Author Topic: Why is the 741 op amp still produced?  (Read 19650 times)

0 Members and 1 Guest are viewing this topic.

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15149
  • Country: de
Re: Why is the 741 op amp still produced?
« Reply #100 on: October 01, 2020, 08:50:25 pm »
As far as I have heard most of today's TL082 OPs are also low enough in noise to meet the TL072 specs, which is the lower noise variant, otherwise identical.  Similar it may happen with other parts - they may not get many of the lowest voltage class transistors. So chances are quite a few may have considerably higher than specified break down voltage. Similar I would not be surprised to get a bag of 1N4001 with all higher than 200 V break-down voltage.
Chances are they decide to mark them also according to orders, not just measurements.
The same may happen to LM358/LM2904 (lower voltage grade).

The PNPs in the old NPN base process are lateral parts with the dimensions defined by lithography. There is no problem making larger structures with a high resolution process. So they should be OK to still build the old style transistors.  A few days ago a found an article showing a lateral BJT in an standard 1.2 µm CMOS process and an OP with 2 such BJTs for the input and the rest MOSFETs. So kind of mixed MOS and BJT the other way around than in the early days. The lateral transistor were supposed to be even good quality (low noise, good gain, though slow) - not the marginal PNPs in the 741.
 
The following users thanked this post: SilverSolder

Offline tszaboo

  • Super Contributor
  • ***
  • Posts: 8217
  • Country: nl
  • Current job: ATEX product design
Re: Why is the 741 op amp still produced?
« Reply #101 on: October 01, 2020, 10:36:58 pm »
Don't forget that evolution usually is also tied to potential revenue. The fact nobody ever refined a process to reach LTZ or LM levels may be due simply to a lack of interest or too narrow of a market to be tapped into. The number of different voltage references available when compared to 30 years ago is impressive, but they are all competing in a segment with more relaxed demands.
Not everything is about profit, or it shouldn't be. I have the feeling that humanity cannot get anything done anymore, because of the search of immediate profit. I've seen companies do nothing, and let engineers sit idle, rather than have them working on a project. They calculated, that the NRE cost of the project is to big, and that would eat up the profit. I guess these things only make sense if you have an MBA.
 
The following users thanked this post: SilverSolder

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Why is the 741 op amp still produced?
« Reply #102 on: October 02, 2020, 02:50:58 am »
Don't forget that evolution usually is also tied to potential revenue. The fact nobody ever refined a process to reach LTZ or LM levels may be due simply to a lack of interest or too narrow of a market to be tapped into. The number of different voltage references available when compared to 30 years ago is impressive, but they are all competing in a segment with more relaxed demands.
Not everything is about profit, or it shouldn't be. I have the feeling that humanity cannot get anything done anymore, because of the search of immediate profit. I've seen companies do nothing, and let engineers sit idle, rather than have them working on a project. They calculated, that the NRE cost of the project is to big, and that would eat up the profit. I guess these things only make sense if you have an MBA.

I've noticed the same thing:  People prefer doing "nothing" to doing "something"...  as you say, probably because "something" is perceived to have a cost associated with it, whereas doing "nothing" has somehow escaped that analysis...
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 11333
  • Country: us
  • $
Re: Why is the 741 op amp still produced?
« Reply #103 on: October 02, 2020, 02:58:10 am »
if it works why change it, its like asking why some things are still made of wood

does a kitchen knife need a neotitaniumsteel handle? no
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7678
  • Country: ca
Re: Why is the 741 op amp still produced?
« Reply #104 on: October 02, 2020, 04:02:04 am »
Look at the other side - the op-amp portfolio manager needs new products, new profits so they constantly roll out new op-amps, per the MBA way.

We think cost is driven by silicon area, but I think it's reverse - they have a pricing matrix using features such as offset voltage, speed, RR etc. to determine the pennies to charge and fill some niche left out by competitors.

