So how much did you spend on these fine old devices not to have faith in what they can do ?
Less than 300 AUD for both instruments.
The first step in acceptance is Performance Verification with is also the first step in Calibration
That explains why these chapters come right before the calibration chapters!
quote author=tautech link=topic=345679.msg4440607#msg4440607 date=1664479610]
There was a time I spent more time repairing my equipment than enjoying the hobby and that really sucks !
Yep, that’s the lesson I’m learning.
For private usage not, that´s right.
For curiousity, I´ve send my new brymen 869s meter for external calibration and got it back with a 8-pages calibration report.
They checked every range and found some deviations and re-adjusted them...
So now I know it really instead trusting the specs blindly.
In your case, in the case of the HP, a calibration would be useful to know if everything is alright.
I don´t know the prices in your country, but I´ve spend under 100 bucks for it and for me it was worth it.
What the tek scope concerns, hmm..
I think I would buy a newer DSO instead of spending time to fix it or to be uncertain about it´s fully functioning.
Maybe I should do some more digging, maybe I’ll find a lab that might just test the devices or something.
Most instruments have service manuals to tell how to verify, but they are not always available to public.
It cost a bit, but I managed to find a service manual for the Tektronix on eBay and I found a free scanned copy for the DMM.
As was suggested in the link my original post, guenthurt gave a link to a cheap Kelvin Varley project. So that will run the DMM through its DC Voltage paces. Just need all the other stuff now.
Only you can answer this. Because you are the only one that knows your use case. And the problem is not solved by buying new equipment. Because you need to understand your use case to know how good that equipment needs to be. Do you need a 3458 to get accurate, many digit measurements of voltage, resistance or current? Or do just need to know that the measurements are in the ball park? Do you need absolute accuracy or do you need linearity? Do you need to compare measurements made months or years apart with high precision?
If you don't know what you will be doing, the safest answer is to buy the top of the line new, and pay for regular calibration and adjustment. For most of us budgetary limits don't allow us to be that safe, and we have to trade the use of valuable brain cell operation for those unavailable dollars. Remember that most of the progress in electronics (starting from Faraday and up through Armstrong and Nyquist) was performed with instruments that provided only two or three digits of absolute accuracy assisted by bridges and other tools that allowed excellent differential accuracy.
As james_s suggested old A list equipment is either close to spec or totally broken. In my case I’m not doing rocket science so I just need to be close. My main concern is that the units are fully functional so that I know 10 volts is roughly 10 volts etc.
Without the instruments for it one can obviously not do a full calibration or performance verification. However there are a few points that can be tested with moderate / low costs.
For the DMM this are things like offsets and zero drift, the input bias current, the same voltage in 2 adjacent ranges, some DNL tests , the turn over error, some INL test in a resistance range (chain of resistors).
Even if no good enough for calibration or with even high uncertainty it also helps to compare to existing meters. This would not detect errors from a more contineous drift over time, but it could detect many cases of defects that usually cause more larger errors if they happen.
Not familiar enough with multimeter internals at this stage. No idea how ADC DAC works in a multimeter, also quickly looked at INL & DNL, will have to do some digging at some point to get acquainted with these topics.
For the scope the situation with the calitration is even easier. The voltage accuracy is not very high anyway, time frequency references (e.g. crystal clock) are relatively easy and dividers down to lower frequency are accurate by design. For the frequency response the simple clean square wave for the probe compensation is a first, quite good test.
Yep, in my original question, others said the same i.e. oscilloscopes aren’t precision devices. I’ve already played around with the probe compensation sin waves. All the sin waves I’ve thrown at it, seem to have made sense so far.