Q: (better) alternative for OPA627 ?

[Kjelt]

Hi since end of the 90's the OPA627 has been a very good audio grade operational amplifier.
Specs: Very Low Noise: 4.5 nV/vHz at 10 kHz but they are pricy (€40/pc) !
They are still manufactured and available.
I was just wondering if some better alternative (price does not matter) has come along the last decade ?
What is the nr1 opamp choice for ultra low noise audio applications today for new designs or is this still the device to use?

[ebclr]

Change supplier to the right country

https://world.taobao.com/item/527517395690.htm?fromSite=main&ali_refid=a3_430620_1006:1122361873:N:OPA627:e196dc0b44da4c07c059afd5b9628a8b&ali_trackid=1_e196dc0b44da4c07c059afd5b9628a8b&spm=a312a.7700714.0.0.WJhWIK

[ebclr]

http://www.aliexpress.com/af/OPA1662.html?ltype=wholesale&d=y&origin=n&isViewCP=y&spm=2114.30010308.0.59.berxq4&site=glo&SearchText=OPA1662&g=y&SortType=price_asc&initiative_id=SB_20160523015819&isAffiliate=y&shipCountry=br


http://www.ti.com/lit/ds/symlink/opa1662.pdf

Regards

[wraper]

Change supplier to the right country

https://world.taobao.com/item/527517395690.htm?fromSite=main&ali_refid=a3_430620_1006:1122361873:N:OPA627:e196dc0b44da4c07c059afd5b9628a8b&ali_trackid=1_e196dc0b44da4c07c059afd5b9628a8b&spm=a312a.7700714.0.0.WJhWIK

LOL, and get crappy fake opamp.
http://zeptobars.com/en/read/OPA627-AD744-real-vs-fake-china-ebay
http://zeptobars.com/en/read/BB-TI-OPA627-opamp-genuine-fake

[Kjelt]

http://www.ti.com/lit/ds/symlink/opa1662.pdf

Thanks for the opa1662 , will check it out. It does not have the offset trim inputs, are the new ones that good they don't need them anymore or whats the story?

[Marco]

Why would you care about offset for audio?

[Kjelt]

Why would you care about offset for audio?

I don't know, I am not an analog expert, I just saw that this is a difference and wondered if it mattered.
Also the 1662 contains two opamps, does that have implications like crosstalk or mutual influence ? Probably also not that big deal in the audio frequencyband.

[Kjelt]

There are many Chinese distributors can give you very low prices than from the western because we believe the cheaper the more competitive, HOWEVER, no one can get goods less than manufacturer's price, so if the price is lower than TI's high volume MSRP (~50%~80% of 1kpcs price), then it MUST be fake.
OPA627 sells for $12.86 in its cheapest package, even at mass quantity it sells for only 50% its 1k price, that's $6.43, or ~42 CNY, anything brand new and cheaper than this price is fake.

Agree, only exception is  overstock or overdate , when a manufacturer finished production and some partly used reels are left they can be very cheap, also when the expiry date on the reel is nearby or passed they can become very cheap since no respectable manufacturer will take a chance that too high part of the production will have problems. For hobbieists this can be real good usable sources of components.

[Kjelt]

No indeed for some of the prices you see at those chinese sites that is almost giving them away and should trigger the "fake" alarm with anybody who is serious.

[Kleinstein]

It depends on the circuit you use the OP. In some circuits even the LM358 could be better. If lowest noise is needed, LT1028 / LT1115 / AD797 could be an option.

[Kjelt]

It depends on the circuit you use the OP. In some circuits even the LM358 could be better. If lowest noise is needed, LT1028 / LT1115 / AD797 could be an option.

I wanted to use it as an audio buffer or I->V converter after a DAC.

[Cerebus]

The Analog Devices ADA4627 is essentially a second source version of the OPA627. £4.64 GBP / €7,06 one off from RS.

