Test Equipment Anonymous (TEA) group therapy thread

[bd139]

Yes that is true. Unfortunately to select a trigger you need to understand what the problem is before the event. Intensity grading did result in the capture of this issue for ref. However I had to capture a large chunk of samples for it to have enough resolution to capture the event more than once to establish delta t and be observable at the same time. We’re talking 1 in 800 pulses was 20ns short here. Capture done in single sweep mode and then eyeball trigger

[tautech]

Yes that is true. Unfortunately to select a trigger you need to understand what the problem is before the event. Intensity grading did result in the capture of this issue for ref. However I had to capture a large chunk of samples for it to have enough resolution to capture the event more than once to establish delta t and be observable at the same time. We’re talking 1 in 800 pulses was 20ns short here. Capture done in single sweep mode and then eyeball trigger

Yeah fine until you need to correlate it to a possible cause/source with additional channels running. That's when an advanced trigger is your friend. DSO's shine at this, not so much CRO's.

[bd139]

Yep exactly that. Once you’ve worked out when something is happening and what it’s doing when it’s happening you can set up a suitable trigger for it then look around the circuit for why it is happening with other channels. I love the DSOs for this.

In the above situation I knew what was happening due to the suspiciously around 18ms glitch frequency which usually points to WDT. One clrwdt instruction added and sorted thus proving with a red/green test that it was the root cause. Not all things are that simple. Debugging weird shit is the most fun you can have I find, probably why the day job ended up being software.

Incidentally this capability is when tek realised they needed to give up the analogue scopes i reckon. Even a 2467 “bright eye” is quite frankly useless at picking this sort of crap up. It’ll tell you what is happening but you’ll have to use another tool to work out why.

[tautech]

And yet in this day and age there are still some that insist your first scope must be a CRO,  it was for me and I learnt a ship load but a DSO is a more capable tool.

[mnementh]

Yep exactly that. Once you’ve worked out when something is happening and what it’s doing when it’s happening you can set up a suitable trigger for it then look around the circuit for why it is happening with other channels. I love the DSOs for this.

In the above situation I knew what was happening due to the suspiciously around 18ms glitch frequency which usually points to WDT. One clrwdt instruction added and sorted thus proving with a red/green test that it was the root cause. Not all things are that simple. Debugging weird shit is the most fun you can have I find, probably why the day job ended up being software.

Incidentally this capability is when tek realised they needed to give up the analogue scopes i reckon. Even a 2467 “bright eye” is quite frankly useless at picking this sort of crap up. It’ll tell you what is happening but you’ll have to use another tool to work out why.

Yeah, but if YOU had been operating in SE debugging mode instead of EE diag mode, you'd probably have guessed that and tried it without ever needing to drag out the DSO.

That's something younger minds are often better at... shifting gears/changing operating modes. The problem is they usually want to do it when it ISN'T called for. 


mnem
As a matter of fact, I DO know jack shit! Plug shit TOO!   

[mnementh]

And yet in this day and age there are still some that insist your first scope must be a CRO,  it was for me and I learnt a ship load but a DSO is a more capable tool.

The reason it's a better tool to LEARN on is the usually-linear relationship between what you configure, what you're measuring and the results you see on the screen. Once you understand the fundamentals, which the CRO is better at teaching hands-down, then you're ready to wrap your brain around more advanced diagnostic techniques; some of which a DSO is better at, and some the CRO is still better at.

As I've said before... the "one or the other" argument is just plain stupid. Each excels at different things, and decent CROs are cheap and plentiful enough that there's simply no good reason NOT to have both on your bench.


mnem
glerp?

[Specmaster]

And yet in this day and age there are still some that insist your first scope must be a CRO,  it was for me and I learnt a ship load but a DSO is a more capable tool.

The reason it's a better tool to LEARN on is the usually-linear relationship between what you configure, what you're measuring and the results you see on the screen. Once you understand the fundamentals, which the CRO is better at teaching hands-down, then you're ready to wrap your brain around more advanced diagnostic techniques; some of which a DSO is better at, and some the CRO is still better at.

As I've said before... the "one or the other" argument is just plain stupid. Each excels at different things, and decent CROs are cheap and plentiful enough that there's simply no good reason NOT to have both on your bench.


mnem
glerp?

Precisely my point and with having both, you'd have the ideal tool for each job as and when required and I for one would love to have a DSO sitting alongside my CRO's.

[tautech]

And yet in this day and age there are still some that insist your first scope must be a CRO,  it was for me and I learnt a ship load but a DSO is a more capable tool.

The reason it's a better tool to LEARN on is the usually-linear relationship between what you configure, what you're measuring and the results you see on the screen. Once you understand the fundamentals, which the CRO is better at teaching hands-down, then you're ready to wrap your brain around more advanced diagnostic techniques; some of which a DSO is better at, and some the CRO is still better at.

As I've said before... the "one or the other" argument is just plain stupid. Each excels at different things, and decent CROs are cheap and plentiful enough that there's simply no good reason NOT to have both on your bench.


mnem
glerp?

Is it ?
The knobs and buttons used for the 'fundamentals' are very similar.

But your word 'understanding' is key and while some might get to gain it from twiddling knobs and punching buttons others already have the basics from just reading books (I did) or from the likes of studying on this forum.
I used my first scope @~13 at high school and for only one opportunity for ~20mins and not again until some 25+yrs later, then in just a few years discovered DSO's.

One might argue, why learn two sets of UI's ?
Sure one is fairly basic but so is fundamental usage of a DSO. It just does more. heaps more.

[mnementh]

And yet in this day and age there are still some that insist your first scope must be a CRO,  it was for me and I learnt a ship load but a DSO is a more capable tool.

The reason it's a better tool to LEARN on is the usually-linear relationship between what you configure, what you're measuring and the results you see on the screen. Once you understand the fundamentals, which the CRO is better at teaching hands-down, then you're ready to wrap your brain around more advanced diagnostic techniques; some of which a DSO is better at, and some the CRO is still better at.

As I've said before... the "one or the other" argument is just plain stupid. Each excels at different things, and decent CROs are cheap and plentiful enough that there's simply no good reason NOT to have both on your bench.


mnem
glerp?

Is it ?
The knobs and buttons used for the 'fundamentals' are very similar.

But your word 'understanding' is key and while some might get to gain it from twiddling knobs and punching buttons others already have the basics from just reading books (I did) or from the likes of studying on this forum.
I used my first scope @~13 at high school and for only one opportunity for ~20mins and not again until some 25+yrs later, then in just a few years discovered DSO's.

One might argue, why learn two sets of UI's ?
Sure one is fairly basic but so is fundamental usage of a DSO. It just does more. heaps more.

Because there is no such thing as too much learning.


mnem
*mic drop*

[Mr. Scram]

This discussion got me wondering. Are CROs still common on benches in proper enterprise or industry settings? Not because they neglected to break, but because their owners intentionally chose them over a DSO? Can you still buy CROs in meaningful numbers and with relevant specifications? How many does Keysight sell each year?

[GerryBags]

This discussion got me wondering. Are CROs still common on benches in proper enterprise or industry settings? Not because they neglected to break, but because their owners intentionally chose them over a DSO? Can you still buy CROs in meaningful numbers and with relevant specifications? How many does Keysight sell each year?

You certainly see many analog 'scopes advertised as: "pulled from a working environment..". They don't often say how many decades it's been since it was pulled, but I can easily imagine small companies or start-ups that simply can't afford as many of the top-end, modern machines as they need and, as you all have so eloquently described, the old stuff can still cover many jobs that then the modern gear doesn't have to be used for, freeing it up.

Sorry for the run-on sentences, I couldn't see where to put a 'period'. 

[Brumby]

I can't say from observation of actual workshops - but I can offer a thought from the bean counter perspective...

If a $100 CRO can do the job capably, you are a lot less likely to get a $1,000 from the budget for a DSO.

[mnementh]

This discussion got me wondering. Are CROs still common on benches in proper enterprise or industry settings? Not because they neglected to break, but because their owners intentionally chose them over a DSO? Can you still buy CROs in meaningful numbers and with relevant specifications? How many does Keysight sell each year?

Good question...

A quick search shows that TEquipment and Arrow both still sell Analog scopes; mostly B&K school-grade gear, though Arrow does have a couple TFT-screen analog models by LeCroy.

I can tell you that our local hackspace has both analog & digital scopes, and the EEs who work and produce commercial product from the space both keep a couple analog scopes apiece on carts in their arsenals, outside the motley collection in the public spaces. But both of them are old-timers like myself, so they're probably not a good representation, even though they are technically still professional EEs.

Of course there are cheap DSOs all over the place; they swarm like mosquitoes in a place like that. And at least half of them still work.    We have a couple Adafruit wannabes doing their thing there too. 


Tautech, I skimmed over Defpom's video on your 1202X-E; I like it overall.

While it does appear quite responsive, I don't see a lot of difference in the usability of the UI vs my 1054Z. I really only notice responsiveness issues on mine when I'm working with multiple signals and decoding; for the usual daily driver stuff, it's pretty much right there with me. For folks who need the extra BW, your 1202X-E is clearly a better choice in this price range; but I think part of what makes it more responsive is it only has to contend with 2 channels and 10K less memory depth. For most of what I need a DSO for, 4 channels is a minimum and I sometimes still wind up dragging out the Salae clone and my tablet.

So it's nice to have a choice in the bottom bracket, depending on what your primary needs are.


Cheers,


mnem
*Toddles off to ded*

[tautech]

Tautech, I skimmed over Defpom's video on your 1202X-E; I like it overall.

While it does appear quite responsive, I don't see a lot of difference in the usability of the UI vs my 1054Z. I really only notice responsiveness issues on mine when I'm working with multiple signals and decoding; for the usual daily driver stuff, it's pretty much right there with me. For folks who need the extra BW, your 1202X-E is clearly a better choice in this price range; but I think part of what makes it more responsive is it only has to contend with 2 channels and 10K less memory depth. For most of what I need a DSO for, 4 channels is a minimum and I sometimes still wind up dragging out the Salae and my tablet.

So it's nice to have a choice in the bottom bracket, depending on what your primary needs are.


Cheers,


mnem
*Toddles off to ded*

We thought the 1202X-E was pretty neat when it came out in April last year and as an entry level unit it was a big step forward on what Siglent had produced in this class before. However the new 4ch X-E's blow them out of the water in terms of capability by virtue really only of much more added functionality. Some of this is stuff many might not need but it's there should you do so these new 4ch  X-E's are a DSO that you can really grow into.

But back to the 1202X-E for a mo....Dave got sent one of the very first and it was a 1102X-E and Dave being Dave when he had it apart for the compulsory teardown decided it would be good form to eyeball the JTAG at boot where he logged it all for all to see and guess what, it was pretty simple to see it was of a series that was BW limited in the boot code. 
Siglent took one look at Dave's vid and then only released the top of the range 200 MHz model. 
The vid is #985 if you want to take a look.

So with another 8mths of development they released the 1004X-E models and that's where the 1054Z becomes seriously challenged, two 1GSa/s ADC's with 14M mem for each plus that same fast processor and the additional functionality over the 2ch models. Again, a sharp responsive DSO.

Anyways, enough of the sales pitch, you can hunt out any further info if you're interested.
Nighty night mnem.

[beanflying]

I managed for over 20 years post Uni without a scope of any sort apart from an occasional drop around to 'borrow' a little time on one. About 3 years ago brought a piece of Hantek USB CRUD which was better than nothing but in hind sight I should have sought out a clunky 20Mhz Analogue of some sort which would have been as accurate and not crashed every other minute  .

Now owning a Migsig 100Mhz Toy would I go back to using an Analogue one day to day - unlikely. I did buy a fixer upper Philips 20Mhz as a play thing and to hack together a curve tracer circuit to dedicate to it at some stage but not for everyday use and I cant see me ever putting money at a high end Analogue ever.

As to day to day use mine is more like weekly or even every other week  I blame the lack of more frequent use to not having owned one for a very large part of my post Uni life and I am used to getting by without one still.

Todays Project just for fun putting the smoke back into my Dad's old Growler so I can use it as a degausser (100W anyone want any data erased?  ) Smoke reinstalled into the new switch 40 year old failed masking tape removed and nail polish added to the partially unwound and rewound coil where it arced out. No CRO was needed in this repair 

Beer Time 

[Specmaster]

I managed for over 20 years post Uni without a scope of any sort apart from an occasional drop around to 'borrow' a little time on one. About 3 years ago brought a piece of Hantek USB CRUD which was better than nothing but in hind sight I should have sought out a clunky 20Mhz Analogue of some sort which would have been as accurate and not crashed every other minute  .

Now owning a Migsig 100Mhz Toy would I go back to using an Analogue one day to day - unlikely. I did buy a fixer upper Philips 20Mhz as a play thing and to hack together a curve tracer circuit to dedicate to it at some stage but not for everyday use and I cant see me ever putting money at a high end Analogue ever.

As to day to day use mine is more like weekly or even every other week  I blame the lack of more frequent use to not having owned one for a very large part of my post Uni life and I am used to getting by without one still.

Todays Project just for fun putting the smoke back into my Dad's old Growler so I can use it as a degausser (100W anyone want any data erased?  ) Smoke reinstalled into the new switch 40 year old failed masking tape removed and nail polish added to the partially unwound and rewound coil where it arced out. No CRO was needed in this repair 

Beer Time 

Haha, neither was a DSO [emoji41]

From mobile device so predictive text might have struck again [emoji83]

[Specmaster]

.

This discussion got me wondering. Are CROs still common on benches in proper enterprise or industry settings? Not because they neglected to break, but because their owners intentionally chose them over a DSO? Can you still buy CROs in meaningful numbers and with relevant specifications? How many does Keysight sell each year?

Good question...

A quick search shows that TEquipment and Arrow both still sell Analog scopes; mostly B&K school-grade gear, though Arrow does have a couple TFT-screen analog models by LeCroy.

I can tell you that our local hackspace has both analog & digital scopes, and the EEs who work and produce commercial product from the space both keep a couple analog scopes apiece on carts in their arsenals, outside the motley collection in the public spaces. But both of them are old-timers like myself, so they're probably not a good representation, even though they are technically still professional EEs.

Of course there are cheap DSOs all over the place; they swarm like mosquitoes in a place like that. And at least half of them still work.  [emoji14]  We have a couple Adafruit wannabes doing their thing there too. 


Tautech, I skimmed over Defpom's video on your 1202X-E; I like it overall.

While it does appear quite responsive, I don't see a lot of difference in the usability of the UI vs my 1054Z. I really only notice responsiveness issues on mine when I'm working with multiple signals and decoding; for the usual daily driver stuff, it's pretty much right there with me. For folks who need the extra BW, your 1202X-E is clearly a better choice in this price range; but I think part of what makes it more responsive is it only has to contend with 2 channels and 10K less memory depth. For most of what I need a DSO for, 4 channels is a minimum and I sometimes still wind up dragging out the Salae clone and my tablet.

So it's nice to have a choice in the bottom bracket, depending on what your primary needs are.


Cheers,


mnem
*Toddles off to ded*

Rapid Electronics in the UK are still selling new CROs, 20MHz 2 ch and 100MHz 2 ch ones, I have one of their 25MHz 2 ch ones myself.

I'm not sure if anyone else is selling any, perhaps I might do a scan on the Web later

We have a lab auction going on of a research company and the only scopes they used are high end CROs so it would seem that they are still in use commercially.

From mobile device so predictive text might have struck again [emoji83]

[tggzzz]

And yet in this day and age there are still some that insist your first scope must be a CRO,  it was for me and I learnt a ship load but a DSO is a more capable tool.

The reason it's a better tool to LEARN on is the usually-linear relationship between what you configure, what you're measuring and the results you see on the screen. Once you understand the fundamentals, which the CRO is better at teaching hands-down, then you're ready to wrap your brain around more advanced diagnostic techniques; some of which a DSO is better at, and some the CRO is still better at.

As I've said before... the "one or the other" argument is just plain stupid. Each excels at different things, and decent CROs are cheap and plentiful enough that there's simply no good reason NOT to have both on your bench.

Precisely.

And in response to tautech's point, "And yet in this day and age there are still some that insist you cannot do anything useful with a CRO, and that 100MHz is sufficient with modern logic".

[bd139]

Indeed. 100MHz is certainly pushing it with modern logic. It’s only good for me because at best I’m using a 16Mhz or so clock and the result of that is much slower. If you have an Intel Core on the bench, out comes your £100k scope

A CRO is perfectly useful. Just a DSO is more useful over a larger problem domain. A CRO is a good learning exercise. I recently sat with an EE student in front of a Rigol DS2000 series unit and it was confusing for him. YMMV but I’d like to see some DSOs on the market with an interface as good as the old 54600’s.

[med6753]

All this talk about CRO's being obsolete has got my boys in full revolt.  They want to kill the Siglent DSO which is currently in hiding.  Please assure them that they still have a function on the bench. I tried and they won't listen.  I'm afraid if this continues much longer they'll start releasing magic smoke. 

Please help!    

[Berni]

2.5GHz thingie, not my fault, blame Berni. 

Nice one. No need to thank me. This is what the TEA group is for.

But im sure you will enjoy the probe. I used my infiniium versions of them a lot and they run circles around any passive probe for high speed digital use.

I'm quite surprised that the probe's body is heavy, if connected directly and hanging horizontally, perpendicular with the equipment front panel, worry that the stress caused by it's weight will wreck the N terminal at the spectrum analyzer or at the scope's BNC terminal using the included N to BNC adapter.

Thinking of using a L shape adapter, to make the probe body parallel with the equipment front panel, and I guess this should help to relieve the stress by resting part of the body weight on the bench surface, rather than hanging horizontally, how do you use yours ? Or I'm worry too much ?

Im sure its fine for N connectors as long as its not in a spot where you could walk by and snag it.

But BNC yeah that's way too big to hang off one of those. Probably best to use a short  BNC to N cable to connect that. You don't really need to worry about adding cable since the signal is driven inside the probe, so the signal is already having to travel all the way down the probe cable. The box on the N connector is for generating the various power supply rails, the signal passes straight trough it.

[tggzzz]

Indeed. 100MHz is certainly pushing it with modern logic. It’s only good for me because at best I’m using a 16Mhz or so clock and the result of that is much slower. If you have an Intel Core on the bench, out comes your £100k scope

Clock rate is irrelevant; all that matters is the transition time. But you know that.

Now if you are working in the digital domain, clock rate is relevant - but you ought to be using a digital domain tool like a logic analyser (optionally plus protocol decoder) of some sort.

A CRO is perfectly useful. Just a DSO is more useful over a larger problem domain. A CRO is a good learning exercise. I recently sat with an EE student in front of a Rigol DS2000 series unit and it was confusing for him. YMMV but I’d like to see some DSOs on the market with an interface as good as the old 54600’s.

I've seen some very misleading results on an HP54621; and removing the misleading result has to be done every time any control was changed. I ought to capture pictures sometime.

[bd139]

That's a fair point. Indeed you are right. The DS1054Z's protocol decode is reasonably efficient for analysing simple serial bus type problems which is what I'm using it for mainly (fecking SPI). It isn't as good as a Salea though but it has more tools in one box that neatly sits on my power supplies. Everything is a compromise on price, performance and quality and this sort of hits the middle of the triangle.

Please note I only said the interface of the 54600. The actual implementation is a different matter. Things have changed a lot since 1994. As with all gear you're looking at the abstraction of a problem, not the problem itself so you need to understand what happens between one and the other.

Perhaps that is a good place for the analogue scope; when you need to intimately understand the difference between input and eyeballs without delving into sampling theorem. You can model an analogue scope as a low pass filter.

[Specmaster]

Quote from: med6753 on Today at 10:43:08 am

All this talk about CRO's being obsolete has got my boys in full revolt.  They want to kill the Siglent DSO which is currently in hiding.  Please assure them that they still have a function on the bench. I tried and they won't listen.  I'm afraid if this continues much longer they'll start releasing magic smoke. 

Please help!   

My your boys sure look angry, and they full set of teeth are in full display for all to see and anyone can see that they are raring for a fight    but please reassure them that is a storm in a tea cup. Is a analogue watch any less useful then a digital one, is a mechanical one any less capable than a quartz one . Do Rolex make and Quartz, digital watches, I don't think so but they are still held up as possibly the best watch maker of all, Breitling, Omega are other top notch makers all of which appear to be mechanical self winding mechanism. Good luck in trying to convince their customers that they have had their day and only Quartz watches cut the mustard these days.

DSO's are in my opinion a very handy and useful tools but I can't help but think that they are gaining such popularity purely through the use of software because we are not equipping our youngsters with engineering skills any longer, schools are focusing purely the world becoming an electronic one. A world where machines do the manufacturing and people sit at desks behind a screen and devise ways of making a piece of silicon do more and more and as a result there is less actual hands on engineering going on these days 

[bd139]

If we're on the subject of the watch analogy, there are two outcomes for a timekeeping piece:

1. The timekeeping piece of keeping time. The pinnacle of the standalone timekeeping device is the Casio F-91W, available for a mere £8 on Amazon. It's waterproof, virtually indestructable, monumentally accurate (my last one drifted 11 seconds a month with no time sync), cheap as chips, disposable and no one will stab you for it. However I haven't worn one for at least a decade, because I have a computer in my pocket which is currently reporting it is in time sync to within 26ms of reference time sources. That computer also allows me to collaborate with other people carrying computers on time related things. And it happens to play music and allow me to pay for things. It's amazing, and was considerably cheaper than a "good analogue watch".

2. The timekeeping piece because you like it or it brings status. If you like it, fair enough! no problems. For ref I had a 1973 Omega Speedmaster from my father. Very nice piece of kit. Really liked it. But it went because I didn't need it, didn't use it, would be stabbed for it in two minutes flat etc. It was also purely a status purchase by him as a flasher of the Omega and Jag keyring. Status for the sake of status is something I dispise. I know currency and derivatives traders who blow silly money on a Rolex entirely for status. Does it help them tell the time? No, they work microsecond timing on their computers! Stick the money in a food bank.

I diagress but analogising, which is the superior for productivity and ROI/cost benefit? That's what matters to an engineering process and organisation and the education system that ultimately drives people into it.

Part 2: On the subject of engineering. Getting everything into the digital domain is important these days. It's easier to change things there and it doesn't have any nasty absolute accuracy problems dependent on physical factors (only mathematical ones) which makes reasoning about signals and data more straightforward. Also you can depend on a variety of work easily shared and collaborated upon while reducing cost and time to market. Even the 3478A tries to get it into the digital domain pretty quick with that dual slope ADC (turns analogue value into a stream of digital pulses which can be counted). All the readings are calculated in the digital domain in memory (read the operating principles chapter of the service manual).

I like it like this if I'm honest. I just appreciate the old kit and like playing with it.