It's a problem now with the mergers and acquisitions where all of a sudden a corporation now owns previously competing portfolios. What a mess. They have to swing the axe and thin out the lineup. Consolidation (PMI, SSM, Burr Brown, NS, LT) will lead to less competition and poorer offerings because innovation is not really needed.  It's ADI vs TI now.

P.S.- on the speed issue, I found LT1494  0.001V/usec slew rate on super low power 1.5uA, a neat part.
 

Offline tszaboo

  • Super Contributor
  • ***
  • Posts: 8217
  • Country: nl
  • Current job: ATEX product design
Re: Why is the 741 op amp still produced?
« Reply #105 on: October 02, 2020, 08:44:03 am »
Look at the other side - the op-amp portfolio manager needs new products, new profits so they constantly roll out new op-amps, per the MBA way.

We think cost is driven by silicon area, but I think it's reverse - they have a pricing matrix using features such as offset voltage, speed, RR etc. to determine the pennies to charge and fill some niche left out by competitors.

It's a problem now with the mergers and acquisitions where all of a sudden a corporation now owns previously competing portfolios. What a mess. They have to swing the axe and thin out the lineup. Consolidation (PMI, SSM, Burr Brown, NS, LT) will lead to less competition and poorer offerings because innovation is not really needed.  It's ADI vs TI now.

P.S.- on the speed issue, I found LT1494  0.001V/usec slew rate on super low power 1.5uA, a neat part.
No, for sure not cost driven. I mean, LT is charging the same amount for a single opamp, than chinese companies for a reel of MC34063 clone.

About the mergers, it is absolutely a mess. They started obsoleting parts left and right. Now we have to look out for parts, like BJTs and zeners, because they might stop producing them. And for sure, you dont care about this in a washing machine. For me these are used in a certified product, and if I have to change it, it is several thousand EUR just to certify the product again.
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 10031
  • Country: gb
Re: Why is the 741 op amp still produced?
« Reply #106 on: October 02, 2020, 09:41:04 am »
Look at the other side - the op-amp portfolio manager needs new products, new profits so they constantly roll out new op-amps, per the MBA way.
No sane portfolio manager wants new products. They want new sales of existing products, since that doesn't involve a lot of NRE and a long time to market. New products are a last resort when pushing the existing products fails. Most new semiconductor products come about from 2 driving forces. The sales team keep feeding back requirements they see from customers, that existing products can't fulfill. The technology side keeps showing some interesting results they can achieve, either today or in the near future. The portfolio manager's role is to find where those 2 inputs match, and an appealing new product can be defined. People rarely develop new semiconductor devices without an extensive survey of what customers are actually looking for, and when they need those parts. One of the key things most companies require for a development project kick-off is some key named customers, their expected volumes, and the market window for each customer.
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 10031
  • Country: gb
Re: Why is the 741 op amp still produced?
« Reply #107 on: October 02, 2020, 09:47:27 am »
About the mergers, it is absolutely a mess. They started obsoleting parts left and right. Now we have to look out for parts, like BJTs and zeners, because they might stop producing them. And for sure, you dont care about this in a washing machine. For me these are used in a certified product, and if I have to change it, it is several thousand EUR just to certify the product again.
Most of the obsoleting is not about mergers, although mergers are often the trigger point for action. Obsoleting is mostly about lack of demand or obsolete production equipment. While some ancient devices, and compatible derivatives, like the 555, still sell by the billion, many old devices still available sell just a few thousand parts per year. How low can you expect demand to go, and still have the part listed as available?
 

