Does the NJM5532 really work at 9VDC?

[fryingpan]

I was perusing the datasheet and apparently they claim that their version of the 5532 works down to +/-3V DC. (And up to +/-22V DC, although the recommended is still +/-15V DC). Do any of you have any experience in that regard?

[R.G.]

The part >IS< the datasheet, not whatever legged-thingie you get in the mail. Legit manufacturers (and JRC is one) go to great lengths to make sure that they meet whatever is on the datasheet.

That being said, what >exactly< does the datasheet say about all of:
- the minimum power supply limits
- input common mode and differential mode range, possibly at the minimum supply voltage
- output swing compared to the power supply
These are where the gotchas will likely hide on the datasheet.

Then there is always the issue of whether the part you have is legit or fake.

I trust the totality of the datasheet, even and especially when read with a very critical eye, far more than some internet anecdotes. There are legal reasons for the manufacturer to try their best to make the datasheet come true.

[ElectricDruid]

+1 agree with RG. If the datasheet says it's ok, it's ok. Because if it wasn't, they'd get sued by someone and it would cost serious money.

Here's the datasheet:

https://www.mouser.com/datasheet/2/294/NJM5532_E-1917664.pdf

And it clearly says in the "RECOMMENDED OPERATING VOLTAGE" section "±3~±22V". So yeah, it's fine. At 9V it'd be more than comfortable, since if ±3V is within the range recommended, the *actual minimum* must be at least a bit lower still, or they wouldn't be able to guarantee it'd work at ±3V. From that datasheet, I wouldn't be surprised if you can run them on 5V, although you'd be extremely foolish to release a commercial product that actually did that.

[R.G.]

Thanks for the datasheet link, Tom.
Interpreting the datasheet charts/curves on a datasheet is always dangerous, as the typical curves are NOT guaranteed. That being said, the chart of maximum output voltage versus power supply voltage on page 4 indicates that at +/- 3V supplies, you might get +/- 2V of output. Maybe, typically, RMS, peak, max.

My take is that yes, it works at +/-3V supplies, but it may not have a big output voltage swing.

[ElectricDruid]

My take is that yes, it works at +/-3V supplies, but it may not have a big output voltage swing.

Sure. It's an op-amp. Unless it's **special** we're not expecting it to go right to the rails, right?! So you typically take a decent chunk off the supply voltage to allow for clipping. Adjust to taste for ones that go close to ground, close to both rails, whatever. I just take a volt off both rails and reckon that's fairly safe as an expectation . Without checking carefully or doing experiments, that's what I'd start with here too - and that fits broadly with your +/-2V output.

The answer to the question in the topic title is pretty clear though, right? This chip would be absolutely fine at 9V. That's well within its capabilities.

[fryingpan]

Actually the fact that (I misread) it states stable, reliable operation up to the operating limits is more interesting. That means that it can be used for (relatively low power) power amps as well. 20V peak is 25W RMS into 8 ohms. At that point, you may as well just hook up the opamp's output to the output transistors. (Well, their drivers).

[PRR]

> 20V peak is 25W RMS into 8 ohms.

And 2.5 AMPs. Doubt it. (Spec is 38milliAmp, voltages not well defined.)

Also OTOO 10 Watts heat, a lot for a DIP8.

Also see graph for output voltage in various loads. 300r does very well for a small chip (why it is a studio workhorse) but a long way from 300 to 8 unless you have buffers.

[PRR]

Remember that datasheet writing is an art. Or Creative Writing. You have to say enough to get bought, yet not promise anything you can't do. It's an intense months-long meeting. The things it does NOT say are sometimes the most important to you.

For "does it really work on a couple penlight cells? A USB feed? A depleted pocket radio battery?", I like the supply current VS supply voltage. It's not too obvious to senior managers (to veto), and not a good clue for all chips. But classic chip op-amps are constant current bias, until they starve and quit.


The typical NJM5532 is +/-15% on current down to just about 6V (+/-3V) and then it falls off quite fast. It "may" be functional (the output flaps about right) at 3V total but the dotted line is the rocky shore at the end of the sea and you best not push it.

[fryingpan]

> 20V peak is 25W RMS into 8 ohms.

And 2.5 AMPs. Doubt it. (Spec is 38milliAmp, voltages not well defined.)

Also OTOO 10 Watts heat, a lot for a DIP8.

Also see graph for output voltage in various loads. 300r does very well for a small chip (why it is a studio workhorse) but a long way from 300 to 8 unless you have buffers.

Of course the opamp would just be the small signal part. A 5532 should be able to power headphones all right, on the other hand.

[merlinb]

That means that it can be used for (relatively low power) power amps as well.

You can make a 16W power amp with nothing but 5532s...
https://forum.diyaudiotr.com/dosyalar/dosya/Douglas%2520Self%2520NE5532%2520amplifier.pdf

A 5532 should be able to power headphones all right, on the other hand.

Yes, lots of commercial products do this.