Good price for an HP 3478A

[tom66]

Sold used, as is, untested. £75.00 plus £15.00 shipping. Picture shows the unit powers up and displays a seemingly random voltage... I guess it's because of the high impedance frontend.

Have been on the look out for a good bench meter for some time, looks perfect if I can get it for cheaper (it is in auction.)

But it could be a lemon? How can I tell?

[joblessalex]

I probably won't be able to tell, but is there a link to this?

[tom66]

Here's a picture of the item in question - the seller says untested, unsure how to test.

[Bored@Work]

the seller says untested, unsure how to test.

Be very careful. While it might be true you also find sellers on the internet who have sold hundreds of instruments and still claim they can't test or don't know how to test any kind of instrument.

[baljemmett]

Here's a picture of the item in question - the seller says untested, unsure how to test.

Hmm - the input selector switch looks a bit odd - on mine it protrudes similarly to the power switch.  Not sure if that was something that varied over the lifespan of the model (it would seem odd to be recessed!) or if it's a sign of mistreatment.

The unit should display SELF TEST OK at power-on, or will show UNCALIBRATED if the battery used to store the calibration has run down.  In the latter case I believe it's possible to recalibrate it following the procedure in the service manual, but you'll need other kit that's suitably accurate if you want to trust it afterwards.

Generally in this sort of situation I'm looking for a seller that a) mentions things like the self test message at startup and/or b) will accept a return if it turns out to be buggered, simply because those give better vibes than the "hey here's a thing I found, I know nothing about it" crowd.

[ModemHead]

The pushbutton switch that selects front or rear inputs is definitely broken. Not much of a problem if you use the front inputs only.

If it passes self-test and then displays a random voltage, it probably works OK. These units are well-designed and made, so I'd say the odds are good that it's useable.  There is a 3V lithium battery in there that maintains calibration constants that could be at or near the end of its life though.

A couple years ago I bought two of these units, each for about 75USD + shipping. One was supposedly calibrated, one wasn't, but they (still) agree within 2 to 3 counts on DC volts, and also with a new Fluke 87V, so I trust them for my purposes.

I do question those sellers that say they can't test a multimeter. How much knowledge does it take to connect a 9V battery and see what it displays?  Check the seller's history. If they've been selling baseball cards and coffee pots, then maybe they really don't know how to test it. But if they've been selling electronic test gear, then the odds go up that they know it has a problem and are just hiding behind the weasel words.

[T4P]

Ah well, apart from the input selector switch i believe there is definitely something very dangerous/rendering it useless that they are hiding

[alm]

The random voltage with no leads connected on DCV mode in ranges <= 3V is normal, that's because of the > 10 Gohm input impedance. I would say there is a good chance that it's working fine, they are very well designed. I would try to get information about the self test that baljemmett suggested, and maybe also some basic tests (short the inputs or connect a battery, for example). They seller may or may not be willing to do any work on this. Small sellers are usually more willing, even if they lack knowledge, than large power sellers that try to shift large amounts of items quickly and with minimal overhead costs. Some of the latter charge a premium price for the privilege of doing business with them.

[bingo600]

This model doesn't like voltages above 300v.

Don't ask an unexperienced seller test it on "mains" ....

Also make sure it doesn't display "UNCAL" (can't remember the exact text)  on power-on , then the battery is dead. You'd need to replace it and then pay for a calibration , witch prob is more than the meterprice.

/Bingo

[radiogeek97]

Digging up an old thread here but I was looking at an hp3478a that is displaying uncalirated after power on.  I was wondering if anybody has calibrated one of these on their own?  Is it possible to just replace the battery and use the instrument for only rudamentary rough measurements without calibration? 

[c4757p]

No, they are designed to require the calibration, so it will be absolute crap with the "default" (zero) values. You'd have better accuracy if you took an old analog voltmeter out of Grandpa's closet and tossed it to the floor a few times before using it.

I have calibrated a Fluke 45 to match my HP 3468A, and the procedure for the Fluke cal seems similar to the procedure for the HP. It's a tedious process, you're going to have to find a lot of stable sources to measure (including at the edges of the meter's capability - max voltage, max current, etc., and stable - do you have a stable source of 300 VAC or 3 AAC?) and a good meter to measure against.

[radiogeek97]

   glad i got outbid fleabay had one uncalibratd for likr 50 bucks BUT it is way more trouble than its worth considering my abilities and the fact that i can just reach into my drawer and pull out a fluke 289      No wonder people go thru great pains to replace the batteries carefully i these


thanks for the heads up