Hi all, I purchased my first scope about two years ago (a Phillips PM3217). It seems to be a very capable scope and has worked well for most all my trivial needs. I recently went about finding a service manual after reading about the need to calibrate analog scopes for accuracy. After reading through the service manual I discovered that I would need several more pieces of test gear just to begin trying to calibrate my scope (I don't need extreme accuracy but having faith in my equipment has to be worth something).
My question is: how often does an analog scope such as mine need to be calibrated? I.E. should I go about purchasing a func generator and freq counter or try and borrow one for a one time calibration. I can hear the obvious answer already "Calibrate when it needs it", but in typical use cases, with maybe 10/15 hrs of use a week living in my non-climate controlled garage how high would the need to re-calibrate regularly be?
Please forgive me if I've overlooked these answers in my searching.
My question is: how often does an analog scope such as mine need to be calibrated? I.E. should I go about purchasing a func generator and freq counter or try and borrow one for a one time calibration. I can hear the obvious answer already "Calibrate when it needs it", but in typical use cases, with maybe 10/15 hrs of use a week living in my non-climate controlled garage how high would the need to re-calibrate regularly be?
Please forgive me if I've overlooked these answers in my searching.
I never calibrate mine.
If it started to give odd results I might look at it.
You could do some limited test using a meter to check AC and DC voltages.
If it started to give odd results I might look at it.
You could do some limited test using a meter to check AC and DC voltages.
wow, really? Even for a scope as old as mine? I felt sure that with the temperature fluctuation, the moving about, combined with age, these old scopes would fall out of spec rather quickly (one or two years?)
I know this dilemma very well. "Calibrate when it needs it" seems an obvious answer to this, however, the problem is how can one know when calibrating is really 'needed'? When it`s already obvious that the measured results can`t be right, it`s not a measuring instrument anymore but just some kind of 'guesstimeter', calibrating is long overdue.
Less obvious errors might go unnoticed. This can be even worse than knowing that the measurement isn`t exactly right, as it could lead to false conclusions.
Unless You intend building up some kind of calibration lab yourself, I`d adwise against buying more gear only to get the scope calibrated because Your problems would just multiply. As You figured already yourself, usually You need a lot of instruments and special calibration fixtures to get this done properly. Those instruments and fixtures, of course, have to be calibrated as well.
You`d end up calibrating the calibration fixtures & instruments for the calibration fixtures & instruments which are needed for calibrating the scope......an endless and costly task.
If the scope is in good condition my advice therefore would be to get it once calibrated in a calibration lab and pay for it (provided that the costs for this are reasonable versus what is worth the instrument). In the commercial or scientific world measurement instruments get calibrated usually about once a year, not necessarily because it is 'needed' but just to verify that calibration is still valid and the instrument performs still inside the specifications. Well cared quality instruments like Tektronix, HP, Philips etc. in good condition can keep their calibration usually for a long time (years, sometimes even decades).
Hence for a hobbyist a one time calibration might be all You need for the next decade or so.
BTW:
Humidy is the among the worst You can do to an electronic instrument, therefore a 'non-climate controlled garage' is NOT a good place for storing quality measuring instruments (or any other electronic instrument for that matter). By any means do Yourself (and the scope) a favour and get it out from there into a dry place.
The only thing which will need periodic adjustment is the trimmer capacitor on the scope probe. There is probably a node on the front of the scope with a square wave output and the probe trimmer capacitor is adjusted for best symmetry.
If you have and use a scope, it's not unreasonable to have a signal generator and a frequency counter. I'm a big believer in frequent reality checks. If the scope agrees with the generator, or better yet, a counter, your confidence level goes up dramatically. You can also put a decent voltage reference chip in a box and measure it with a good DVM. If the scope agrees with that, and if you adjust the probes properly, your measurements should be pretty good. Does the scope by any chance have it's own signal to check probe compensation? Use that. A good tech will do a quick reality check between the equipment in use often, maybe every time a new measurement setup is done, or before anything really important. Once a year calibrations are nice, but you need to be more diligent than that to catch problems early when they're easy to fix.
CH
CH
This is great info! There is indeed a cal terminal on the front panel, the spec is 2kHz @ 1.2v +/- %5. In the calibration procedure, this terminal is among the last to be adjusted (I assume due to the effects previous adjustments would have), the wave doesn't quite sit at 2kHz, though it is within the 5% spec. I guess the perfectionist in me wants to see that wave split those division lines down the middle. Is this unreasonable?
I do like the idea of "reality checks", in that if multiple instruments agree with one another then things are relatively good.
I do like the idea of "reality checks", in that if multiple instruments agree with one another then things are relatively good.
Remember, until modern digital scopes, scopes were not the preferred device for precision quantitative measurements if there was anything else that could do the job. That's why we have triggered frequency counters and nice DVMs. HF AC amplitudes are often a bit iffy depending on strays, so a scope is/was quite useful even at 5% accuracy. I go through my old Tek boat anchor tube scopes every 5 years or so, and they stay plenty accurate. They have nice new digital scopes at work and people use all the built in measurement features about 98% of the time. I'd rather have a real counter, but the few digits they get on the scopes seem to be enough for most tasks.
CH
CH
I have a Tek 453 to which I've never done a thing except for running some No-Noise into a couple of the pots that were beginning to lose it. Aside from the built-in fixed sig gen and probe-trimming, are there other things to consider with this old beast?
