NE5532 of unknown origin
Many of the recent post forget the context of the 2011 original post. If you are like many DIYers, the chance of you ordering a DAC, preamp, line amp, crossover, headphone amp etc. from China through the bay or aliexpress is very high. The chance of getting fake active component, such as NE5532 from unknown foundry is also very high. The genuine NE5532 is a good choice for many audio applications. It has good bandwidth, slew rate and reasonable low noise to meet most challenge. But the performance of a fake NE5532 can be anything. Replacing the fake NE5532 is necessary to get good result. In most case, the rest of the PCB, passive components and workmanship are acceptable.
I ordered a bunch of LME49720 from DigiKey to be on hand. It is a lower noise, higher slew rate substitute for NE5532. Stability is NOT a problem in all case I did the substitution.
If you buy your opamp IC also from China, using fake part replacing fake part is not a smart move. Newark, DigiKey and Mouser are US electronic distributor who are willing to sell in smaller quantity.
When you receive the circuit board, if you see laser engraved marking on the IC, it is likely to be genuine. If the marking is a white paint stamp or no marking, you are most likely getting fake parts.
Many of the recent post forget the context of the 2011 original post. If you are like many DIYers, the chance of you ordering a DAC, preamp, line amp, crossover, headphone amp etc. from China through the bay or aliexpress is very high. The chance of getting fake active component, such as NE5532 from unknown foundry is also very high. The genuine NE5532 is a good choice for many audio applications. It has good bandwidth, slew rate and reasonable low noise to meet most challenge. But the performance of a fake NE5532 can be anything. Replacing the fake NE5532 is necessary to get good result. In most case, the rest of the PCB, passive components and workmanship are acceptable.
I ordered a bunch of LME49720 from DigiKey to be on hand. It is a lower noise, higher slew rate substitute for NE5532. Stability is NOT a problem in all case I did the substitution.
If you buy your opamp IC also from China, using fake part replacing fake part is not a smart move. Newark, DigiKey and Mouser are US electronic distributor who are willing to sell in smaller quantity.
When you receive the circuit board, if you see laser engraved marking on the IC, it is likely to be genuine. If the marking is a white paint stamp or no marking, you are most likely getting fake parts.
Last edited:
No sure what you are referring to when you said "fakes for a 5532 are extremely rare".
I agree. The genuine NE5532 in DIP-8 is $1.01 each in single quantity from Mouser. It is a good drop in replacement for the fake NE5532 when you buy a DAC on ebay. See the first post on this thread. But the LM4562 and LME49720 are better general purpose choices for just a little more.
Last edited:
When you add the total cost of your project BOM, using LME49720 instead of NE5532 may add 2%, 5% or 10% of your project cost? Not a big deal for DIYer. I am not talking about commercial manufacturing.
NE5532 is totally adequate in most audio project. Replacement is needed only when you worry about fake IC.
No one is going to fake a $1 chip - its not worth it. like money fakers - thery don't fake $1` bills - they do $50 and $100 bills.
I stupidlu bought a DAC chip (ESS9023 IIRC) on eBay when I was living in Taiwan. I complained to eBay and quickly got the right part. The supplier of course was just trying his luck and must have thought I was a noob.
My recommendation is that you never, ever buy any semiconductors from eBay. Always buy from reputable US and European distributors (Mouser, Digikey, RS Components and Reichelt are who I use). Not cheap - but quality is always good and all products are branded goods from world class suppliers.
I stupidlu bought a DAC chip (ESS9023 IIRC) on eBay when I was living in Taiwan. I complained to eBay and quickly got the right part. The supplier of course was just trying his luck and must have thought I was a noob.
My recommendation is that you never, ever buy any semiconductors from eBay. Always buy from reputable US and European distributors (Mouser, Digikey, RS Components and Reichelt are who I use). Not cheap - but quality is always good and all products are branded goods from world class suppliers.
I had an amp full off 5532 and 4558's and replaced them with 2068's and the sound was improved 1000%. They best part they don't oscillate
Well, most any of the newer opamps will outperform a crappy 4558. But the old 5532 still holds its own. I will say, though, at 60¢ each, the 2068s are quite the bargain!
Last edited:
I have a AK4490 DAC with a 5523 in the output. What's the current favourite substitution? I see a lot of recommendations for LME49720, at a max of 17v and well shielded. Also OPA2604. Both supposed to be stable. Do you recommend these or something else? Need advice since I usually build with tubes.
Attachments
Last edited:
Nothing wrong with NE553x.LM would be fine. You get lower distortion. A problem I see are the SMD resistors and caps. They produce a lot of distortion. WIMA ist here the solution. Don't use the OPA2604. It has much more distortion than LM.
LM4562 = LME49720. They are the same chip exactly. These have much higher bandwidth and may misbehave where ann LM5532 was perfectly happy.
Nothing wrong with the NE5532, and yes. The surface mount components can be a much greater contributor to distortion. You can buy the right surface mount components, they aren't all bad ones. Even the PCB layout can mess things up. That you cannot change.
Make certain that there is enough power supply decoupling capacitors in there, be very careful which ground trace you use. (when is ground, not ground?)
-Chris
Nothing wrong with the NE5532, and yes. The surface mount components can be a much greater contributor to distortion. You can buy the right surface mount components, they aren't all bad ones. Even the PCB layout can mess things up. That you cannot change.
Make certain that there is enough power supply decoupling capacitors in there, be very careful which ground trace you use. (when is ground, not ground?)
-Chris
The NJM2068 doesn't have a current noise spec, its hardly encouraging when that's a key performance number for many opamp circuits. It may be fine, it may be 20 times worse than a 5532, has anyone measured this? And 5532's are about the cheapest opamps anyway for audio, I bought 100 for £35 from a reputable supplier - cheaper than many passives. You must decouple every 5532 with a 100nF ceramic between the rails right at the chip - but those caps cost next to nothing in SMT.
If you are using SMT resistors use thin-film, not thick-film for ones where distortion matters. And PPS film capacitors are low distortion and reasonably cheap and obtainable as SMDs. There are no polypropylene surface mount caps because it melts before solder does. PPS are high temperature stable.
Not sure anything comes close to the 5532 still for the combination of low voltage and current noise, low distortion at high output drive and low cost still - one day it will be bettered for this combined feature set, but ATM its in a great place.
If you are using SMT resistors use thin-film, not thick-film for ones where distortion matters. And PPS film capacitors are low distortion and reasonably cheap and obtainable as SMDs. There are no polypropylene surface mount caps because it melts before solder does. PPS are high temperature stable.
Not sure anything comes close to the 5532 still for the combination of low voltage and current noise, low distortion at high output drive and low cost still - one day it will be bettered for this combined feature set, but ATM its in a great place.
Last edited: