the ubiquitous Radio Shack one.
Note to post # 7, better slew rate is not good for audio. As above experience proves. 5532 is about perfect, but was not for sale the day I went looking to purchase low noise op amps. 33078 was.
How about NJM2122? It's a quite recent Opamp with ultra-low noise figure. And it is cheap and easy to find.
https://www.nisshinbo-microdevices.co.jp/en/products/operational-amplifier/spec/?product=njm2122
https://www.nisshinbo-microdevices.co.jp/en/products/operational-amplifier/spec/?product=njm2122
This post is a mix of true and false statements. I have about 20 different opamps, each one sounds different. Opamp rolling pays off, but with a function generator, an oscilloscope and a little necessary knowledge. Replacing the wiring and passive components also has a big impact on the sound of the device. I'm not saying that it's not dangerous if you're not careful. This amplifier requires special care, an autotransformer for testing, and the like. Opamps in auxiliary circuits that have nothing to do with the signal should not be touched. There are many good modern opamps with fet and bipolar inputs that sound better than the NE5532 and TL072.
Last edited:
I'll disagree with some of your statements NIXIE62.
Wiring changes will only cause problems, and for single wires (no shield) there will be zero difference. A shielded wire has to be spectacularly bad to make a difference by being replaced at audio frequencies. If the wire is that bad, probably the equipment is junk and you're wasting your time.
Components may make a difference, but only in certain circuit locations (and those aren't coupling capacitors). As for op amps, most don't make a difference unless they are unsuited for the application (or they are fakes). You may hear a difference between a mis-usd op amp and the proper type. Not unless you're comparing something really bad like a 4558 to NE5532 in some applications.
The NE5532 is difficult to beat in audio, in high gain situations then an LM4562 / LME49720 will be better. I haven't looked today, but the TL072 may be used in the DC servo. Replacing it shouldn't affect the sound unless the servo isn't well designed. But then you have larger problems to deal with.
I would leave the op amps alone. Bigger gains can be had elsewhere in that amplifier.
Wiring changes will only cause problems, and for single wires (no shield) there will be zero difference. A shielded wire has to be spectacularly bad to make a difference by being replaced at audio frequencies. If the wire is that bad, probably the equipment is junk and you're wasting your time.
Components may make a difference, but only in certain circuit locations (and those aren't coupling capacitors). As for op amps, most don't make a difference unless they are unsuited for the application (or they are fakes). You may hear a difference between a mis-usd op amp and the proper type. Not unless you're comparing something really bad like a 4558 to NE5532 in some applications.
The NE5532 is difficult to beat in audio, in high gain situations then an LM4562 / LME49720 will be better. I haven't looked today, but the TL072 may be used in the DC servo. Replacing it shouldn't affect the sound unless the servo isn't well designed. But then you have larger problems to deal with.
I would leave the op amps alone. Bigger gains can be had elsewhere in that amplifier.
My knowledge is a little different on these issues, after a lot of trials, experiments and built and rebuilt devices. There are always different opinions on this and similar topics. Luckily for us in the DIY community the freedom of choice of components is unlimited. 🙂
Absolutely, it is DIY. You do what you want. There is also information that isn't entirely subjective.
Yes, my comment is about subjective impression. I have some measuring equipment, but the measurements don't explain some things. As for this particular amp, any modifications should be left to an expert, it's not exactly a toy.
A couple of months ago I was building a Whammy headamp and experimenting with opamps. There was a lot of oscillation and the need to adjust values of the compensation capacitors. I see in that thread that people are just poking opamps, which is completely wrong.
A couple of months ago I was building a Whammy headamp and experimenting with opamps. There was a lot of oscillation and the need to adjust values of the compensation capacitors. I see in that thread that people are just poking opamps, which is completely wrong.
Absolutely! Just sticking parts in is ill advised as you noted.
Depending on the equipment you have access to, you may not be able to see the differences. Given really good spectrum type analysers for audio, yes you can measure the differences.
PCB layout is also critical, power supply quality as well. That doesn't mean huge transformers and massive capacitors.
Depending on the equipment you have access to, you may not be able to see the differences. Given really good spectrum type analysers for audio, yes you can measure the differences.
PCB layout is also critical, power supply quality as well. That doesn't mean huge transformers and massive capacitors.
My equipment is modern but modest, I don't do anything professionaly. I have enough to be able to check that everything is working as it should and to be able to try out the devices without much risk of destruction.