Calibration Lab Chat

[@boonedockjourneyman7979]

I enjoy your content. You add value to the community. I hope you come back.

[@tgsoapbox]

Thanks appreciate that - Things have been in great flux for the last 2 years but I have a bunch of content ideas so I hope to get back in the saddle soon.

[@SeanBZA]

I had that problem of obsolete parts. ordered 10 transistors, and got a call around 3 months later that they were obsolete, but that the original manufacturer was willing to make them, but would require a minimum order of 1000 trays of the specific TO3 transistor, 100 per tray. Also the per unit price would be, in the 1990's, $15 each, with a 6 months lead time. as there were acceptable substitute devices available ( NJPN power transistor, Ic 6A, Vce 100V, hfe min 30@5A) that met or exceeded the required spec, the order was cancelled, and we did the paperwork to get the alternates approved. They were in stock, and also available in trays of 100, and were already qualified for the airframe as well, plus they were perfect as ground support spares as well, exceeding the old ancient 2N3055 in all respects by far, making repairing the GSU a lot easier, as they would not die from cooking themselves. The equipment in the plane also was wire wrapped, as they had a prototype, and the production run was maximum 90 units from the serial numbers, so all were simply wire wrapped. Most reliable part was the wire wrapped backplane, though if a pin broke you had a lot of work to get that pin out of the socket and replace it, but that was easier than removing 200 wraps and redoing them, especially as the majority were wrapped with a modified wrap, then soldered. Test bench was wire wrapped as well, single prototype, and repairing faulty sockets was easiest by simply soldering another into the old one, as that wire wrap was up to 10 wires deep in many places. Lots of TTL, and Numitron displays. We did upgrade one to red HP common cathode displays, and after soldering in the near 200 330R 1/8W resistors into a cordwood panel, we found the LED's were too bright, so a quick upgrade consisting of a pair of 1N4001 diodes, in series with the common ground for the non multiplexed display. Each segment was driven via a pin on a CD4050 buffer, and the inputs came from a latch. Those diodes ran hot on display self test, but otherwise were fine with less than all segments on. Did learn how to program fusible link PROMS there, using a breadboard and a power supply, and lots of resistors and wire links to set up address, data and pop the fuses manually. Only around 20 bits max per unit, so not too onerous, and most actually worked as well.

[@tgsoapbox]

Thanks for the comment - min production run quantities are sometimes simply staggering in the amounts required.

[@laurens4359]

The 8481 and 8482 have very different low end ranges due to the coupling caps.

[@tgsoapbox]

Good point completely forgot about that - 8181A is 10MHz and the 8482A is 100KHz.

[@CoolMusicToMyEars]

Fluke 5220A was taken off due to not meeting the CE Rules the Fluke Rep informed me, that was a Great unit, BUT would not go well with the Fluke 5500A Coil, due to low compliance voltage output on the Fluke 5220A, There are great GPS unit's like Jackson Labs that make a great GPS Standard with low drift Oven Oven Xtal, when signal is lost it can look at the drift, for minumum drft over the time when signal is lost, Come the year 2000 not many manuals had full service information like the old days 1960s test equipment, I will be honest I prefer the older generation test equipment, as it Works just as well as modern equipment, but you have more chance to repair older generation equipment !

[@tgsoapbox]

Thanks for the comment - I'll have to look at the Jackson Labs units as I'm using a HP one that came out of a cell tower installation - I'm happy with it but you can't go to far wrong by getting a better timebase - I also agree that it is a shame that they don't make manuals like they did back in the 60s to 80s