Offline tszaboo

  • Super Contributor
  • ***
  • Posts: 8217
  • Country: nl
  • Current job: ATEX product design
Re: Why is the 741 op amp still produced?
« Reply #108 on: October 02, 2020, 06:22:51 pm »
About the mergers, it is absolutely a mess. They started obsoleting parts left and right. Now we have to look out for parts, like BJTs and zeners, because they might stop producing them. And for sure, you dont care about this in a washing machine. For me these are used in a certified product, and if I have to change it, it is several thousand EUR just to certify the product again.
Most of the obsoleting is not about mergers, although mergers are often the trigger point for action. Obsoleting is mostly about lack of demand or obsolete production equipment. While some ancient devices, and compatible derivatives, like the 555, still sell by the billion, many old devices still available sell just a few thousand parts per year. How low can you expect demand to go, and still have the part listed as available?
Well, kinda. Remember Fairchild? They had no obsolescence policy, until ON bought them. ON is obsoleting like half their portfolio as we speak.
Microchip throw out entire product lines from Atmel.
Littlefuse just throw out IXYS's solar panel portfolio
IRF was decimated by Infineon. When they bought them, it was synergy (sound of angels in  the background), after a year it was too much overlap.
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 10031
  • Country: gb
Re: Why is the 741 op amp still produced?
« Reply #109 on: October 02, 2020, 08:19:46 pm »
Well, kinda. Remember Fairchild? They had no obsolescence policy, until ON bought them. ON is obsoleting like half their portfolio as we speak.
A blanket no obsolescence policy is ridiculous, and obviously won't be kept to. When they sell just a few hundred a year, do you really think its viable to not obsolete the part? When the fab is unmaintainable, do you really think its viable to not obsolete the part?

The first time I got to see from the inside the point at which old parts get dropped I was astonished how long that vendor had continued with supply. Many of the parts dropped on that occasion sold just a few thousand per annum, and hadn't sold in decent quantities for years. Vendors serving things like the consumer and computer markets tend to have fairly short windows of opportunity for their parts, and are dropping them all the time. Vendors serving industrial customers are at the opposite end of the scale. Large vendors, serving many markets, may have different practices in each division.

Microchip throw out entire product lines from Atmel.
Littlefuse just throw out IXYS's solar panel portfolio
IRF was decimated by Infineon. When they bought them, it was synergy (sound of angels in  the background), after a year it was too much overlap.
I wonder how many of those parts were sold per annum? I doubt too many high volume, decent margin parts were dumped at the drop of a hat. Political infighting obviously plays a part, but its mostly down to volumes, margins, and the technology still being available to make the parts.
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Why is the 741 op amp still produced?
« Reply #110 on: October 02, 2020, 08:55:45 pm »
[...] How low can you expect demand to go, and still have the part listed as available?

At some point, it must make sense to do a "final production run" and put a couple of crates into a warehouse, to sell out of over the next e.g. 5-10 years before finally selling the last remnants to stockists of esoteric components.  That has to make a lot more sense than keeping an obsolete production line operating for a very small number of parts annually...
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 10031
  • Country: gb
Re: Why is the 741 op amp still produced?
« Reply #111 on: October 02, 2020, 09:51:47 pm »
[...] How low can you expect demand to go, and still have the part listed as available?

At some point, it must make sense to do a "final production run" and put a couple of crates into a warehouse, to sell out of over the next e.g. 5-10 years before finally selling the last remnants to stockists of esoteric components.  That has to make a lot more sense than keeping an obsolete production line operating for a very small number of parts annually...
I bet you could start a business providing that end of life stuff as a service. Keep banks of dies in controlled conditions, and do packaging and test on demand. Maybe you could call it something like "Rochester Electronics".  :)
 
The following users thanked this post: SilverSolder

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 6105
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: Why is the 741 op amp still produced?
« Reply #112 on: October 02, 2020, 10:22:32 pm »
Don't forget that evolution usually is also tied to potential revenue. The fact nobody ever refined a process to reach LTZ or LM levels may be due simply to a lack of interest or too narrow of a market to be tapped into. The number of different voltage references available when compared to 30 years ago is impressive, but they are all competing in a segment with more relaxed demands.
Not everything is about profit, or it shouldn't be. I have the feeling that humanity cannot get anything done anymore, because of the search of immediate profit. I've seen companies do nothing, and let engineers sit idle, rather than have them working on a project. They calculated, that the NRE cost of the project is to big, and that would eat up the profit. I guess these things only make sense if you have an MBA.

