DSO triggering issues (Lecroy WaveAce224)

[madworm]

I'm currently testing my new-ish scope with a test-jig that creates hard-to-trigger digital signals, aperiodic stuff, missing pulses and the likes.

Unfortunately quite a lot of things don't seem to work or are at least very very finicky with regards to the parameter values. Now either I'm too impatient and fail to set all necessary options or something is wrong with the DSO itself.

One trigger option is "pulse", which is supposed to trigger on pulses of known length. Sounds simple enough. The test signal I use is a pulse-train of short pulses of increasing length. Every pulse is unique in length, so no ambiguities that could confuse the thing.

Just to make sure I haven't completely lost my senses, please tell me what you think about this screenshot. Personally I think it is a bug.



In the tested range 1µs - 9µs the scope is consistently off by 2µs, in the sub-µs-range it gets worse (e.g. it triggers on a 250ns pulse when set to something completely different).

I know that this is the bottom-end of the Lecroy scopes and just a re-branded Chinese something, but I would have hoped they had done some more testing. It's pretty annoying that there will probably be no firmware fixes for this. Not enough money in it for them... only reputation. Does that still matter these days? Who knows.

[amspire]

I think

A. You have the trigger level set very high - well within range of the noise level of the high value. So it may have triggered on what looked like a 6uS pulse to the trigger, and

B. The people who wrote the LeCroy scope software are the same as the ones who wrote the Rigol DS1052E software.

Richard.

[madworm]

Well, I can lower the trigger level. Doesn't change a thing. And sometimes I have to move it close or even above the edge to make it trigger.

But I have more issues. Slope-trigger has its quirks too ;-( It works acceptably with a sine, but triggering on a runt-pulse (capacitor charge curve) in the presence of other bigger digital pulses... very odd behaviour sometimes.

[free_electron]

I'd go even further than amspire... The poeple who made the lecroy are the same that made the rigol.... And the ,now defunct, 3000 series of agilent. ( not the 3000 that dave reviewed. Go to agilent.com go scopes click at the bottom on 'dicontinued' you will find it there )


 Here is a bit of backgound info... Agilent wanted acheap scope and partnered woth rigol. The deal was such that, rigol would make x variants. Agilent picked a number to be branded agilent. The rest could be sold under various names. Rigol did the hardware design, assisted by agilent. Agilent did a lot of the software and scope algorithms.

Take the users manual of a agilent 3000 , rigol , lecroy and some other brands.. Look at the screencaptures. Apart from the logo they are all identical.

Agilent used this to bridge the gap until they could release their own machines ?
I suspect their 1000 family may still be a crossbreed

[cybergibbons]

I found the triggering on Rigols to be less than reliable. More options than the GW Instek but they don't work well.

[madworm]

Well, I don't mind re-branding or out-sourced design at all. IF the resulting device lives up to the brand name. So far I have only known the older Lecroy WaveRunner series (the gray-ish ones with amber screens) and they simply did the job. Of course I can't expect this scope to behave like Lecroy's latest 65GHz DSO for 450.000$, but if it offers certain features I do expect them to work without fuss. If some "feature" doesn't work, don't put it in.

The best example for rotten design is the "Adjust" knob. It is used for trigger hold-off, pulse-width settings for pulse-trigger and so on. It does NOT have a coarse/fine setting. You can imagine how long it takes to get from the ns-scale up to ms! You may spend minutes twisting this darn thing. They could have used the push-button of it to simply copy the current timebase s/div, but no... I bet they never really worked with this scope before it was released into the wild.

Or in other words: when I fork out the money for expensive Swiss dark chocolate I simply don't expect to get Hershey (which simply sucks).

[free_electron]

Lecroy WaveRunner

 Hershey (which simply sucks).

You mean rebranded Iwatsu's ... oh wait .. amber screens ... don't know. i'm thinking of the 540 series waverunners. i took oner apart a couple of years ago to repair the touchscreen. every single board had 'iwatsu' stamped on it ...

Hershey .. poo flavoured sawdust you mean ...

