3Peak Opamps

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Kent, Joe., analogdigital, infradead, lisa, parasitk, plord, sduck

User avatar
555x555
Common Wiggler
Posts: 231
Joined: Wed Jun 08, 2016 12:59 pm
Location: California

3Peak Opamps

Post by 555x555 » Mon Nov 25, 2019 2:19 pm

Scrolling through lcsc.com, stumbled on to these opamps which seem to have pretty excellent specs, especially for the money:

https://datasheet.lcsc.com/szlcsc/3PEAK ... 126446.pdf

Noise isn't great, but not much worse than the TL0s. Everything else seems to be much better for most applications I can think of.

Wondering if anyone knows anything about or has any experience with these, or this company generally? (or even any op amps besides the TI/AD duolith...)

Or if not, just a second pair of eyes to make sure I'm not missing something would be helpful :tu:

jorg
Wiggling with Experience
Posts: 379
Joined: Fri Apr 03, 2015 9:38 am
Location: East Coast USA

Post by jorg » Wed Nov 27, 2019 9:39 am

They do look interesting. For Euro (large signals and large power supply) I expect noise won't be a major issue. I have always had a soft spot for CMOS analog; these appeal to me for that reason.

MikeDB
Common Wiggler
Posts: 60
Joined: Fri Feb 08, 2019 8:28 am
Location: UK

Post by MikeDB » Wed Nov 27, 2019 3:34 pm

OPA2134 is cheaper - stick to known things unless it doesn't meet your unstated requirements.

User avatar
BugBrand
Knowledge of Bugs
Posts: 7086
Joined: Tue Jul 15, 2008 11:59 am

Post by BugBrand » Wed Nov 27, 2019 4:14 pm

MikeDB wrote:OPA2134 is cheaper - stick to known things unless it doesn't meet your unstated requirements.
?? It doesn't seem to be ??
OPA2134 price is high.

User avatar
Orgia Mode
Common Wiggler
Posts: 106
Joined: Mon Nov 04, 2019 5:24 pm

Post by Orgia Mode » Wed Nov 27, 2019 4:17 pm

TL072's are in JLCPCB's basic-parts list, so no extra charge to use them in the their assembly service.
One guy I design PCB's for gets a lot of his part reels from them and I've had uhh, two designs assembled by them:
First:

Second:


Very happy with both.

MikeDB
Common Wiggler
Posts: 60
Joined: Fri Feb 08, 2019 8:28 am
Location: UK

Post by MikeDB » Wed Nov 27, 2019 4:19 pm

LCSC is in China so I'm assuming OP is happy with dealing with Chinese suppliers
Last edited by MikeDB on Fri Dec 06, 2019 9:19 am, edited 1 time in total.

MikeDB
Common Wiggler
Posts: 60
Joined: Fri Feb 08, 2019 8:28 am
Location: UK

Post by MikeDB » Wed Nov 27, 2019 4:22 pm

Orgia Mode wrote:TL072's are in JLCPCB's basic-parts list, so no extra charge to use them in the their assembly service.
One guy I design PCB's for gets a lot of his part reels from them and I've had uhh, two designs assembled by them:
First:

Second:


Very happy with both.
Nice. Though I hadn't realised they only assembled on green. How much was assembly per board as I always assemble in the UK on their red version ?

User avatar
Orgia Mode
Common Wiggler
Posts: 106
Joined: Mon Nov 04, 2019 5:24 pm

Post by Orgia Mode » Wed Nov 27, 2019 4:31 pm

MikeDB wrote:
Orgia Mode wrote:TL072's are in JLCPCB's basic-parts list, so no extra charge to use them in the their assembly service.
One guy I design PCB's for gets a lot of his part reels from them and I've had uhh, two designs assembled by them:
First:

Second:


Very happy with both.
Nice. Though I hadn't realised they only assembled on green. How much was assembly per board as I always assemble in the UK on their red version ?
Super cheap. They want 7USD for the base service fee and then cost for parts.
I had to redesign a lot of my boards to accommodate their library of basic parts:
https://jlcpcb.com/client/index.html#/parts

But it was really worth it since their extended parts are 3USD extra.
I'll PM you with further info.

User avatar
Mungo
Super Deluxe Wiggler
Posts: 1077
Joined: Fri Jan 04, 2013 8:12 pm

Post by Mungo » Wed Nov 27, 2019 4:46 pm

BugBrand wrote:
MikeDB wrote:OPA2134 is cheaper - stick to known things unless it doesn't meet your unstated requirements.
?? It doesn't seem to be ??
OPA2134 price is high.
The "magic value" OPA1678 is a modern 30c part competing with TL072/TL052 and the very similarly positioned TP2604 part mentioned.

MikeDB
Common Wiggler
Posts: 60
Joined: Fri Feb 08, 2019 8:28 am
Location: UK

Post by MikeDB » Wed Nov 27, 2019 4:53 pm

Mungo wrote:
BugBrand wrote:
MikeDB wrote:OPA2134 is cheaper - stick to known things unless it doesn't meet your unstated requirements.
?? It doesn't seem to be ??
OPA2134 price is high.
The "magic value" OPA1678 is a modern 30c part competing with TL072/TL052 and the very similarly positioned TP2604 part mentioned.
1678 needs very low impedances to minimize noise. Better for those applications but I find the 2134 is best for general use.

Both are about same price and under 20cents so price isn't really an issue.

User avatar
Mungo
Super Deluxe Wiggler
Posts: 1077
Joined: Fri Jan 04, 2013 8:12 pm

Post by Mungo » Wed Nov 27, 2019 5:57 pm

MikeDB wrote:
Mungo wrote:
BugBrand wrote:
MikeDB wrote:OPA2134 is cheaper - stick to known things unless it doesn't meet your unstated requirements.
?? It doesn't seem to be ??
OPA2134 price is high.
The "magic value" OPA1678 is a modern 30c part competing with TL072/TL052 and the very similarly positioned TP2604 part mentioned.
1678 needs very low impedances to minimize noise. Better for those applications but I find the 2134 is best for general use.

Both are about same price and under 20cents so price isn't really an issue.
You keep talking about OPA2134? Ti lists their volume pricing just under $2 per unit. And the current noise is close enough between the OPA1678 and OPA2134 their noise vs input impedance plots look identical. Superior to TL072/TL052 and TP2604 doesn't even mention it.

Are you really sure you mean OPA2134, because everything you say about it doesn't seem to match up with the information we can see.

MikeDB
Common Wiggler
Posts: 60
Joined: Fri Feb 08, 2019 8:28 am
Location: UK

Post by MikeDB » Wed Nov 27, 2019 6:14 pm

Mungo wrote:
MikeDB wrote:
Mungo wrote:
BugBrand wrote:
MikeDB wrote:OPA2134 is cheaper - stick to known things unless it doesn't meet your unstated requirements.
?? It doesn't seem to be ??
OPA2134 price is high.
The "magic value" OPA1678 is a modern 30c part competing with TL072/TL052 and the very similarly positioned TP2604 part mentioned.
1678 needs very low impedances to minimize noise. Better for those applications but I find the 2134 is best for general use.

Both are about same price and under 20cents so price isn't really an issue.
You keep talking about OPA2134? Ti lists their volume pricing just under $2 per unit. And the current noise is close enough between the OPA1678 and OPA2134 their noise vs input impedance plots look identical. Superior to TL072/TL052 and TP2604 doesn't even mention it.

Are you really sure you mean OPA2134, because everything you say about it doesn't seem to match up with the information we can see.
TI aren't the only supplier of OPA2134s. As I mentioned earlier, the OP appeared to be OK with Chinese suppliers as they referenced a Chinese op-amp from a Chinese distributer, so I gave the link to the OPA2134s I use which I'll repeat below. Are they an exact copy of the TI part - possibly not Neither of course are the numerous suppliers of 5532s or 072s.

But have I determined any difference - nope. Behringer have been working with Chinese parts manufacturers for over 30 years to make sure they are suitable for audio applications but at lower cost than Western suppliers so we may as well take advantage of that.

But if you want to use TI and source from a Western distributer, then I agree the 1678 is probably the most cost-effective choice.
Last edited by MikeDB on Fri Dec 06, 2019 9:18 am, edited 1 time in total.

User avatar
devinw1
Super Deluxe Wiggler
Posts: 1079
Joined: Tue Aug 07, 2018 11:20 am
Location: Portland, OR
Contact:

Post by devinw1 » Wed Nov 27, 2019 8:23 pm

Why would you get 072s from a questionable supplier ? They are dirt cheap.legit. why risk it?

2134s are spendy.. it's one of the fancy SoundPlus chips. Not for every day use IMO.

User avatar
555x555
Common Wiggler
Posts: 231
Joined: Wed Jun 08, 2016 12:59 pm
Location: California

Post by 555x555 » Wed Nov 27, 2019 10:30 pm

Thanks for all the opinions.

And yeah, TL0s are great and cheap for the JLC PCBA service, so I’m still mostly using them. But sometimes a design can benefit from better specs.

Re: China, I think it’s maybe time to stop treating it as a monolith. lcsc has a good reputation, and seems to me they’re angling to break into a lot of the Mouser market. They’d have a lot to lose if it turned out their supply chain was shitty. Never heard anything about 3peak though, which is why I asked.

China is the clean, high-tech, innovative future, and if that doesn’t ring true, just wait a few years. Probably time we started talking about companies instead of nations.

User avatar
thetwlo
Super Deluxe Wiggler
Posts: 4676
Joined: Wed Apr 16, 2008 4:05 pm
Location: East Bay, CA US

Re: 3Peak Opamps

Post by thetwlo » Wed Nov 27, 2019 10:34 pm

555x555 wrote:Scrolling through lcsc.com, stumbled on to these opamps which seem to have pretty excellent specs, especially for the money:

https://datasheet.lcsc.com/szlcsc/3PEAK ... 126446.pdf
are these an OPA2604 clone? Weird that they'd use the same #.

User avatar
thetwlo
Super Deluxe Wiggler
Posts: 4676
Joined: Wed Apr 16, 2008 4:05 pm
Location: East Bay, CA US

Post by thetwlo » Wed Nov 27, 2019 10:36 pm

MikeDB wrote: TI aren't the only supplier of OPA2134s.
but the ones on Aliexpress and ebay(from china) are fake.

User avatar
thetwlo
Super Deluxe Wiggler
Posts: 4676
Joined: Wed Apr 16, 2008 4:05 pm
Location: East Bay, CA US

Post by thetwlo » Wed Nov 27, 2019 10:41 pm

MikeDB wrote: Behringer have been working with Chinese parts manufacturers for over 30 years to make sure they are suitable for audio applications but at lower cost than Western suppliers so we may as well take advantage of that.
no, they bought the factory. Behringer is CoolAudio, they manufacture many of their own chips, avoiding all the shady chip dealers in China.

User avatar
thetwlo
Super Deluxe Wiggler
Posts: 4676
Joined: Wed Apr 16, 2008 4:05 pm
Location: East Bay, CA US

Post by thetwlo » Wed Nov 27, 2019 11:10 pm

555x555 wrote: lcsc has a good reputation, and seems to me they’re angling to break into a lot of the Mouser market.
they do sound legit as they charge $2+ for an OPA 2134.

ixtern
Common Wiggler
Posts: 168
Joined: Mon Jan 22, 2018 4:34 am

Post by ixtern » Thu Nov 28, 2019 2:00 am

Once I've bought 20 cheap OPA2134 from the local supplier. All they looked like the TI (BB) one's. But then I've measured them input bias current (indirectly, measuring output offset voltage with 100MB resistors).
ALL of them were fakes - all had input bias current about 1-2 nA, what suggests they were rebranded cheap bipolar op amps.
If someone simply puts them into circuit they will be working so it's hard to notice they are not JFETs op amps if you not put your attention at the tiny differences.
From this time I am looking for OPA2134 only at the reputable suppliers and these chips are never cheap.

MikeDB
Common Wiggler
Posts: 60
Joined: Fri Feb 08, 2019 8:28 am
Location: UK

Post by MikeDB » Thu Nov 28, 2019 5:52 am

Lots of comments here so a general reply. Behringer were in China long before they bought CoolAudio. They were already the biggest competitor to the established companies such as the one worked for in the late 90s and have grown massively since then. Whilst we always used transistors from Rohm, opamps from nJRC, connectors from Neutrik and faders/pots from Alps even in our cheapest mixers, Behringer often used Chinese copies, but because they had their own factory in China they could work on quality issues quickly and the Chinese parts soon became as good as the originals. Were their products as reliable as ours - possibly not, but far better than the rubbish places like Maplin sold. Did they sound the same - unfortunately yes. And did they sell far more units than us - most definitely yes !
So in the end they bought Midas and other companies, and have a huge range of products including numerous synth clones, whilst my old company is now owned by Samsung who don't really know what to do with it.

The key reason I use 2134s is they work well on +/- 2.5V whereas most other devices don't. The 2604 for instance needs +/- 4.5V to bias itself properly. If the devices were counterfeits they simply wouldn't function in my designs.

