Keithley 2015THD Initial Tests

[G0HZU]

Hi
I just bought one of these 2015 THD meters on ebay and it arrived yesterday and I have a few questions...

I'm mostly interested in using it for automated measurements via GPIB to exploit the onboard DSP to measure S/N and also to measure distortion.

I've started playing with the GPIB interface today using Visual Studio and a USB/GPIB cable and I wonder if anyone else has played with this to measure THD etc? It's pretty easy to make basic measurements using SCPI commands for DCV or ACV etc but I'm not doing so well with THD.

The THD function works OK but I'm struggling with some of the more exotic SCPI commands. eg I want to measure the frequency of any external test tone when it is in auto mode for THD and display the result on the PC screen.

Has anyone been here already and got these THD commands to work? I can get it to tell me what THD mode it is in and I can do a basic THD or SINAD measurement via GPIB but not all of the SCPI commands seem to be as easy to use correctly and the system is very unforgiving when it dislikes my attempted command

I have another question... When using the SCPI commands for stuff like DCV the meter appears to have a 9 digit resolution on the data.

The documentation hints that this meter has the same 'engine' as other Keithley meters so does this mean the meter can deliver more than 6 1/2 digits via GPIB?

I don't really need 6 1/2 digits let alone 9 but I've got it repeatedly measuring a cheap 5V voltage reference via GPIB and it is reporting it to 9 digits via GPIB.

Are the extra digits just digital gunk or can the meter actually deliver higher resolution via GPIB?

Thanks
Jeremy





 

[G0HZU]

It's OK, I've managed to work out how to get the more exotic THD commands to work

For example, it's possible to also ask it over GPIB for the amplitude and frequency of the test signal whilst it is doing THD measurements if it is put in AUTO acquire mode. eg using commands like :SENSe:DISTortion:FREQ?
and  :SENSe:DISTortion:RMS?

It's quite a versatile meter and I can inject my old function generator in and change the frequency and the 2015 can auto track the frequency change and still measure THD and also the dBc of each harmonic and the PC can display all this info onscreen as if it is all in real time.

This is much better than my old system because I can now see all this data onscreen and log all this data to a file on the PC.

[G0HZU]

I managed to do a few tests of the meter for THD on a really high quality square wave.


If the meter is only set to measure THD for 8 harmonics then the THD can be predicted to be -7.657dB with a simple excel spreadsheet. (2015THD meter actually measures -7.650dB THD)

If the meter is only set to measure THD for 20 harmonics then the THD can be predicted to be -6.804dB
(2015THD meter actually measures -6.797dB THD)

If the meter is only set to measure THD for 50 harmonics then the THD can be predicted to be -6.503dB
(2015THD meter actually measures -6.497dB THD)


You can see from the screenshots of the simple harmonic viewer program for each case the meter pretty much nails the measurement spot on. You can also see how good the 'test' square wave is because the even harmonics are very well suppressed.

Everything in the screenshots is based on lots and lots of measurements taken with the 2015THD  and the results plotted in the graphs below. So the viewer looks like a spectrum analyser even though it is really just a lot of measurements at each harmonic inside the DSP in the meter

It can produce that screenshot in about a second although the 50 harmonic capture takes a little longer because I have to take lots of data from the meter via the GPIB interface.

[G0HZU]

Here's a couple of measurements taken with the Keithley 2015THD where it tests the THD of two of my audio generators. The 2015THD is being completely controlled (via GPIB) by the PC software and it can produce that harmonic spectrum plot from the 2015THD in under 1 second. It also auto senses and measures the frequency of the source and the true  rms voltage of the source and displays this info on the PC screen.

The first generator is an old Black Star Function generator set to give a sine wave at about 1kHz. It's pretty good with harmonics down at -50dBc.

The second plot is from my Marconi 2024 signal generator at the AF output socket. This 1kHz AF signal is generated in DSP inside the Marconi sig gen and it has very low distortion as you can see. Typically -90dBc in the measurement!

This wonderfully versatile 2015THD meter is an absolute steal on ebay at the minute because you can get them for just over £300 from ebay seller KJDS. I plan to explore the SINAD measurement next and also see how good its own inbuilt signal source is in terms of THD. I expect it to perform spectacularly well

[quarks]

Very interesting, thanks for sharing!

[bingo600]

Very nice ...

I have a 2015 also ... Would it be possible to get the sourcecode made for the meter readings ?

TIA

/Bingo

[KJDS]

They are good meters. I'm glad that people are using it's full functionality.

[nickm]

Can you post measurements for a 20kHz signal showing how many harmonics it can do?
Maybe THD+N measurements too.  It seems nice, I'm just wondering it it can do some of the tests than more expensive measurement equipment can do.

