ON Semiconductor 5534

[Gus]

I missed that thread.

Is the On 5534 made the same as the Phillps?

Thanks for the link.

 

[JohnRoberts]

Not to be snarky, but why throw good money at tooling the old 5534 other than to be able to sell parts under a popular number?

The 5534 was great back in the '70's but there are too many better parts available today to count. For low performance GP use there are better parts too, so it's main value IMO is as a well known part number, and perhaps to support legacy designs where no engineer is around to approve a change.

Note: I can image some oddball applications that relied upon idiosyncracies of that particular design but they are surely too rare to worry about, and it had it's rough edges.

JR

Edit- FWIW that schematic on the ON data sheet looks like the original 5534 or pretty close. I never did care for that compensation scheme./edit

 

[JohnRoberts]

I like the 4560 for higher current but relaxed audio spec, applications. I'm embarrased to say how cheap those were when you buy a million a year.

I think it's all about the well known part number... Engineering management is typically rewarded for reusing old designs especially when reinventing the wheel doesn't deliver some merchantable difference.

I've worked both sides of that fine line..

JR

 

[CJ]

Does TI wanna know?

 

[bcarso]

[quote author="JohnRoberts"]I like the 4560 for higher current but relaxed audio spec, applications. I'm embarrased to say how cheap those were when you buy a million a year.
JR[/quote]

The 4560 is my "house wine" opamp. I've seen as low as 9 cents...

 

[CJ]

7613

audio workhouse.

dual ver of the 12.

precirsor to 5512/13

 

[CJ]

7613

audio workhouse.

dual ver of the 12.

precirsor to 5512/13

2 1/2 cents

 

[Samuel Groner]

The 5534 was great back in the '70's but there are too many better parts available today to count.

Can you elaborate? I'm not aware of an opamp at the same price as the NE5532 which has as low distortion and noise. If you allow $5 or even $10 instead of $1 per IC there's plenty of choice for sure though.

Samuel

 

[JohnRoberts]

There should be no debate that there are many better parts today. I can't intelligently respond to "good for the money" since I've been out of the loop and not beating up reps for pricing on such parts for over a decade now.

Even back then I was using a selected* version of 5532 with our own house number on it, and then only as needed. For most low noise-gain line level crunching the bi-fets were my go to part. FWIW I even brought an ungraded 5532 into the system for use in lower noise-gain applications like balanced output drivers where I needed the drive capability (and slew rate) more than noise floor. Many lower performance designs actually used 4560 in the audio path, while I don't consider this audiophile, it was adequate and comparable in competitive budget products.

If I were an IC manufacturer today I would target the 5532 as a high volume opportunity. While the line is blurring between bifet and bipolar input opamps I still see a place for both. While I have defended the performance of the 5532 in general there are a few wrinkles in it, I'd like to see ironed out, and I don't consider myself overly tweaky. Since I was already using a selected version of the 5532 there was an easier price bar to meet at my old day job. If the price was right, that improved part would get the entire 5532 business and perhaps some of the 4560 business too, but I wouldn't expect to ever meet that part's excellent price.

Sorry for such a wordy non-answer.

JR

* Our house number 5532 was selected for the spectral envelope of the noise floor not the amount of noise, but this selection effectively weeded out parts with high 1/F noise and process problems. By this selection the noise floor, when audible, was uniform and not nasty sounding.

 

[clintrubber]

--

 

[clintrubber]

--

 

[JohnRoberts]

Was that the purple plague?

JR

 

[CJ]

ok all you opamp brainiacs, which opamp has the worst frequency response of all time?

 

[SSLtech]

[quote author="mediatechnology"]Peter - You may find this story interesting. Back when I worked for SSL we, like many other manufacturers, had to use Raytheon 5534s. These parts had this sputtering 1/f popcorn whatever you want to call it noise. They sounded awful.

...<snip happens>...

Raytheon channels, when nulled to each other, sounded very distorted. Pure grunge. Very IM-ish sounding.

Each channel from line in to mix went through perhaps 30 5534 gain stages.

Clients quickly understood what we were trying to accomplish replacing, usually at no charge, their Raytheon parts with Signetics.[/quote]

Word.

Wayne came to town (before I moved here) and commissioned a console for a mutual friend. Later on, this op-amp sonic performance cropped up, and I've a suspicion that our friend noticed the issues for himself...

As a result, Wayne prevailed upon SSL to send down a FedEx package of 5534s. A side-by-side comparison was performed, and the console was COMPETELY refitted. It was subsequently noted that some of the modules ALREADY had the better 5534s in them, while others contained the nasty Raytheon versions. Some modules were a mixture.

After the refit was completed, there were quite a few left over. They were kept as service replacements. The last of these are STILL in the original FedEx tube, which can -if necessary- date this documented occurrence, should anyone care to know more.

Wayne speaks the truth.

[quote author="CJ"]ok all you opamp brainiacs, which opamp has the worst frequency response of all time?[/quote]
A dead one.

Keith

 

[Samuel Groner]

Should anyone care to know more.

I do (the work on my grand opamp test series has begun). Any chance for a shipping to Switzerland? A pair of both is enough.

Samuel

 

[dale116dot7]

[quote author="CJ"]ok all you opamp brainiacs, which opamp has the worst frequency response of all time?[/quote]

I'm not 100% sure but I'd hazard a guess at a Philbrick P2.

 

[JohnRoberts]

[quote author="Samuel Groner"]
I do (the work on my grand opamp test series has begun). Any chance for a shipping to Switzerland? A pair of both is enough.

Samuel[/quote]

Coincidentally if my memory serves me right, the selected 5532 we used under a house number was (is?) purchased from Raytheon. The grading program was set up before my time there so I don't know the history but in hindsight perhaps this noise spectral balance test was just screening for 1/F process related contamination. It was explained to me as a noise floor thing but who knows?

If you want bad opamps to test I'm sure there's lots out there... Sorry I throw mine away. I trashed a bunch of noisy TL074s in the '70s.

Later at PV I disapproved one of the majors as a vendor on a jelly bean 3 terminal regulator because of high noise levels. When I discussed it with them they declined to provide a noise spec for the part. Perhaps they didn't want to tarnish their 6 sigma performance by looking too closely. There was a number of other companies with identical but quiet parts ready to sell us parts.

JR

 

[SSLtech]

I'm with you John.

Seems that there's always been more than just a part number to consider.

Mercifully, we can actually measure these things, unlike the audiophile community, who have to worry about the soundstage, moon phase, relative humidity, temperature and football league results and their influence upon cable orientation.

Ah how easy we simple fools have it... what with the luxury of being able to actually MEASURE things...

Keith

 

[clintrubber]

--

 

[JohnRoberts]

The 5534/2 is respectable but not without imperfection. It's much too late in the game, and I've designed far too many 5534/2 into sockets to start bad mouthing that part now but there's no reason we can't take advantage of newer processes and improved internal circuitry.

I could still hammer a nail with a 1970 hammer, but why not use a new improved hammer if it doesn't break the bank? If I'm on a budget I'll hit the nail with a rock.

JR