Hi
I am thinking of trying opa445's in place of either lm4562's that are in my pre/dac or swapping out some NE5532's that are in my crossover, just to see if and how they may affect the music... however, in checking out the specs of each, I am a little unsure that they align well enough not to cause any issues.
Yes you have guessed, I am a beginner with electronics, if it is broken or I don't mind if it won't work again, have a go at soldering or fixing it, so far okay with minor issues, but my pre/dac and crossover are valuable to me and I do not want to break them! , so thought I had better ask advice! Yes the dip8 sockets in both, allow the original op amp just to be pulled out and the new pushed in 🙂
I am thinking I shall need some of these:
https://www.cimarrontechnology.com/...alled&attribute_pa_bdog_1=4f-nocomponentadded
I can see the bandwidth is somewhat lower on the 445, but am unsure how this will change things...
I am not sure what the current balance is tbh, I know it only looks like a tiny number but the 445 is 1000 x different to the 4562
Once again I am not sure how the V offset will affect things in the cct, or the sound coming out the speakers
As far as I can see if I am using two 445's the current supply with be okay... not too far off
The Current OP bothers me most, as this is less than half, of the op amps already in place...
Many years ago I attempted to swap out the 4562 for a Burson op amp, the supply V was 22V if I remember correctly and the Burson tech support said that their op amps would be okay at 22V but they were not!! I have not yet measured the supply to the 5532! to check the min 20V of the 445 is met.
Any comments much appreciated, I have had a read, to try and understand what the specs mean, but I think I need a very simple explanation, and of course to know if one number differs a little or wildly what could happen :-0
Cheers for any thoughts
I am thinking of trying opa445's in place of either lm4562's that are in my pre/dac or swapping out some NE5532's that are in my crossover, just to see if and how they may affect the music... however, in checking out the specs of each, I am a little unsure that they align well enough not to cause any issues.
Yes you have guessed, I am a beginner with electronics, if it is broken or I don't mind if it won't work again, have a go at soldering or fixing it, so far okay with minor issues, but my pre/dac and crossover are valuable to me and I do not want to break them! , so thought I had better ask advice! Yes the dip8 sockets in both, allow the original op amp just to be pulled out and the new pushed in 🙂
| NE5532 (crossover) | OPA445BM (to try) | LM4562 (pre amp/dac) | |
| slew V/uS | 9 | 15 | 20 |
| gain bandwidth MHz | 10 | 2 | 55 |
| Current IP balance | 10pA | 10nA | |
| V IP offset | 500uV | 1mV | 100uV |
| current supply | 8mA (2ch) | 4.2mA (single) | 10uA (2ch) |
| Current OP | 38mA / ch | 15mA | 26mA / ch |
| V supply min | 6 | 20 | 5 |
| V supply max | 40 | 90 | 34 |
I am thinking I shall need some of these:
https://www.cimarrontechnology.com/...alled&attribute_pa_bdog_1=4f-nocomponentadded
I can see the bandwidth is somewhat lower on the 445, but am unsure how this will change things...
I am not sure what the current balance is tbh, I know it only looks like a tiny number but the 445 is 1000 x different to the 4562
Once again I am not sure how the V offset will affect things in the cct, or the sound coming out the speakers
As far as I can see if I am using two 445's the current supply with be okay... not too far off
The Current OP bothers me most, as this is less than half, of the op amps already in place...
Many years ago I attempted to swap out the 4562 for a Burson op amp, the supply V was 22V if I remember correctly and the Burson tech support said that their op amps would be okay at 22V but they were not!! I have not yet measured the supply to the 5532! to check the min 20V of the 445 is met.
Any comments much appreciated, I have had a read, to try and understand what the specs mean, but I think I need a very simple explanation, and of course to know if one number differs a little or wildly what could happen :-0
Cheers for any thoughts
yeah.... hence the link to the adapter ;-)
https://www.cimarrontechnology.com/...alled&attribute_pa_bdog_1=4f-nocomponentadded
should work, no? haha as long as I get the things the right way around :-0
https://www.cimarrontechnology.com/...alled&attribute_pa_bdog_1=4f-nocomponentadded
should work, no? haha as long as I get the things the right way around :-0
Sorry, missed that. I wouldn't consider the OPA445 especially interesting for audio - it's mainly interesting where you need the high-voltage capability. If you want to try something that might be an upgrade soundwise I would look at the various OPA16xx-types instead. They are all SMD so you need adapters and you need to watch for instability, but soundwise they should be at least be different (and possibly better).
OPA445 strong point is high supply rail voltage. But, it has higher distortion, rising with frequency. I did use it as a driver voltage amplifying stage in a class A amp with output stage based on high power diamond buffer. It was working well, reliably. Depends if you accept distortion rising with frequency. Noise is also higher.
