Electronics Surplus Warehouse Find!

[Terabyte2007]

Bored and sick of the arctic cold and snow, I decided to take a trip to the local surplus warehouse and found this unit tucked away on a shelf, a Yokogawa 2560 DC Calibration Set. Made a deal with the guy for $100 bucks and brought it home. Cleaned it up, turned it on and let it bake for a few hours and behold, it is still damn accurate! Most of the reference voltages are within 10 uV or less to 5 digits. The unit has impressive specs, .1uA to 10uA current resolution and 1uV to 1mV resolution at .02% across all ranges. It also has a %deviation for very easy display of your calibration tolerance of the DUT. HV module from 100mV to 1200V and precision uA current ranges. EMF compensation, Voltage divider and more! The unit is fully loaded with all options...

All ranges checked against my Agilent 34461A are still in tolerance! Amazing!

[calexanian]

Nice find. That sort of find is always the best!

[Vgkid]

Nice find,teardown requested.

[Terabyte2007]

Nice find,teardown requested.

Actually the teardown pics should be easy. There are two modules in the frame and they slide in and out from the front!

[xrunner]

Love that output meter! 

[lowimpedance]

Neat looking unit, will be nice to see the insides particularly since we don't see much of these less common brands on here.

[codeboy2k]

Nice find! I just glanced at the manual, the 2553 seems to use a DAC and a code value from the CPU to set the desired output voltage, but it uses an FPGA to shape the code run up, and then uses an ADC in the feedback loop to servo the FPGA/DAC combination to keep the setpoint.  Cool and very similar to the Jim Williams app note 86 using a 24 bit adc to servo a 20-bit dac.

http://cds.linear.com/docs/en/application-note/an86f.pdf



Their code comparator is inside the FPGA and it also does code shaping on code changes from the main cpu and in the feedback loop, to control rise times, overshoot, etc.

EDIT: the best thing about this architecture is that it's only dependant on the accuracy and stability of the 24-bit ADC and it's reference.  So an expensive 24 bit ADC with < 1 lsb INL/DNL and a stable reference can correct for a cheap, non-linear DAC, cheap non-precision output op-amps, and less than perfect output drivers.

EDIT 2: This application probably doesn't even need the ADC to be as good as I said above, since they only have 5 decades

[Smokey]

That's a pretty new looking cal sticker.  When was it last calibrated?

[radiomog]

looks like a nice find. congrats 

[EEVblog]

Sweet score! 

[rodpp]

Nice! Great price.

Does anyone have manuals of this DC standard?

[codeboy2k]

Nice! Great price.

Does anyone have manuals of this DC standard?

You can google for the manual, that's all I did. But the manual I found had no schematics, just a block diagram.

It would be nice to find the complete schematics.

EDIT: the spec sheet for the 2553 is now attached.  The manual I downloaded is 4MB, so it's over the 1MB limit.

I haven't been able to find anything on the other half, the 2563 .

User Manual is here

[rodpp]

Thank you. Please note that this manual is for the new version of the calibrator, 2553A and not 2553.

[Terabyte2007]

Did a bit more testing last night to check the %deviation compared to known tolerances in calibrated DMM's I have and the % of deviation was dead on to the tests. The cool part is you can easily check either +/- deviation. Also the voltage/current adjustments are actually optical encoders switches. The divider seems to also be right on the money too. I'll get some internal teardown pics shortly, I don't have time right now but I may tomorrow.

[rodpp]

Please see attached the 2560 set datasheet too.

But I could not find to download any manuals for the 2560 set or for the 2553 unit.