HP3400B RMS Volt Meter

[Steve Gibson]

Hi All

Does anyone have any experience with the HP3400B RMS Voltmeters?

I'm looking for the most cost effective route for obtaining an RMS voltmeter from 100mV to 300V up to the 20Mhz range?

I've not found anything of good quality that's not silly expensive yet, advice would be most welcome.

And if anyone has an old HP3400B surplus to requirements, please message me.

Many Thanks

Steve

[G0HZU_JMR]

The little Racal 9300 (B) true rms voltmeter works up to 20MHz and is a worthy contender. It works across an amplitude range of about 30uV to 300V.

I own one and use it quite often. It works really well across all waveform types including noise and modulated RF signals up to about 20MHz. There is also a BNC output connector on the rear panel and this can feed into a modern DVM if you need to log any readings or if you want less measurement uncertainty compared to the analogue meter.

The Racal 9300 dates back to the 1980s and typically costs about £50 to buy today. I think Stewart of Reading may still have some for sale in their shop.

[G0HZU_JMR]

I think the 3400B has an input impedance of 10Meg and the Racal is just 1Meg so this might be an issue for you. The specs are generally better for the HP3400B but I find that the Racal 9300 works really well for noise or for modulated signals.

I think the crest factor spec is superior on the 3400B but you may have to pay several hundred pounds for a 3400B today. The prices for old HP gear like this seem to have gone up a lot in recent years.

[IanG3XYV]

Do you actually need the B model rather than the perhaps more common 3400A which I believe has a bandwidth of 10Mhz rather than 20Mhz?

[Steve Gibson]

Thanks guys. I am primarily looking at how I might use it in switch mode power supplies, so I would ideally like the bandwidth - although not sure what sort of an issue the impedence might be.

I could probably look at designing a probe where high impedence might be useful

[factory]

Also consider the Fluke 8920A or 8921A, good for 20MHz too.
Never seen a cheap 3400B, perhaps that is because it's more recent than the 3400A, you probably never will now "The Signal Path" has done a video on one. https://www.youtube.com/watch?v=NfxY74W7rNY

David

[Steve Gibson]

Hi David

Yes, I saw that video and that's part of the reason why I am looking for one.

Although and alternative doner device may be just as good for my needs



Regards

Steve

[G0HZU_JMR]

One thing to note is that I believe the HP 3400 is connected to chassis earth on the outside of the input BNC connector.

The Racal is the same and this means that these meters are similar to a scope in terms of the BNC outer being wired to chassis earth. The Racal 9300 does have an isolation switch but it only isolates via some diodes. Therefore, both of these meters need to be used with care as the input isn't floating.

My Racal 9300 has a 1Meg input impedance and can be used with a regular x10 scope probe if the scope probe's internal trimmer has enough compensation range for the 30-40pF input capacitance of the input of the 9300. The frequency response is pretty good like this although I rarely use a scope probe with it.

[factory]

Just looked at both active & sold listings in ebay uk & ebay com, only one on there in the last two months & it sold for over $400.

Shahriar's previous video on the HP counter, resulted in all of them selling out, even the crazy priced ones.

David

[G0HZU_JMR]

I am primarily looking at how I might use it in switch mode power supplies, so I would ideally like the bandwidth - although not sure what sort of an issue the impedence might be.

I would have thought that a digital scope would be a much better tool for SMPS design work. The Vrms meter will just blindly tell you what it measures across its full bandwidth. Sometimes this can cause confusion if a small, low frequency signal is expected to be the only signal yet there is another unexpected signal at a higher frequency that is also present. At least the scope can display this problem. The 3400/9300 operator won't know this is happening unless they are experienced in issues like this.

I only skimmed the youtube video about the 3400 meter. I'm not particularly convinced by the LCD display modification. The meter will already have a DC output that can be fed to a logging DVM. Also, it is 'very' useful if the external multi-meter has a dB/REL function built in. This allows the reading to be normalised to 0dB on the external meter.

I'd expect the piggy LCD module to support a feature like this as this is (for me) the main reason for wanting to couple up a DVM to the back of the meter. This allows signal to noise ratio to be measured. It seems odd that the LCD meter doesn't support this common/basic function.

[trobbins]

I'm also a bit intrigued by how such a meter could be used for smps design/development, as compared to using a modern isolated digital scope. Perhaps an rms ripple voltage measurement of a port that needs to be wideband, such as from mains to 30Mhz to align with conducted versus radiated transition.