Arr I have soldered a couple of SMD inductors and lost quite a few of them when they pinged out the tweezers! haha 😛 I did not like the way they soldered... or rather I have not had enough practice at SMD and so would by far prefer through hole or push fit when it comes to these two bits of kit as I would not trust my SMD soldering,,,,, although it has seemed to hold up now for about 6 months with the dinky inductors so far! Thanks for the recommendation though I may have a little look!
hmmm so it could work without causing issues to the kit !? it might just sound a bit iffy... if I hear the distortion....
At the moment it does sound quite clean, and maybe could be more musical and involving, I wonder if the distortion / noise may encourage foot tapping..... bit weird ay, aiming for clean detailed, silent background, accurate sound, and then introduce noise and distortion to maybe get foot tapping... crazy, but who knows, if it will go in with out causing issues!? I may give it a go.....
It will be a very strange replacement. If TS wants he can do it of cause, but reasons are unclear. There is nothing interisting or special in opa445 except high voltage capabilities.
🙂 yes my initial excitement of trying something unusual and getting the soldering iron out, has faded a little tbh, if all the parts were local and at least some one said the audio qualities were good, then I may continue in my quest 😛 but the parts are not local and the only good things that are said about it is its high voltage capabilities... so yes I may well listen to more experienced folk! and look somewhere else for a more worth while tinker 🙂 Cheers! 🙂
Firstly the LM4562 is only rated to +/-17V, yet you claim to be running them at +/-22V - not a good idea. The NE5532 isn't rated for +/-22V either (that's its absolute maximum, +/-20V is about the sensible limit for them). Yes Burson's are +/-16.5V IIRC, even less than the LM4562
So I'd suggest if possible fixing your power supply for the LM4562 to the standard +/-15V or so and then you have a much wider range of opamps available, some much less EMI-sensitive than the LM4562.
Several problems with randomly switching out opamps you should be aware of, for instance
1) decoupling requirements differ, some opamps oscillate at RF if not adequately decoupled.
2) output drive limits - the NE5532 can drive down to 500 ohm loads, some opamps can't, so substituting them might lead to clipping or raised distortion.
3) input bias current varies enormously between FET and bipolar opamps, many circuits assume FET opamps and will misbehave with bipolar opamps.
So I'd suggest if possible fixing your power supply for the LM4562 to the standard +/-15V or so and then you have a much wider range of opamps available, some much less EMI-sensitive than the LM4562.
Several problems with randomly switching out opamps you should be aware of, for instance
1) decoupling requirements differ, some opamps oscillate at RF if not adequately decoupled.
2) output drive limits - the NE5532 can drive down to 500 ohm loads, some opamps can't, so substituting them might lead to clipping or raised distortion.
3) input bias current varies enormously between FET and bipolar opamps, many circuits assume FET opamps and will misbehave with bipolar opamps.
Leave them alone, it is not as easy as some claim.
The input and output impedances need to be modified for different chips.
Best get ready made cross overs, and swap them out, must be sold at fairly low prices from the usual sources.
It is very fiddly to swap out SMD parts repeatedly, and the result is usually not a big deal of improvement, again most people will tell you a lot of hype (and lies).
It is mostly a sales pitch to sell you junk.
The input and output impedances need to be modified for different chips.
Best get ready made cross overs, and swap them out, must be sold at fairly low prices from the usual sources.
It is very fiddly to swap out SMD parts repeatedly, and the result is usually not a big deal of improvement, again most people will tell you a lot of hype (and lies).
It is mostly a sales pitch to sell you junk.
I have a socketed NE5532 in my DAC and the best substitute has been LT1364. LM6172 was good and OPA2134 was OK too. LM4562 was percussive and not smooth on voices.
I'd be interested in trying any other DAC or DAC board (populated) with a socketed NE5532 in the output. Do you know of any?
I'd be interested in trying any other DAC or DAC board (populated) with a socketed NE5532 in the output. Do you know of any?
Again, opamp swapping is fun to try BUT you MUST observe some precautions for best results:
- 100nF MLCC X7R caps across rails (pins 4 and 8)
- 22pf COG caps across feedback resistors (pins 1&2, 6&7) to limit bandwidth and curb oscillation tendencies
- MillMax gold-plated sockets
- BiPolar or film caps at input pins, NOT polarized caps.
I'd recommend not using gold plated sockets for standard ICs - IC pins are tin plated, you should never mix gold and tin as that leads to electrolytic corrosion and early failure.
No issue with polarized caps on the inputs - well certainly no issue that affects swapping out opamps.
An issue that can bite is replacing a FET opamp with bipolar and finding the original circuit produces large offsets due to the much higher bias currents on the inputs.
The LM4562 needs much more decoupling than NE5532, BTW - a single 100nF from V+ to V- is not good enough with the LM4562, you have to decouple each rail to ground for stability.
No issue with polarized caps on the inputs - well certainly no issue that affects swapping out opamps.
An issue that can bite is replacing a FET opamp with bipolar and finding the original circuit produces large offsets due to the much higher bias currents on the inputs.
The LM4562 needs much more decoupling than NE5532, BTW - a single 100nF from V+ to V- is not good enough with the LM4562, you have to decouple each rail to ground for stability.