It depends on how you define profit. In the cold war era, profit was the gain over the ever present danger of the other side and money was not really a problem for these state of the art projects. Fast forward to today and the scenario changed a lot.

My own example is a cold war era Power Designs 5015A power supply that once belonged to General Dynamics. You would be very hard pressed to find a modern analog linear power supply with that level of build quality and finesse. Oh well... Modern times.
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Offline schmitt trigger

  • Super Contributor
  • ***
  • Posts: 2431
  • Country: mx
Re: Why is the 741 op amp still produced?
« Reply #113 on: October 03, 2020, 03:18:54 am »
Coppice;
Today I was wondering precisely about Rochester Electronics.
Wondering from where do the wafers come from.

I wondered if they had purchased obsolete fabs, shipped them to the Far East, and continued production of old parts.
But your assessment of Rochester packaging and testing NOS wafers makes much more sense.
Do you know this as a fact?

EDIT: searching the web, apparently Rochester is doing it both ways.
« Last Edit: October 03, 2020, 03:25:15 am by schmitt trigger »
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 10031
  • Country: gb
Re: Why is the 741 op amp still produced?
« Reply #114 on: October 03, 2020, 11:23:30 am »
Coppice;
Today I was wondering precisely about Rochester Electronics.
Wondering from where do the wafers come from.

I wondered if they had purchased obsolete fabs, shipped them to the Far East, and continued production of old parts.
But your assessment of Rochester packaging and testing NOS wafers makes much more sense.
Do you know this as a fact?

EDIT: searching the web, apparently Rochester is doing it both ways.
Rochester's web site tells you the things they do. They work with the original vendors of a part in most cases. They take either wafers or completed parts from them, plus all the packaging and test documentation needed to complete the parts to their original spec. Assembly and test is a commercial enterprise these days. Even the biggest silicon vendors outsource much of their assembly and test to specialist A&T houses in Asia. Packages don't go obsolete very often, and testers don't change very fast. So, as long as the wafers area stored in optimal conditions, Rochester can get them assembled and tested as the would have been originally.

I am not aware of Rochester operating any fabs. I can't imagine how that would work. If you have a hit product and need more capacity there are usually only one or two other fabs that can run the same process, which you might be able to expand into. The older things get the less compatibility you see between fabs. You probably can't even put the masks into the lithography machine, unless you have just the right machine to match your masks. There is very little compatibility in the fab business.
 
The following users thanked this post: schmitt trigger

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Why is the 741 op amp still produced?
« Reply #115 on: October 03, 2020, 01:26:01 pm »
Wondering why it is cheaper to store them as wafers and build the ICs later, rather than just asking the original supplier to make a batch of product on the lines that have already been set up for it, before powering down for the last time...   and storing the output of that run?   Is it because you then have flexibility to put that chip into different packages etc.?
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Why is the 741 op amp still produced?
« Reply #116 on: October 03, 2020, 02:20:28 pm »
I have a batch of transistors whose actual breakdown voltage is astronomical compared to the specification.   Does/did it ever happen in production, that the process improved to the point where even the lowest specced parts easily achieved the performance of the (older) highest rated parts - so the lower rated parts effectively became price/performance "bargains" (but not advertised, so nobody would ever know)?

I have seen that happen with some transistors where the lowest specified parts became unavailable, and were eventually discontinued in favor of the higher voltage parts.

As far as I have heard most of today's TL082 OPs are also low enough in noise to meet the TL072 specs, which is the lower noise variant, otherwise identical.

Noise tests tend to be time consuming which adds cost to production, so it is not surprising that TL072s often meet the noise specification for the TL082.  On the other hand, I have hardly ever seen any price difference when buying these parts so I buy the presumably noise tested TL082 series in favor of the TL072.