However I totally agree there are counterfeit devices on the market so you do need to be VERY careful. But in fact the counterfeit ones will probably have a TI or BB logo - why go to the effort of making a counterfeit that didn't look like the 'real thing' ? There are many of those on AliExpress but the ones I buy aren't and do perform moreorless the same as the TI ones. They usually do this by using the services of a company based in Singapore who take an IC and slice it layer by layer and produce a new set of masks. So whilst not a perfect copy, they are usually pretty close. I expect the 3PEAK device mentioned was produced in this manner from the original 2604. But it's hardly something TI can complain about - all the American companies did it to each other in the 1960s to 1980s ! And as device numbers are not copyrightable, it makes sense to use the same number on your device.

User avatar
BugBrand
Knowledge of Bugs
Posts: 7086
Joined: Tue Jul 15, 2008 11:59 am

Post by BugBrand » Thu Nov 28, 2019 6:17 am

Very interesting discussions! Thanks for commenting on your views of the 2134s you've sourced.

3Peaks seems a bit strange - I mean, the TP2604 the OP linked to doesn't appear to be on their site..?!.. I started looking at the TP2302 - until I finally saw it only runs on +5V, as do most of the other opamps listed on their site. Kind of puts me off delving deeper at this stage.

PS - agreed that the OPA1678 is a good recent addition at affordable prices

MikeDB
Common Wiggler
Posts: 60
Joined: Fri Feb 08, 2019 8:28 am
Location: UK

Post by MikeDB » Thu Nov 28, 2019 6:52 am

BugBrand wrote:Very interesting discussions! Thanks for commenting on your views of the 2134s you've sourced.

3Peaks seems a bit strange - I mean, the TP2604 the OP linked to doesn't appear to be on their site..?!.. I started looking at the TP2302 - until I finally saw it only runs on +5V, as do most of the other opamps listed on their site. Kind of puts me off delving deeper at this stage.
I hadn't noticed that - 5V operation gets me very interested ! I can only find the TP24xx which are similar but think I'll order some and give them a try.

Regarding the website, I often find the English translated sites are years out of date. I use several high voltage MOSFET drivers for Class D amps from another Chinese company called EGMicro which I have the owner of a local Chinese restaurant translate the key parts of the datasheets for me :-)

loki
Common Wiggler
Posts: 89
Joined: Fri Apr 25, 2014 4:20 pm

Post by loki » Thu Nov 28, 2019 8:33 pm

MikeDB wrote: They usually do this by using the services of a company based in Singapore who take an IC and slice it layer by layer and produce a new set of masks. So whilst not a perfect copy, they are usually pretty close. I expect the 3PEAK device mentioned was produced in this manner from the original 2604. But it's hardly something TI can complain about - all the American companies did it to each other in the 1960s to 1980s ! And as device numbers are not copyrightable, it makes sense to use the same number on your device.
Compare the data sheets. It is obvious that the TP2604 is not a clone of the OPA2604. The TP2604 is CMOS while the OPA2604 is a FET input op-amp and the specs are quite different.

User avatar
555x555
Common Wiggler
Posts: 231
Joined: Wed Jun 08, 2016 12:59 pm
Location: California

Post by 555x555 » Fri Nov 29, 2019 12:38 am

Yeah, the new design is part of what’s interesting about them. CMOS, 36V, low distortion, decent precision. Idk much about semi fabs, but offhand this seems like a pretty innovative chip.

MikeDB
Common Wiggler
Posts: 60
Joined: Fri Feb 08, 2019 8:28 am
Location: UK

Post by MikeDB » Fri Nov 29, 2019 7:59 am

loki wrote:
MikeDB wrote: They usually do this by using the services of a company based in Singapore who take an IC and slice it layer by layer and produce a new set of masks. So whilst not a perfect copy, they are usually pretty close. I expect the 3PEAK device mentioned was produced in this manner from the original 2604. But it's hardly something TI can complain about - all the American companies did it to each other in the 1960s to 1980s ! And as device numbers are not copyrightable, it makes sense to use the same number on your device.
Compare the data sheets. It is obvious that the TP2604 is not a clone of the OPA2604. The TP2604 is CMOS while the OPA2604 is a FET input op-amp and the specs are quite different.
Okay - I hadn't looked at the OPA specs. Very silly to use the same number then - some large company buyer is going to mess up.

Post Reply

Return to “Music Tech DIY”