Purchasing first oscilloscope for serious research/commercial electronics work

[3rj4]

Say you measure result at 10ns, actually is 15ns ? 

It seems very unlikely.

Should I be worried? Or is he just exaggerating?

They are winding you up. Accuracy outside calibration from the manufacturer is as good as you need it to be, if you want to be sure any specific measurement is accurate you start by checking a known accurate source with the scope.

Thanks . I was starting to get worried .

[Fungus]

Yeah, still possible though, my point is, with this serious gear, one must have confidence in the 1st place right ? Especially its the only scope around.

I though the whole point of buying "serious gear" was that you always have that confidence?

[BravoV]

Yeah, still possible though, my point is, with this serious gear, one must have confidence in the 1st place right ? Especially its the only scope around.

I though the whole point of buying "serious gear" was that you always have that confidence?

Only new one and covered by warranty.

[tggzzz]

Say you measure result at 10ns, actually is 15ns ? 

It seems very unlikely.

Should I be worried? Or is he just exaggerating?

They are winding you up. Accuracy outside calibration from the manufacturer is as good as you need it to be, if you want to be sure any specific measurement is accurate you start by checking a known accurate source with the scope.

Thanks . I was starting to get worried .

The point at which you have to be worried depends on the purposes for which a measurement will be used - and we don't know that.

To take a different example... If you have several voltmeters, they will always read slightly different values - so which is correct?

Usually the absolute value is not very important; stability is more important. It is cheap and easy to get them to all read the same value: just have one stable voltage source, and twiddle each meter until they read the same. 

It becomes expensive and difficult when you want your meters to read the same value as everybody else's voltmeters. At that point you have to twiddle your meter to the same value as other people's meters - and that means the voltage source has to be an external common standard, which is expensive.

[Shock]

Let me explain it...

Calibration is a check to see if the device is operating within it's defined parameters.

When scopes are new they often have a manufacturers calibration and or a certificate (which may include a report or the results). They may also specify the frequency or time at which the calibration or certification expires and is no longer considered accurate or valid.

After several calibrations are performed this builds up confidence that the equipment is operating normally (as you can see drift from the specifications over time). Some scopes have a self calibration, they may check themselves or even adjust themselves to remain as accurate as possible. This does not replace a full calibration check from a manufacturer or third party.

If the equipment fails calibration, to restore it back within it's specified accuracy, it either needs adjusting to be within calibration or repairing. Otherwise it will be only as accurate as it's last check or possibly worse. Calibration equipment typically is collectively more expensive than the equipment being calibrated as it is also been checked periodically for calibration and has a higher accuracy and tighter specification.

Traceability is another aspect of calibration/certification it and it's basically to ensure noone fudges the figures. The flow goes something like this. Standard ==> Transferred Standard ==> Calibration Device ==> Equipment being calibrated.

[3rj4]

Say you measure result at 10ns, actually is 15ns ? 

It seems very unlikely.

Should I be worried? Or is he just exaggerating?

They are winding you up. Accuracy outside calibration from the manufacturer is as good as you need it to be, if you want to be sure any specific measurement is accurate you start by checking a known accurate source with the scope.

Thanks . I was starting to get worried .

The point at which you have to be worried depends on the purposes for which a measurement will be used - and we don't know that.

To take a different example... If you have several voltmeters, they will always read slightly different values - so which is correct?

Usually the absolute value is not very important; stability is more important. It is cheap and easy to get them to all read the same value: just have one stable voltage source, and twiddle each meter until they read the same. 

It becomes expensive and difficult when you want your meters to read the same value as everybody else's voltmeters. At that point you have to twiddle your meter to the same value as other people's meters - and that means the voltage source has to be an external common standard, which is expensive.

Let me explain it...

Calibration is a check to see if the device is operating within it's defined parameters.

When scopes are new they often have a manufacturers calibration and or a certificate (which may include a report or the results). They may also specify the frequency or time at which the calibration or certification expires and is no longer considered accurate or valid.

After several calibrations are performed this builds up confidence that the equipment is operating normally (as you can see drift from the specifications over time). Some scopes have a self calibration, they may check themselves or even adjust themselves to remain as accurate as possible. This does not replace a full calibration check from a manufacturer or third party.

If the equipment fails calibration, to restore it back within it's specified accuracy, it either needs adjusting to be within calibration or repairing. Otherwise it will be only as accurate as it's last check or possibly worse. Calibration equipment typically is collectively more expensive than the equipment being calibrated as it is also been checked periodically for calibration and has a higher accuracy and tighter specification.

Traceability is another aspect of calibration/certification it and it's basically to ensure noone fudges the figures. The flow goes something like this. Standard ==> Transferred Standard ==> Calibration Device ==> Equipment being calibrated.

Interesting. Thank you all for the information.

This scope was last calibrated in 2004: https://i.ebayimg.com/images/g/vxIAAOSwGzBd22-u/s-l1600.jpg
So I probably should go and get it calibrated.

[Shock]

Do you need secondhand equipment to come calibrated? Yes if you want to ensure it's accurate. No if it's still in calibration or was recently or you don't care. Ensuring it at least passes internal calibration and self tests is a good start though.