Quote
A few days ago a found an article showing a lateral BJT in an standard 1.2 µm CMOS process and an OP with 2 such BJTs for the input and the rest MOSFETs. So kind of mixed MOS and BJT the other way around than in the early days. The lateral transistor were supposed to be even good quality (low noise, good gain, though slow) - not the marginal PNPs in the 741.

A couple weeks ago I searched for examples of operational amplifiers which support a high differential input voltage range that are built on a modern process and found nothing.  I was specifically looking for modern replacements for the 741 and 301A types of operational amplifiers.  The closest ones I found are single supply 324/358 type improved parts like the LT1006/LT1013/LT1014 which obviously make use of high voltage PNPs but are hardly modern despite their high performance.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15149
  • Country: de
Re: Why is the 741 op amp still produced?
« Reply #117 on: October 03, 2020, 03:05:42 pm »
For OPs with a larger input differential the main option to look at are FET input types (e.g. OPA196/7). Even there some JFET types (e.g. OPA170-172) have a limited input differential. A moderate cost (can be comparable to 741), slightly improved lm358 is the MC33172.

The TL072 is the version with lower noise specs, the TL082 was the low cost version that allowed more noise. Chances are these are not individually tested - probably just a few per batch. I also noted essentially no price difference.

The lower grade transistors (lower gain bins or lower voltage bins) can become rarer though still in the program but less likely at the distributors. There however one exception: the higher supply variants for some OPs are not taking over. For some reasons there are still LM2904 ( = lower voltage grade of LM358) available - could be just LM358 with less testing for the penny pitchers.
 

Offline tszaboo

  • Super Contributor
  • ***
  • Posts: 8217
  • Country: nl
  • Current job: ATEX product design
Re: Why is the 741 op amp still produced?
« Reply #118 on: October 03, 2020, 03:18:50 pm »
Well, kinda. Remember Fairchild? They had no obsolescence policy, until ON bought them. ON is obsoleting like half their portfolio as we speak.
A blanket no obsolescence policy is ridiculous, and obviously won't be kept to. When they sell just a few hundred a year, do you really think its viable to not obsolete the part? When the fab is unmaintainable, do you really think its viable to not obsolete the part?

The first time I got to see from the inside the point at which old parts get dropped I was astonished how long that vendor had continued with supply. Many of the parts dropped on that occasion sold just a few thousand per annum, and hadn't sold in decent quantities for years. Vendors serving things like the consumer and computer markets tend to have fairly short windows of opportunity for their parts, and are dropping them all the time. Vendors serving industrial customers are at the opposite end of the scale. Large vendors, serving many markets, may have different practices in each division.

Microchip throw out entire product lines from Atmel.
Littlefuse just throw out IXYS's solar panel portfolio
IRF was decimated by Infineon. When they bought them, it was synergy (sound of angels in  the background), after a year it was too much overlap.
I wonder how many of those parts were sold per annum? I doubt too many high volume, decent margin parts were dumped at the drop of a hat. Political infighting obviously plays a part, but its mostly down to volumes, margins, and the technology still being available to make the parts.
The no obsolescence policy was rolled out in 2013, and ON bought them in 2016, so we will never know if this would've worked. In all fairness, Fairchild had a fairly "established" portfolio, they weren't chasing trends, just making parts for industrial applications. Only a few new IC was for new applications, like SIM card switch and other low-tech parts.

IDK, how many parts Atmel or IRF was selling. I think the part which got canned was Atmels FPGA portfolio, that was very old designs. So you can imagine, anyone using those had a very good reason to use them.
Like that company, open sourcing their respirator. They used some 20 year old ST microcontoller.

The reason to keep these going is simple. They are used in the train your are taking to work. In the hospital equipment. In the nuclear powerplant controller board. Even if you only sell 100 of these parts, we need these parts.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf