Calibrating an OLD oscilloscope

[windozeuser]

Hey, I just picked up an old teleqipment S51B Oscilloscope for free. I was just wondering how I can make sure it's calibrated accurately. It works and powers on. I've feed it a simple square wave and it seems to be working ok.

 

[ljcox]

There is a variety of simple tests you can do.

Set it to DC and connect a say a 1.5 Volt battery to the input. The line on the screen should go up or down by 1.5V depending on which direction the battery is connected. Repeat with other batteries - ie. 3V, 4.5V, 6V, etc.

Connect say 6V AC from a transformer to the input.

6V is 16.97 V peak to peak (ie. 6 * 2 sqrt 2). Your AC is 60 Hz, so you can check the sweep rate calibration by measuring the period of the signal. 60 Hz has a period of 1000/60 = 16.67 millisec.

If you have an oscillator and frequiency meter, you can check the sweep calibration at other rates. eg. 1000 Hz has a period of 1 ms.

Hope this helps, Len

 

[zachtheterrible]

You will have to compensate your probe as well. The scope should have a little pin or something on it that you can connect the probe to, that pin outputs a square wave.

There should be a little slot to adjust the compensation on the probe with a screwdriver. Turn it until the square waves are just that: square. If its over-compensated, you will get very sharp peaks, undercompensated just hte opposite.

Set it to DC and connect a say a 1.5 Volt battery to the input. The line on the screen should go up or down by 1.5V depending on which direction the battery is connected. Repeat with other batteries - ie. 3V, 4.5V, 6V, etc.

test the battery with a voltage meter, its unlikely that the battery is going to be exactly 1.5Volt.

 

[ljcox]

Set it to DC and connect a say a 1.5 Volt battery to the input. The line on the screen should go up or down by 1.5V depending on which direction the battery is connected. Repeat with other batteries - ie. 3V, 4.5V, 6V, etc.

test the battery with a voltage meter, its unlikely that the battery is going to be exactly 1.5Volt.

Correct, this also applies to the AC voltage. A 6 Volt transformer will not necessarily give 6 V, particularily with no load.

Len

 

[Nigel Goodwin]

Hey, I just picked up an old teleqipment S51B Oscilloscope for free. I was just wondering how I can make sure it's calibrated accurately. It works and powers on. I've feed it a simple square wave and it seems to be working ok.

The problem with calibration is finding something to calibrate it against!.

Also, unless you can get the workshop manual, I wouldn't bother touching any adjustments, from what I remember they are quite obscure in those Telequipment scopes? - with adjustments affecting each other.

For 99% of use there's no need for it to be accurately calibrated anyway!.

 

[windozeuser]

Thanks for your tips.

Also, unless you can get the workshop manual, I wouldn't bother touching any adjustments, from what I remember they are quite obscure in those Telequipment scopes? - with adjustments affecting each other.

I've found the Oscilloscope in the Dumpster! From what I see it works perfectly. I don't have the lab manual unforunately.

For 99% of use there's no need for it to be accurately calibrated anyway!.

Can you elaborate on this a little more?

thanks

 

[Nigel Goodwin]

For 99% of use there's no need for it to be accurately calibrated anyway!.

Can you elaborate on this a little more?

It's usually only required for specific institutions and large businesses, it's NOT about the scope being accurately calibrated, it's about having a certificate saying it is! - even on a brand new scope you have to pay for calibration if you require it!.

So your scope might be more than accurate enough for your needs anyway, considering the actual visible reading accuracy from the screen, if it's within 5% or 0.5% you wouldn't be able to tell the difference.

I've been using scopes professionally for 34 years, we've NEVER had a scope calibrated, nor had any desire or need to.

Everything in live has tolerences, and things you do should bear that in mind.

As already suggested, connect it to a battery and measure it's voltage, compare it to your multi-meter, and see if it's fairly similar, it should be, and that's accurate enough for a scope.

Bear in mind, how do you know your meter is accurate?, do you have it professionally calibrated and certified every year? (at great expense!). Digital meters have made people rather complacent, your meter might read 1.765V - but it's only going to be 1% or 2% accurate (even when new!), so you can throw the last digit away, and be very dubious about the next one as well!.

 

[JimB]

Bear in mind, how do you know your meter is accurate?, do you have it professionally calibrated and certified every year? (at great expense!). Digital meters have made people rather complacent, your meter might read 1.765V - but it's only going to be 1% or 2% accurate (even when new!), so you can throw the last digit away, and be very dubious about the next one as well!.

That all depends on how much the meter cost in the first place, and el-cheapo will not be very accurate, something like a Fluke will have much better accuracy, but you pay for it!.

JimB

 

[Nigel Goodwin]

That all depends on how much the meter cost in the first place, and el-cheapo will not be very accurate, something like a Fluke will have much better accuracy, but you pay for it!.

A quick check in RS Components shows Flukes varying from +/-1.5% (3%) to slightly better than +/-0.1% for the really expensive ones.

 

[windozeuser]

Nigel, The X-Axis line is Slightly Rotated at an angle, is there a setting to get it level and center

I tore the cover off, and on the bottom board theres six Variable resistors?

They are labeled RV89 - Link

RV92 - BLKG
RV88 - IPS
RV6 - SET Y GAIN
RV72 - TIME/CM
RV48 - Trig SENS

 

[jbeng]

None of those sound to me like they will help your trace problem. Newer scopes often have an adjustment labelled "trace rotate", some even have it on the front panel. That's the one I would be looking for.
JB

 

[windozeuser]

Don't see anything inside or out like that

 

[ljcox]

Perhaps you could loosen the screws holding the CRT and rotate it by hand. Be sure to turn the power off and discharge the EHT first.

Len

 

[zachtheterrible]

None of those sound to me like they will help your trace problem. Newer scopes often have an adjustment labelled "trace rotate", some even have it on the front panel. That's the one I would be looking for.
JB

yeah, there is a little hole drilled in the front of my scope and i can adjust trace rotation through it with a small screwdriver.

 

[Nigel Goodwin]

Nigel, The X-Axis line is Slightly Rotated at an angle, is there a setting to get it level and center

I've got an old D61b here at work, and it has two 'tabs' sticking out the back of the scope which can rotate the CRT.

 

[mechie]

oscilloscope manual

I've got a service manual for the S51B - no mention of trace rotate - it must be simply a case of rotating the tube - careful, there's 3kV in there!

pm me an email address and I can send you a .pdf scan - 3.5Mb
ps. windozeuser, I tried following you WWW link and my anti-virus went Berserk ! - rife with trojans ! (and a million advertising pages to close down!)

 

[Dean Huster]

Tektronix at one time bought out Telequipment and it was a wholly-owned subsidiary of Tek. I don't know if they still keep up on the microfiche distribution of their manuals, but the Telequipment line, if the paper manuals were no longer available, were available on microfiche.

https://www.tek.com

Only the top-end TQ scopes had a trace rotation adjustment. All others had to have the CRT loosened in the back and rotated. This can be a bit testy, as a lot of times, the CRT socket glued itself into the clamp and made rotation difficult without disassembling the back end of the thing and breaking it loose. Yes, you do have to be careful of the high voltage, but it's pretty difficult to rotate the CRT and know where you are without the scope running and the calibration procedure will have you do it that way. Doing it with the power off is a very hit-and-miss operation and will take you 200 attempts to get it right -- not only that, but the very act of retightening the clamp usually moves the rotation just a little bit.

RV92 - BLKG This will be the blanking adjustment and without the calibation procedure, you won't be able to set it correctly. If you have a good trace and can't see the retrace, I wouldn't worry about it.

RV88 - IPS I don't recognize or remember this one.

RV6 - SET Y GAIN If you have a KNOWN ACCURATE vertical voltage input, this trimpot will adjust the vertical gain of the scope so that it reads correctly. If you don't know how accurate your source is, then don't mess with it. But a DMM measuring a sine wave off the secondary of a transformer and then calculating the peak-to-peak value (two times RMS times the square root of 2) will give you a short-term accurate voltage, although it may be a hard value to read on a scope display.

RV72 - TIME/CM If you have a know accurate frequency (10KHz is about the best), then adjust this trimpot at a sweep speed of 1ms/DIV so that you have one complete square wave for each division. A 1KHz frequency will give you one squarewave over the entire 10 divisions, but that isn't as telling a display as it doesn't take into account the non-linearity that all scopes have at the extremes of the display.

RV48 - Trig SENS This adjusts the triggering circuit for the best sensitivity. You need the manual to get this one right, too. If a trigger is adjusted so that it isn't very sensitive, then it'll be difficult to trigger the scope on small signals. On the other hand, if it's too sensitive, then it'll trigger on anything that comes around so that even the smallest amount of noise on an audio signal will keep the scope from triggering solidly.


You CAN tell the difference if a scope is off by 5% vs. 0.5%. A 5% error with a six-division signal (whether time or amplitude) will give you an error of 1.5 minor divisions which can be disturbing if you have a known-accurate signal. A 0.5% error would give you a 0.15 minor division error which is about a trace-width on a TQ scope. Bear in mind that the low-end TQ scopes are rated for an accuracy of no better than 3% and may be as bad as 5%. But those are specifications. A typical Tektronix scope with a spec of 3% was usually better than 1% after sitting around for 5 years with no calibration and 0.5% right after a fresh calibration.

Dean

 

[windozeuser]

THANK YOU VERY MUCH FOR ALL YOUR ADVICE AND REPLIES

Dean, your post will help me when I begin to test the oscilloscope

Thanks nigel for your wisdom

Thank you everyone else for your tips.

If theres one think I've learned, theres usually a easy solution to a complex problem. Last night I decided to rotate the entire crt yoke approximately 10 degrees to the left. I found a hose clamp at the neck of the yoke, Perfectly striaght now! Thanks EVERYONE!
8)

THANK YOU SO MUCH mechie

I'd have to say that telequipment is one solid peice of equipment, built to last, with the test of time now tekronix

 

[Dean Huster]

Looking back over my post, I might make a comment and a correction.


RV88 - IPS I don't recognize or remember this one. Actually, with the "PS" in there, this may be a power supply adjustment. Definitely don't mess with a power supply adjustment if you don't have the manual, because the power supply will affect every other adjustment in the scope.

And in the next one, after the bold-faced adjustment label, any bold-faced material is a correction to what I originally posted. I just wasn't thinking right:

RV72 - TIME/CM If you have a known accurate frequency (1KHz is about the best), then adjust this trimpot at a sweep speed of 1ms/DIV so that you have one complete square wave for each division. A 100Hz frequency will give you one squarewave over the entire 10 divisions at 1ms/DIV, but that isn't as telling a display as it doesn't take into account the non-linearity that all scopes (except LCD panel digital scopes) have at the extremes of the display.

Sorry for the goof-up. I'm just glad I caught it.

Dean

 

[Helloween]

Hi,

I just googled and came up with this site. I have a similar problem.

I am now the proud owner of an Telequipment S51B. My plan is basically to use it until I don't have any other options. My first plan is just to see if I can figure out what the outputs of my car's ECU do, e.g. injector signal, vss, etc. Hopefully the power consumption is less than 300 Watts (since I plan on using it with an inverter in the car...

**broken link removed**

Anyway, it doesn't have a working probe. And the connectors are not the usual ones for probes. Here is the manual. From what I can tell, the bottom two connectors are for ground and signal.

Any idea how I can get this thing working with a probe?

Thanks in advance.