[G0HZU]

They are good meters. I'm glad that people are using it's full functionality.

Hi Kev

I'm very impressed with this meter you sent me I bought it for the THD/SINAD functionality and also the decent AF source it produces. It should also serve as a pretty good gold standard to help keep all my other meters in reasonable calibration. I'm not really a calibration junkie so it will do me fine for quite a while before I think about having it checked formally.

It's also going to be useful for remote logging via GPIB from some of my older analogue gear that produce a DC output at the rear. Up until now I've been using a Fluke 45 for this but it only has RS-232. Having GPIB on everything makes the software a lot easier to manage.

Hi Bingo and Quarks
The little GUI I've produced is very basic and has virtually no error handling so I don't think it's fit for any kind of release.
It's also only a day or so old and maybe when I've smartened it up a bit I'll offer it for trial. I've since added a marker facility to it so I can now click on each harmonic and get its distortion contribution in -dBc and % and also the number and frequency of the relevant harmonic.

I've also had a quick play with another little remote GUI that sets the meter up to do lots of averaging internally. I've done this so I can have a look at the 9 digit readout from the SCPI command I'm using for raw DC voltage.

With lots of averaging (in the meter) the 9 digits that are sent from the meter do appear to climb as if the data really is to this resolution. i.e. digits 7, 8 and 9 do appear to count up as the average settles. I don't know yet if these extra digits are just due to the averaging maths inside the meter or if the meter can get this resolution from a single shot measurement. I can't see me ever wanting more than 5 or 6 digits from the meter but it is interesting to see what it can deliver

If I turn off the internal filtering/averaging with the command Keithley2015.WriteString ":sens:volt:dc:aver:stat off"
then it still spits out 9 digits but the last three digits jump around a lot (presumably due to noise) and are effectively of no use unless I do some averaging in the GUI on the PC.

Hi nickm

I think the THD is limited to 50kHz max and also 64 harmonics max. So if I want to see 64 harmonics I have to make sure the fundamental tone is less than about 750Hz. If I set the GUI to 64 harmonics and inject in something like 2kHz then it will show a blank for all harmonics above 50kHz.

If I feed in 20kHz then the meter can only see the second harmonic at 40kHz. I think it really is just for audio analysis.

[Clint]

Hi Jeremy, I was also so impressed by these meters I have purchased a second uint !

I would be more than interested in your software, should, when you are ready be willing to share.

9 Digits is very interesting, I too would have no need for it but its nice to be able to add it into a conversation with my wife, I already bore her silly with my designs.

Regards
Clint

[G0HZU]

Hi Clint

I'll have another look at the SW to see if I can make it install on Win7 etc. It's written in a very old version of Visual Studio so I doubt this will be that straightforward because I might miss some support files that won't be on your PC etc.  The other thing is there is no method to change the GPIB address so it is locked to GPIB address 22 because that's what my 2015 is set to. But this is easy to change to address 22 in the meter anyway.

Note: I also wrote a Windows app for the 2015 that makes it act like a Toneohm over GPIB. i.e. the meter measures down to milliohms and reports this over GPIB very quickly.

So by exploiting the speed of the 2015 over GPIB I was able to get a really decent ToneOhm sound from the PC speakers for tracing shorts on a circuit board It really does sound like a real ToneOhm (or at least how I remember a ToneOhm) The tone from the speaker lowers in pitch the lower the resistance measured by the meter.

To get a smooth audio effect in Visual Studio I had to use MS DirectX8 and this is a bit of a pain because the VB DLL for DirectX8 isn't supported in Win7. It is possible to manually install it and register it because I got it working for my Win7 PC here. But it does make it difficult to produce a smooth install process

Also, I didn't bother making the ToneOhm GUI look good because I doubt I'll use it very often. It was just fun to create it and prove it could be done

Regards
Jeremy

[casinada]

Now you have to figure out if you can convert the 2015 to 2015P
Great job with the software

[RandallMcRee]

I know this is an old thread, but....

Can anyone share a simple SPI script to retrieve THD measurements from a Keithley 2015?

Don't need full-blown software with GUI, just some examples would be nice.

Thanks!
Randy

[Tony_G]

I know this is a really late reply but I answered a similar question on a different thread just now - here is a very simple C# program that gets the THD and first 4 harmonics:

Program.cs

TonyG

[frankmhowell]

Hi Jeremy,

Is your Windows application for the Keithley 2015 THD available anywhere? Just got this unit and would like to use it on my workbench.

73,

Frank
K4FMH

[emartine]

I will purchase this meter and upload an app to github, so anyone can use it..
Im already working on it.
This is my github page if you wanna help:
https://github.com/andmarti1424

[emartine]

This is the progress, including auto update.
Data is simulated.
Now i should get the unit..