micro review - Keithley model 230 programmable voltage source

[Marcus_S]

May I introduce myself with a micro-review (please keep in mind, I am only a little hobbyist)?

For playing around with multimeters I wanted to have an adjustable voltage source with acceptable stability and a voltage range up to approx. 100 V. Although I found nearly nothing in the internet about the performance and although the noise-specs seemed not very impresssive to me I tried a Keithley model 230 programmable voltage source.

In principle this instrument seems to be a microprocessor which controls an A/D-converter (DAC80) with an internal reference which feeds an amplifier. That's all.

The voltage source has four ranges (+/- 199,9 mV, +/- 1,9995V, +/- 19,995V, +/- 101,0V) and can deliver and sink up to 100 mA. The output voltage can be dialed in with a resolution of 12 bit using the buttons on the front panel. The complete documentation with schematics can be found in the internet.

Most of the functions of the instrument can be programmed using the front panel...


(K230_front.jpg)

but unfortunately all the connectors are on the back...


(K230_back.jpg)

When powered on the voltage source needs approx. one hour to reach constant output voltage:


(K230_10V_warm_up.jpg)

For increased resolution I measured the output of the K230 as the difference to my cheap 10 V-voltage standard (Ref-102) with an old Rohde & Schwarz UDL45 Multimeter (uncalibrated, 5,5 dig, = Analogics DP-100). Currently the UDL45 ist the only multimeter which I can connect to a computer.

When warmed up the voltage stability is satisfactory:


(K230_100mV.jpg)


(K230_1V.jpg)


(K230_10V.jpg)


(K230_100V.jpg)

For higher resolution in the 10 V-range I measured the stability as the difference to the 10V-Standard:


(K230_10V_vs_Ref102.jpg)

This noise of approx. 5 ppm (peak to peak, sum of K230 + UDL45 + Ref-102) corresponds to measurements using a HP3456A (uncalibrated) and its statistic program:


(K230_noise_3456A.jpg)

And for special applications it might be interesting to use the program mode. Up to 99 steps can easily be programmed using the front panel and the program can be executed stepwise or once or continously.


(K230_program.jpg)

I am happy with this instrument  .

[TiN]

Nice box, thank you for efforts and tests.
Perhaps some internal photos would compliment micro-review into full review later on? 

[quarks]

Very nice review

Thanks for sharing

[Performa01]

Finally a review that deserves the name, demonstrating the aspects of the instrument that really matter. 

At some point I thought about getting one of these old Keithley voltage/current sources, but the noise and stability specifications always held me back, especially considering the ridiculously high price offers usually found at 2nd hand dealers and even eBay.

So your review confirms that his unit can be a nice tool for many tasks, including calibration, but certainly nothing close to volt-nuts territory

Thanks a lot for the effort to share your findings!

[Marcus_S]

Thank you very much for your comments. Yes, very little performance reviews can be found regarding Keithley 220/230. And I was a little bit worried about the noise and the drift specifications too. In reality it is a nice instrument with good stability. Due to the 12-bit D/A-converter the accuracy and the resolution are not exceptional. But it is useful for experiments in the 5,5-digit-(10 ppm) range.

As an example I used it for playing around with adjustment of a Solartron 7150plus (reference meter and 7150plus connected in parallel): no problem with the higher ranges (100V, 20V, 2V). But the noise of the K230 is too high to adjust a 7150plus in the 200 mV-range directly, therefore I used an additional divider. With some luck the K230 is even usable to adjust a 3456A to another meter (both connected in parallel) to be within 1 ppm at 10V.

It would be interesting to compare this old instrument with some newer source meters regarding noise and drift. Unfortunately, I do not have one.


Best regards

Marcus