Stanford Research DS360 calibration failure

[Jwalling]

A friend of mine just sent out a DS360 for calibration at Tektronix. It failed.
Oddly enough, a number of the parameters were at the edge of passing, but still within SRS specification, yet Tektronix listed them as FAIL.  Some others were a bit on the high side. Even the ones that were passing were at the hairy edge (high)

I've attached the calibration report.

My questions:
1. Why did Tek list the passing parameters as FAIL?
2. Did they really attempt a calibration, or did they just run a performance test?
3. If there is something wrong, my guess is maybe a resistor may have gone out of tolerance, yet I think a *true* calibration would have compensated, yes?
And the longshot:

4. Does anyone happen to have schematics? In my past dealings with SRS, they are not helpful at all with parts or service data unless you are the original owner and bought direct from them. In other words, they suck.

Thanks.
Jay

[tec5c]

1. Why did Tek list the passing parameters as FAIL?
2. Did they really attempt a calibration, or did they just run a performance test?
3. If there is something wrong, my guess is maybe a resistor may have gone out of tolerance, yet I think a *true* calibration would have compensated, yes?

1. Because the tests didn't meet the standards... I'd be more concerned if Tek. passed it. At what point then would you say enough is enough.

2. & 3. A calibration is a comparison between a set/known standard and another piece of equipment. Commonly mistaken for servicing the equipment in order for it to pass the calibration test.

[ElektronikLabor]

See: JCGM 200:2008 International vocabulary of metrology Page 28 §2.39:

calibration
operation that, under specified conditions, in a first
step, establishes a relation between the quantity
values with measurement uncertainties provided
by measurement standards and corresponding
indications with associated measurement uncertainties
and, in a second step, uses this information
to establish a relation for obtaining a measurement
result from an indication.

NOTE 1 A calibration may be expressed by a statement,
calibration function, calibration diagram, calibration
curve, or calibration table. In some cases, it may consist of
an additive or multiplicative correction of the indication
with associated measurement uncertainty.

[dom0]

<goes totally off topic>

That's an interesting device you have there. Looking at the THD vs frequency graph:



Do you see that blip around 1 kHz? Over 20 dB change in distortion back-and-forth in a couple dozen Hz span! Very interesting. I guess they implemented either some very delicate noise shaping in their DDS or added some additional analog filters for specific frequencies to boost distortion performance at exactly 1.000 kHz. Also look at the distortion peaks at n*1 kHz.

They don't mention this explicitly in their specs except for that graph and a small "Anti-distortion filters" block in the block diagram.

[Jwalling]

1. Why did Tek list the passing parameters as FAIL?
2. Did they really attempt a calibration, or did they just run a performance test?
3. If there is something wrong, my guess is maybe a resistor may have gone out of tolerance, yet I think a *true* calibration would have compensated, yes?

1. Because the tests didn't meet the standards... I'd be more concerned if Tek. passed it. At what point then would you say enough is enough.

I don't understand this. For instance, unbalanced output at 1 kHz, Nominal = 0.400Vrms. The Max is listed at 0.404 and the measured amount is 0.404, yet this fails.

Thanks.

Jay

[dom0]

The measured value might be rounded down for the spreadsheet.

[Macbeth]

Do you see that blip around 1 kHz? Over 20 dB change in distortion back-and-forth in a couple dozen Hz span! Very interesting. I guess they implemented either some very delicate noise shaping in their DDS or added some additional analog filters for specific frequencies to boost distortion performance at exactly 1.000 kHz. Also look at the distortion peaks at n*1 kHz.

Are you thinking what I'm thinking? Some morally bankrupt engineering going on here? Is SRS a division of Bosch? 

As for OP, you should ask for a calibration with adjustment - then all the columns headed "as left" would hopefully be within specification.

[tec5c]

Or rather if you want a definitive answer, why wouldn't you (your friend) ask the people at Tek who calibrated it why it failed?