If the seller is unscrupulous they may sell/ditch the equipment because it cannot be easily recalibrated. So there is some advantage in getting a calibration check done sooner than later. Obviously if you need a calibrated instrument, you need to factor calibration costs ahead of time.

If you have access to similar or better calibrated equipment and you have or can make a suitable circuit to test, then you can make a comparative calibration check. This is generally the poor mans calibration route but at least gives you a rough idea.

[Fungus]

Oscilloscopes aren't multimeters. Oscilloscopes have self-calibration against built-in references.

The chances of the internal references going out of spec is very very very small and usually detected by the diagnostics.

ie. These:


Unless you see something obviously wrong when it arrives then I'd give it the benefit of the doubt.

[tautech]

Unless you see something obviously wrong when it arrives then I'd give it the benefit of the doubt.

When the OP has little idea of how to even operate a scope let alone an advanced one. 

His safest course of action is to have its performance verified as meeting datasheet spec.

[Mechatrommer]

Thanks . I was starting to get worried .

tell you what. there is possibility when its arrived, its showing half the magnitude, or its PSU smoked, or its clock non locked, or its simply DOA you ask people here, they will tell you the worst possible nightmare. if its me, i will just enjoy the purchase, if not, return it. or spend half the money i can afford, if its not expected, buy another one with more cautious. anyway, did you win the AUD3000+ DSO?
https://www.ebay.com.au/itm/HP-Agilent-54845A-4-Channel-1-5-GHz-8-GSa-s-Infiniium-Oscilloscope-0445/254433035190?hash=item3b3d6403b6:g:DYoAAOSwzCZd22-W#rpdCntId

[3rj4]

Thanks . I was starting to get worried .

tell you what. there is possibility when its arrived, its showing half the magnitude, or its PSU smoked, or its clock non locked, or its simply DOA you ask people here, they will tell you the worst possible nightmare. if its me, i will just enjoy the purchase, if not, return it. or spend half the money i can afford, if its not expected, buy another one with more cautious. anyway, did you win the AUD3000+ DSO?
https://www.ebay.com.au/itm/HP-Agilent-54845A-4-Channel-1-5-GHz-8-GSa-s-Infiniium-Oscilloscope-0445/254433035190?hash=item3b3d6403b6:g:DYoAAOSwzCZd22-W#rpdCntId

That's the one, but it's only 2400 AUD (including shipping) - not 3000+ AUD.

[3rj4]

Unless you see something obviously wrong when it arrives then I'd give it the benefit of the doubt.

When the OP has little idea of how to even operate a scope let alone an advanced one. 

His safest course of action is to have its performance verified as meeting datasheet spec.

Since you seem to work in this space, and you're based in NZ, what is your estimation of how much this will cost me (I'm based in AUS)?

[Fungus]

Since you seem to work in this space, and you're based in NZ, what is your estimation of how much this will cost me (I'm based in AUS)?

If you have to ask, you can't afford it.

[BillB]

Congratulations on winning your auction for that used superbike really nice oscilloscope!  Given that you have no specific job yet for which you'll need this scope, and you have some sensitivity to budget, why worry about calibration yet? 

As others have suggested, examine the scope for obvious problems, but otherwise if it passes self-tests and sensibly performs whatever measurements you can figure out at this point, just spend some time with it learning about oscilloscopes before you go sinking even more money into it.

As you explore the scope and its capabilities, you can use this forum as a resource for troubleshooting if things don't behave like they should.

[3rj4]

Since you seem to work in this space, and you're based in NZ, what is your estimation of how much this will cost me (I'm based in AUS)?

If you have to ask, you can't afford it.

I've seen catalogues price it at $180 USD? That's 10% of the cost of the scope itself.

[3rj4]

Congratulations on winning your auction for that used superbike really nice oscilloscope!  Given that you have no specific job yet for which you'll need this scope, and you have some sensitivity to budget, why worry about calibration yet? 

As others have suggested, examine the scope for obvious problems, but otherwise if it passes self-tests and sensibly performs whatever measurements you can figure out at this point, just spend some time with it learning about oscilloscopes before you go sinking even more money into it.

As you explore the scope and its capabilities, you can use this forum as a resource for troubleshooting if things don't behave like they should.

Sounds good. Gaining a deep understanding of how the scope works will surely have a lot of crossover with other aspects of electronics, so I think it will be a very productive endeavour!

[JohnnyBerg]

When the OP has little idea of how to even operate a scope let alone an advanced one. 

These kind of threats pop up from time to time, with a lot of response all giving the "best advice". In the early days we had a name for that .. 

[Mechatrommer]

I've seen catalogues price it at $180 USD? That's 10% of the cost of the scope itself.

that probably only for renting a space in their calibration lab, Keysight et al are known for their "fabulous" after sale service charge. you are talking 10% of "used" price. its new price is like 8K i guess, so expect something like AUD800. but thats just a wild guess. i recently bought an used spectrum analyzer, its calibration cost is 150% from what i paid for the unit. now that should start to worry you

[tggzzz]

Since you seem to work in this space, and you're based in NZ, what is your estimation of how much this will cost me (I'm based in AUS)?

If you have to ask, you can't afford it.

I've seen catalogues price it at $180 USD? That's 10% of the cost of the scope itself.

You need to understand the difference between "calibration" and "adjustment". Roughly speaking calibration means comparing to specification and noting where in/out of spec, whereas adjustment means twiddling it until it is in spec.

[3rj4]

I've seen catalogues price it at $180 USD? That's 10% of the cost of the scope itself.

that probably only for renting a space in their calibration lab, Keysight et al are known for their "fabulous" after sale service charge. you are talking 10% of "used" price. its new price is like 8K i guess, so expect something like AUD800. but thats just a wild guess. i recently bought an used spectrum analyzer, its calibration cost is 150% from what i paid for the unit. now that should start to worry you

Since you seem to work in this space, and you're based in NZ, what is your estimation of how much this will cost me (I'm based in AUS)?

If you have to ask, you can't afford it.

I've seen catalogues price it at $180 USD? That's 10% of the cost of the scope itself.

You need to understand the difference between "calibration" and "adjustment". Roughly speaking calibration means comparing to specification and noting where in/out of spec, whereas adjustment means twiddling it until it is in spec.

Thanks for the clarification. If it's really that expensive, then I'll just have to make do without it. Anyway, as the guys above said, as long as it passes the self-tests and has no obvious problems, I'm guessing it will be fine for my work.

I'm curious to learn about how such a device is maintained to function well. These are expensive devices in the first place, so I'm guessing that you guys don't go around paying for calibration in addition to that, so what do you do to calibrate it when you notice that there's problem? Given the amount of time people put into personally repairing their expensive scopes, surely there's something you guys do for calibration too?

I'm keen to start studying some oscilloscope-specific material (in addition to the other electronics material I'm studying), so if you guys have any textbook suggestions, I'd appreciate it. Something less technical to start off would probably be best, but then I'd like to jump into some technical material that will really give me a deep understanding of the device, so those recommendations are also appreciated.

[tggzzz]

These are expensive devices in the first place, so I'm guessing that you guys don't go around paying for calibration in addition to that, so what do you do to calibrate it when you notice that there's problem? Given the amount of time people put into personally repairing their expensive scopes, surely there's something you guys do for calibration too?

First you decide what you need to achieve.

Second you understand the consequences of different courses of action.

Third you decide which course of action is appropriate.

No, it isn't easy. Yes, it does take time and experience.

[Shock]

After double checking all the self test pass yourself, and then perform functional tests you can do the self calibration. It will explain all this in the manual with the procedure to follow and what you need, if you get stuck I would start a new thread.

As explained it's not a definitive calibration but it might be able to detect if any problems are found. You might also be able to look at logs and stats to determine the history of the oscilloscope. Just depends on the oscilloscope they vary model to model.

Some people on the forum acquire calibration equipment so they can perform their own but it can get expensive especially the more advanced the oscilloscope is. If you have a service manual for your oscilloscope it may detail the calibration adjustment procedure. Sometimes the service manual/manual are one in the same, sometimes the manufacturer does not make the service manual public, sometimes it's just the schematic not released and often you can find it all on the web.

Eevblog has a US calibration club, I'm not sure whats in the latest one or even if it's suitable for oscilloscopes. Besides that you aren't in the US, you're new and you need more experience. My advice is to completely absorb the manual and service manual and you will learn a lot. If you are unsure of terms look them up on the Keysights website and look for literature around the time the oscilloscope came out and about it's capabilities. If you know where every function is on the oscilloscope and can watch a video of someone using one and understand where that feature is on yours and your scopes limitations, that is half the battle. The rest is practical measurement techniques and interpreting what you are looking at.

[tautech]

When the OP has little idea of how to even operate a scope let alone an advanced one. 

These kind of threats pop up from time to time, with a lot of response all giving the "best advice". In the early days we had a name for that .. 

Err, well the OP clearly stated he had zero scope experience and requires ns accuracy yet still spent $2k+ on a used scope. Could it have previously been abused or damaged from an instant of carelessness we may never know.

Formal performance verification (advanced checks) by one of the many Cal labs will identify potential faults and then give the owner options to re-Cal (adjust) or have it repaired.

Without a good understanding of the risks of buying something used and the potential costs for correction of a faulty instrument the poorly informed buyer can be setting themself up for unforeseen additional costs.

[Mechatrommer]

as a beginner, we just measure whatever we want to measure. if there is signal, we are happy enough. forget about anything that cost us $$$ that we have less appreciation about. he will know when he needs a full scale calibration. by that time, he will and should be willing to find money for that. if not enough money, we will find workaround, improvise and use every bits of our imagination to deal with it. at least thats how myself is..

[nctnico]

Without a good understanding of the risks of buying something used and the potential costs for correction of a faulty instrument the poorly informed buyer can be setting themself up for unforeseen additional costs.

That is true up to the point where a piece of equipment can do a thourough self test. The Agilent 54835A / 54845A / 54846A can do such a self test. If it passes this self test then you know you have a working scope.