It’s so nice to hear folks talk about doing it right. My current company and it’s parent both have(had) a policy that if we actually repaired an instrument it was call’s to zero. I’ve seen many instruments sent to us as BER, from a customer who is required to use a specific cal lab, that only required going through the published procedure with attention to detail.
toggle quoted messageShow quoted text
Datagate Systems, LLC
On Feb 5, 2021, at 17:54, Jim Ford <email@example.com> wrote:
Hi, Don. Yeah, I said that in jest. Calibration professionals are professionals after all. Not that there aren't cal shops that do pencil whip the job or don't know what they're doing, of course. I had a 453A scope calibrated some years ago, and when I got it back, I noticed that channel 1 and channel 2 had very different behavior. One channel showed much different low-frequency rolloff than the other, but only in AC coupled mode, and only at IIRC in 0.1, 0.2, and 0.5 V/div settings. I traced it down to a variable cap being misadjusted and was able to tweak it to make both channels the same. I never used that cal shop again.
I used the CSA803 and SD-24 plug-in back about that same time frame, and I can only imagine how detailed and time-consuming the cal procedure must be.
------ Original Message ------
From: "Don Bitters via groups.io" <firstname.lastname@example.org>
Sent: 1/31/2021 5:24:11 PM
Subject: Re: [TekScopes] bandwidth
I was one of those HP/Agilent field cal guys doing the MTE, never did it for Raytheon, but did it for 100’s of other customers. We verified that every instrument made published specs - no matter what manufacturer HP, TEK, Fluke, etc. we never adjusted any MTE without notifying the customer - he/she had to authorize it and quite often had to provide the adjustment procedure if it was a manual procedure.
I did the manual verification procedures on about 100ea. 11032’s, CSA803’s and plugins. One time I was pressured by a high level manager to “pencil whip” the test procedure to speed it up. I pointed out to him that I would not/could not do that because of the damage to our reputation, let alone the legal liability if I passed something that actually failed. You know how many pages the manual test performance for the 11032/CSA803 is - about 75, and the SD-24, SD-26 plugins
- about 30 pages. Doing the performance procedure on an 11032 with 2ea. SD-26 plugins was most of a day calibration without the additional paperwork. I did not know of any HP/Agilent/Keysight techs, engineers that would “cheat” on a calibration, and I knew well over 100 of them personally to talk to. I also doubt that any of the Tektronix, or Fluke, or Dayton calibration people that I knew, and even a large number of the independent cal labs that I dealt with would also “cheat” on a cal.
On the other hand I did know of a few labs that were notorious for selling the customer the cal stickers and the certificates for a fee.
One of our NA customers went looking for a cal lab in Mexico (Agilent-Mexico declined, they had neither the ETE nor the trained people to do the required tasks).
They found a guy that had a small shop above a pharmacy in Guadelejara that said he could do the job calibrating production line optical test systems. He had a 100MHz o-scope and a 4-1/2 digit DMM. The NA customer hired Agilent-Midwest to go down to Reynosa, Mexico for 3 weeks to calibrate his calibrating production line optical test systems, for a princely sum, but got it done to NIST traceable, with data.