[splin]

The LME49990 is a very low noise, ultra low distortion audio opamp and relatively cheap. Unfortunately it's another victim of TI retiring one of its National Instrument fabs, terminating many fine devices. You could stockpile - not sure if they have ceased production yet.

Alternatively the NE5532/NE5534 are fine low cost, low noise amps used extensively in audio. This is well worth reading:

http://nwavguy.blogspot.co.uk/2011/08/op-amps-myths-facts.html

[The Soulman]

TLE2071, OPA604, could be better could be worse, it depends.

[nctnico]

The venerable TL074 also comes to mind. On paper more noisier but the parts around it can also add noise.

[Zeranin]

Hi since end of the 90's the OPA627 has been a very good audio grade operational amplifier.
Specs: Very Low Noise: 4.5 nV/vHz at 10 kHz but they are pricy (€40/pc) !
They are still manufactured and available.
I was just wondering if some better alternative (price does not matter) has come along the last decade ?
What is the nr1 opamp choice for ultra low noise audio applications today for new designs or is this still the device to use?

I'm fond of the OPA627, though not convinced that it's strengths (low Ib and current noise combined with fairly low voltage noise) are usually required in audio applications.

I once bought some off the net because my normal supplier was out of stock, and they were fakes. I would recommend buying only from well-known, reputable suppliers, and any price that seems too good to be true almost certainly is. 

[BarsMonster]

Alternatively the NE5532/NE5534 are fine low cost, low noise amps used extensively in audio. This is well worth reading:

Early warning: there are lots of fake (!!!) NE5532 on ebay. We'll feature that soon on the site.

[Cerebus]

Alternatively the NE5532/NE5534 are fine low cost, low noise amps used extensively in audio. This is well worth reading:

Early warning: there are lots of fake (!!!) NE5532 on ebay. We'll feature that soon on the site.

Unless it's from a known and trusted supplier it would be wise for one to assume that from ebay, aliexpress etcetera that any op-amp that's in any way 'premium' is a fake. The stepping point from probably good to probably fake is somewhere just above LM358. Incredibly I've got some TL072s that are fakes, with obvious 'blacktoping', parts that I really didn't think it was worth anyone's time to fake.

[ebclr]

To good to be true.

Try before label as fake...

Many reasons to find chips undervalued, like excess inventory, end of production, company bankruptcy and many other ones.

Fake or not fake is not the matter the matter is work or not.

By experience all so called fakes works fine, and at least 80% do exactly as the datasheet describes, You are fool enough to pay 20 times more for something to be real not fake etc etc it's up to you.

Ei incumbit probatio qui dicit, non qui negat    even for ic's this can be true

[Cerebus]

To good to be true.

Try before label as fake...

Many reasons to find chips undervalued, like excess inventory, end of production, company bankruptcy and many other ones.

Fake or not fake is not the matter the matter is work or not.

By experience all so called fakes works fine, and at least 80% do exactly as the datasheet describes, You are fool enough to pay 20 times more for something to be real not fake etc etc it's up to you.

Ei incumbit probatio qui dicit, non qui negat    even for ic's this can be true

The number of sources of low cost parts for the legitimate reasons you've listed are dwarfed by the number of fakes. I do buy lost cost parts that are low cost for legitimate reasons from British and Polish suppliers that I've used before. I have a, thankfully small, collection of fake parts from trial purchases made from, lets say, 'countries in the Pacific Rim'.

If you've had fake parts where "80% do exactly as the datasheet describes", what do you do about the 20%. Solder them into a circuit and pray? Do 100% incoming component inspection on them? The latter, if you're equipped to do it, rather takes away the cost advantage of grey market parts.

Plus the usual fundamental arguments about fakes apply: you're taking money away from the legitimate developers of the part, and hence stunting future R&D, and funding, frankly, criminals. Your application might be non-critical, but god help us if some of those "80% OK" parts find their way into a safety critical system. etc. etc.

What about Caveat Emptor, rather more germane I'd have thought than a Latin tag about criminal law.

[wraper]

Fake or not fake is not the matter the matter is work or not.

I have read like this in one audio forum  . "If it sounds good, does not matter if it is fake" 
Why the hell would you buy some unknown cheap opamp labeled as more expensive stuff, if you can just buy decent cheaper opamp without any lottery.

Many reasons to find chips undervalued, like excess inventory, end of production, company bankruptcy and many other ones.

Rarely the case with China, usually you just get cheap crap. Yeah, they work, but not as well as genuine ones, and not up to the datasheet spec. Probably this will remain unnoticed in audio applications, but if they go in some kind of precise equipment, can cause tons of weird problems.

[zlymex]

For my own use/test, I buy reclaimed ICs with original marking and original leads. They are very cheap and genuine most of the time.

[Zeranin]

To good to be true.

Try before label as fake...

Many reasons to find chips undervalued, like excess inventory, end of production, company bankruptcy and many other ones.

Fake or not fake is not the matter the matter is work or not.

By experience all so called fakes works fine, and at least 80% do exactly as the datasheet describes, You are fool enough to pay 20 times more for something to be real not fake etc etc it's up to you.

Ei incumbit probatio qui dicit, non qui negat    even for ic's this can be true

What on earth does 'works fine' mean?? Sure, the fake OPA627's that I bought 'worked', but they did not meet the specs for an OPA627, so were no use in any critical application that required the OPA627 specs, so I threw them in the rubbish bin where they belong.

If all you want is an op amp that 'works', then you are much better off just to buy a cheaper op amp in the first place.

If your demanding application requires an expensive, high-performance op amp, then fakes are also of no use.

Summary, fakes are not of any use to anyone.

What exactly is a 'fake' IC? I always assumed that it was a re-badged product, and done as a 'backyard' industry.

One cannot literally make an IC, and especially a high-performace state-of-art IC, in a backyard, or even in a small tinpot factory. Is there even such a thing as a 'fake' IC made in some no-name, large factory in China or similar? If any company and factory has the engineers and expertise to actually make IC's, then why would they need or want to make 'fake' ones anyway?

I tend to think that fake ICs are simply backyard rebadged ICs. Does anyone have evidence of fake ICs being manufactured, rather than re-badged?

[zlymex]

.....
I tend to think that fake ICs are simply backyard rebadged ICs. Does anyone have evidence of fake ICs being manufactured, rather than re-badged?

That's right, they all from underground. There are actually three types of fake ICs:
 - retreated. they are the same ICs(could be different suffix), re-tinned, re-marked so that they look new.
 - shoddy, functionally are the same but vast and cheaper to get. For instance use uA741 for OP07
 - fake, they use arbitrary ICs of the same package, re-mark anything they want.
I've encountered all three.

[wraper]

Does anyone have evidence of fake ICs being manufactured, rather than re-badged?

FTDI FT232RL, Nordic nRF24L01
As of opamps, IMO in 95+% of the cases they are just rebadged.

[Kjelt]

And there is this chinese company that makes reverse engineered STM32's.
An sich nothing wrong with that (if they had a valid license or agreement which they don't) but as long as no-one removes the Chinese markings and re-brands them as original ST people know what they are buying.
However in China the latter it seems is not a question of IF but WHEN.

https://olimex.wordpress.com/2015/11/09/chinese-clones-attack-stm32-microcontrollers/

[wraper]

And there is this chinese company that makes reverse engineered STM32's.
An sich nothing wrong with that (if they had a valid license or agreement which they don't) but as long as no-one removes the Chinese markings and re-brands them as original ST people know what they are buying.
However in China the latter it seems is not a question of IF but WHEN.

To be frank, they are compatible but not exactly clones. Also they have higher per clock performance than STM32 MCUs. They don't need to clone actually, buy the licence from ARM https://community.arm.com/community/arm-partner-directory/partner-gigadevice-semiconductor, throw in compatible peripherals and here you go. AFAIK they cannot be rebadged unnoticed either, because firmware programming will reveal it.

GigaDevice GD32 series of MCU got the trade mark in world wide such as US, Korea& Japan

https://olimex.wordpress.com/2015/11/09/chinese-clones-attack-stm32-microcontrollers/

https://olimex.wordpress.com/tag/gd32/

[ebclr]

China handle 80% of the electronic market, just because parts come from china, does not mean this parts are sold by criminals, I went in China several times and never have a dime stolen, when i went in London They hit my wallet on my 1st day there.

Have a lot of good people in China, also some bad guys as in any place it's not true that everything that comes from China is Fake and from criminal origin, this a label easy to make for group of people that can't handle to compete with those guys on a price performance basys.

China have a Hugh network of honesty business. Even in china you get what you paid form this means if you want 1st class pay 1st class, But can assure you 1st class in China is cheaper that 3rd class in Europe.

[wraper]

China handle 80% of the electronic market, just because parts come from china, does not mean this parts are sold by criminals

If you bay opamps at <20% of the cost they normally cost, you ARE buying counterfeits from the criminals. Same goes for mobile phone chargers. You have not so high chance getting genuine branded charger from china even if paying reasonable price and zero chance if buying very cheap. Very low chance to buy genuine memory cards either.

[Cerebus]

If you bay opamps at <20% of the cost they normally cost, you ARE buying counterfeits from the criminals.

Not necessarily. As a previous poster has pointed out there are many good reasons why parts can be discounted. It's not the price, it's not the country, it's the supplier that determines whether a part can be considered likely to be fake. Prices, as of writing, from some reliable suppliers for an identical part LT1013DDR:

Farnell, (Stock number 1623220), price each £2.129 (1-10 parts), £1.846 (11-100 parts)
Digikey, (Stock number 296-14615-1-ND) price each £1.36 (1- 2500 parts)
RS, (Stock number 660-9681), price each £0.366 (5 - 20 parts, no 1 off price available)

By your logic RS with a 5 off price 17.2% that of Farnell and 26.9% that of Digikey must be criminals selling counterfeits when I think we all know that's not true.

[wraper]

I will, on the other hand, be surprised if you got a fake card if you pay Amazon price on eBay.

Actually quiet easy to get a fake at genuine price, it won't be like 2 GB one which appears as 64 GB but it won't be from the brand written on it either, just some cheap slow crap. You can even get counterfeit Mitutoyo calipers from non Chinese sellers at the price of genuine.

[wraper]

If you bay opamps at <20% of the cost they normally cost, you ARE buying counterfeits from the criminals.

Not necessarily. As a previous poster has pointed out there are many good reasons why parts can be discounted. It's not the price, it's not the country, it's the supplier that determines whether a part can be considered likely to be fake. Prices, as of writing, from some reliable suppliers for an identical part LT1013DDR:

Farnell, (Stock number 1623220), price each £2.129 (1-10 parts), £1.846 (11-100 parts)
Digikey, (Stock number 296-14615-1-ND) price each £1.36 (1- 2500 parts)
RS, (Stock number 660-9681), price each £0.366 (5 - 20 parts, no 1 off price available)

By your logic RS with a 5 off price 17.2% that of Farnell and 26.9% that of Digikey must be criminals selling counterfeits when I think we all know that's not true.

By normal price I mean one of the cheapest prices from distributors. Not average among them because there is always some the price bloated multiple times. As of RS part number, they get rid of the last 5 pcs in their stock. Do you think all of those sellers in Chine get rid of the last pieces they have by selling the part for the same cheap price for years? Heck I even got fake TB6560AHs for not ridiculously cheap, which turned out being remarked TA8435H. They are both stepper motor drivers but not compatible at all. It would blow up with rest of the circuit if I hadn't measured the pins before soldering them in. Opamps get counterfeited because it is very easy to do (tons of opamps with the same pinout) and difficult to check.

[wraper]

So far I did not have the "luck" to get genuine priced fake things. I only shop at big sellers that has thousands of feedbacks, or shops with reputation among the community.

Who knows, you might as well just bought fake memory cards but just didn't notice it. It is not easy to notice, especially if this in not something that you extensively check for.

For extreme cases, Dave even got fake parts from Digikey, but similar cases are rare.

He got SMD resistors in the cut tape which were out of spec. More likely they were wrong parts in the bag with the right label, rather than counterfeits. If he got a full reel, that would be a different story.
Read the negative reviews http://www.amazon.com/Samsung-Memory-Micro-UHS-I-Class/product-reviews/B00MNT86NE/ref=cm_cr_arp_d_hist_5?ie=UTF8&filterByStar=five_star&showViewpoints=0&pageNumber=1
They bought at genuine price.

[wraper]

Bunnie investigates microSD memory cards:
http://www.bunniestudios.com/blog/?p=918

[Cerebus]

As of RS part number, they get rid of the last 5 pcs in their stock.

No they don't. I bought some the other day - which is how I know they have a good price at the moment - and it said "5 pieces in stock" both before and after my purchase. Just a quirk of their systems.

I don't know about you but I trust people, not prices. The point I'm making, which you seem to be determined to ignore, is that it is the individual supplier that is the crux of honesty not the whole country or people of a particular ethnic group. Please remember that it was me who issued this warning earlier in the thread:

Unless it's from a known and trusted supplier it would be wise for one to assume that from ebay, aliexpress etcetera that any op-amp that's in any way 'premium' is a fake. The stepping point from probably good to probably fake is somewhere just above LM358. Incredibly I've got some TL072s that are fakes, with obvious 'blacktoping', parts that I really didn't think it was worth anyone's time to fake.

Note the qualification, "Known and trusted supplier", and the suspect group "ebay, aliexpress etcetera".

And I don't like the crypto-racism that's creeping in here. You can't just pick on the poor old Chinese and blame them for the behaviour of some of their tribe. I had a Latvian family living in the downstairs maisonette here for six months and after they left the police came around looking for them in connection with the fraudulent purchase and sale of some cars. Should I base my judgement of all Latvians on them? Of course not.

[wraper]

And I don't like the crypto-racism that's creeping in here. You can't just pick on the poor old Chinese and blame them for the behaviour of some of their tribe. I had a Latvian family living in the downstairs maisonette here for six months and after they left the police came around looking for them in connection with the fraudulent purchase and sale of some cars. Should I base my judgement of all Latvians on them? Of course not.

Franly usually we export not the best people.

[wraper]

As of RS part number, they get rid of the last 5 pcs in their stock.

No they don't. I bought some the other day - which is how I know they have a good price at the moment - and it said "5 pieces in stock" both before and after my purchase. Just a quirk of their systems.

Either way something is
off about that price, they sell reel of 2500 for more than that.
http://uk.rs-online.com/web/p/operational-amplifiers/9233987/

[Cerebus]

As of RS part number, they get rid of the last 5 pcs in their stock.

No they don't. I bought some the other day - which is how I know they have a good price at the moment - and it said "5 pieces in stock" both before and after my purchase. Just a quirk of their systems.

Either way something is
off about that price, they sell reel of 2500 for more than that.
http://uk.rs-online.com/web/p/operational-amplifiers/9233987/

You can often find weird counter-intuitive pricing with RS. Identical parts with identical quantity and reeling requirements with different stock numbers and prices; industrial temp range parts cheaper then commercial temp range etc. etc. I've given up trying to figure it out and just take advantage of it when it crops up. Hence I ended up with LT1013Ds in the industrial temp range when commercial would have done me quite nicely but would have cost more.

[Kleinstein]

With SMT chips one could also get old chips if they are cheap. Old chips might have to much humidity and thus could turn into popcorn in some production lines.

After drying they can be OK for hobby use and hand soldering.

I would consider this a major reason to find discounted chips, that are not fakes - they are just overstock and limited shelf life.