--Bob.
Remember a scope measures voltage 🙂 so use a DVM and DC PSU and make sure the trace moves the appropriate amount... easy 😉
For horizontal timebase depending on your location the mains should be pretty accurate at 50 or 60 hz... OK for a checking a few lower ranges, for higher speeds use a crystal oscillator and divider.
Or... many multimeters have a frequency range, so use that and a signal generator.
You should be able to get it to "as accurate as you can see the trace" taking into account parallax errors.
)
For horizontal timebase depending on your location the mains should be pretty accurate at 50 or 60 hz... OK for a checking a few lower ranges, for higher speeds use a crystal oscillator and divider.
Or... many multimeters have a frequency range, so use that and a signal generator.
You should be able to get it to "as accurate as you can see the trace" taking into account parallax errors.
)
I don't remember specifics on the 453, but if it has a fan, clean the crud out and clean the filter. I use compressed air to blow off everything else, but not in the winter when the humidity is low and static potential is high. The auto stores often sell little 3/4" diameter "dashboard brushes", and I use those to get between tubes and such. Any controls getting stiff get some Breakfree or similar lube at the bushings. Noisy controls and switches get a shot of DeOxit. Everything gets a good visual inspection with a magnifier- look for cracked solder joints and wires on the verge of breaking. Look for toasted resistors and leaking caps. If the PS voltages aren't very close to factory specs, I check caps. I don't believe in replacing anything that isn't proven to be faulty- I want to see low value or high DF before pulling things apart because you can do a lot of hidden damage flexing old wiring and such. Those old Tek, HP, GR, EH, Kepco, Keithley and Fluke instruments performed brilliantly when new and it's important to know what's normal so you can get after problems when they're still minor.
Happy New Year!
Conrad
Happy New Year!
Conrad
Yea, really. We hobbiests can't afford meteorology services. It cost more than the used scope to start with. So, used 4 1/2 digit true RMS multimeter, used freq counter, used HP generator, and PC based tools for everything else. Then just dig out the o' Tek when I want to see the waveshape or relationship. No software scope can work anywhere as well as a real one for my pulser I use to measure speaker time alignment. Zalescope is close. Just wish they made one that could run 24 bit and 192K.
So, if you can set your probes to a nice square wave, be happy, don't worry. They were never that close to start with.
So, if you can set your probes to a nice square wave, be happy, don't worry. They were never that close to start with.
Remember scopes measure and display peak to peak voltage and should be calibrated as such.
I don't know the prices outside Norway, but I'm responsible for a setup of appx 5-800.000 $ worth of instruments, all under ISO 9001 certification. The yearly checkup routine costs around 20.000$, - but then we do have some fairly expensive RF instruments.
The cal. checkup for an average scope is around 600$ a piece, - any repairs or adjustments will add up extra. All these checks are run by an accredited cal.lab. in our premises.
Thus if your old scope really needs a recal, I guess you are easily up for 1000$++ bill........
If you know someone who have access to calibrated instruments, I guess the easiest way is to pay someone for a quick checkup. Certified check and calibration is REALLY expensive.........
A good cleanup with a damp rag and some switch cleaning is usually almost free.....😉
The cal. checkup for an average scope is around 600$ a piece, - any repairs or adjustments will add up extra. All these checks are run by an accredited cal.lab. in our premises.
Thus if your old scope really needs a recal, I guess you are easily up for 1000$++ bill........
If you know someone who have access to calibrated instruments, I guess the easiest way is to pay someone for a quick checkup. Certified check and calibration is REALLY expensive.........
A good cleanup with a damp rag and some switch cleaning is usually almost free.....😉
Last edited:
A quick check of the input amps can be done using DC, and a regular multimeter. It will at least tell you if it's way off. Comparing freq. against a counter is another quick check. Waveforms are more tricky, as you really need a good generator, and most cheap ones are not, IME......
BTW - last year we decided to remove all the cheaper hand multimeters from the cal list....what's the use of paying 250$ to check a 200$ meter...😕
instead we set up an internal check service against certified instruments...😛
BTW - last year we decided to remove all the cheaper hand multimeters from the cal list....what's the use of paying 250$ to check a 200$ meter...😕
instead we set up an internal check service against certified instruments...😛
I remember having an "argument" with the company I worked for... we were the "consumer electronics" side of a major power company, and they saw fit to insist our DVM's were calibrated, and that the "calibration" was to be performed by themselves .
I asked what standards they were using... and was amazed to hear something along the lines of... well it has to read 240volts to within 10%... absolutely unbelievable.
I asked what standards they were using... and was amazed to hear something along the lines of... well it has to read 240volts to within 10%... absolutely unbelievable.
240V +/- 10% used to be the spec for mains power , not an instrument! 🙄
Someone clearly had misunderstood grossly....😉
Someone clearly had misunderstood grossly....😉
It got worse... they then "insisted" that our workshop was "unsafe" as it was earth free (all on isolation transformers)... earth leads snipped off in plug tops etc.
A little demo with earthed scope and leads and soldering iron and live TV chassis soon showed them the error of their ways... they left us alone after that 🙂
A little demo with earthed scope and leads and soldering iron and live TV chassis soon showed them the error of their ways... they left us alone after that 🙂
- Status
- Not open for further replies.