[madworm]

The old Lecroys I knew had amber cathode-ray screens, no touch screen anything. And most importantly NO windows. A bit prone to burning in, but that's what you get for looking at the same trace all the time. When we wanted an upgrade they insisted to exchange our scope with a new "all improved" one that came with windows 2000 - and a virus scanner. Ewwww. The most shocking thing was the boot-up time. In practice it wouldn't have mattered, but still shocking.

Regarding Hershey's. Yes, saw-dust. That was my first impression too. I wonder how people manage to eat it. Probably because they simply don't know what the real stuff tastes like, which can be had for little money too.

[T4P]

Regarding Hershey's. Yes, saw-dust. That was my first impression too. I wonder how people manage to eat it. Probably because they simply don't know what the real stuff tastes like, which can be had for little money too.

Awful . It's not even remotely similar to Swiss Dark Chocolate .

[madworm]

I've had a few words with Lecroy support.

After exchanging a few emails and testing on their side, they have confirmed that there is indeed a bug in the pulse-trigger. They've notified their dev. people about it. Fingers crossed.

Maybe (hopefully) they'll also fix the darn adjust button.

[EEVblog]

I've had a few words with Lecroy support.
After exchanging a few emails and testing on their side, they have confirmed that there is indeed a bug in the pulse-trigger. They've notified their dev. people about it. Fingers crossed.

Nice work.
It's the result diligent users that this stuff gets found and fixed.

Dave.

[saturation]

Those screens look very much like Rigol 1000 series, i.e., the 1052e and such.  I've noticed too a bit of a weirdness in the trigger circuits, but I haven't tried to quantify them.  Other posts suggests these scopes maybe far more related than just firmware.

I wonder since they look like they share the same firmware, or even more, if the same exact bug exists there?

I have been meaning to test it for months but haven't had the time.

But will make a note of this thread and hopefully get to it this summer.

I've had a few words with Lecroy support.

After exchanging a few emails and testing on their side, they have confirmed that there is indeed a bug in the pulse-trigger. They've notified their dev. people about it. Fingers crossed.

Maybe (hopefully) they'll also fix the darn adjust button.

[Rufus]

I've had a few words with Lecroy support.
After exchanging a few emails and testing on their side, they have confirmed that there is indeed a bug in the pulse-trigger. They've notified their dev. people about it. Fingers crossed.

Nice work.
It's the result diligent users that this stuff gets found and fixed.

Dave.

If companies were diligent in the first place we wouldn't have this stuff to find and fix.

This doesn't look like some obscure condition, pulse width triggering isn't right so it was never tested properly or was broken and never re-tested.

[madworm]

A small update:

They've done some more testing. They claim the trigger itself does work correctly ( = it only triggers if a certain condition is met ), but the signal is then displayed with some time offset. This then makes the trigger symbol appear at a wrong spot. At least that's what they say.

[FreeThinker]

A small update:

They've done some more testing. They claim the trigger itself does work correctly ( = it only triggers if a certain condition is met ), but the signal is then displayed with some time offset. This then makes the trigger symbol appear at a wrong spot. At least that's what they say.

Oh that's all right then, It's not broken... simply retarded. Great!

[T4P]

A small update:

They've done some more testing. They claim the trigger itself does work correctly ( = it only triggers if a certain condition is met ), but the signal is then displayed with some time offset. This then makes the trigger symbol appear at a wrong spot. At least that's what they say.

Oh that's all right then, It's not broken... simply retarded. Great!

Lecroy .... some say the best oscope maker ... eh 

[madworm]

Well, I'm sure the 60GHz models are better than this one. And probably tested rigorously before shipped. 450k$ pays for a lot of testing.

Still I wonder how it can be that these issues are still present in the current firmware. This scope is on the marked for 2-3 years now. Maybe I'm the only one who bought it... and I dare complain!

[T4P]

Well, I'm sure the 60GHz models are better than this one. And probably tested rigorously before shipped. 450k$ pays for a lot of testing.

Still I wonder how it can be that these issues are still present in the current firmware. This scope is on the marked for 2-3 years now. Maybe I'm the only one who bought it... and I dare complain!

Yeah their 60GHz models are sex on a stick, but this is ridiculous.

[A Hellene]

That's funny!

This is a snippet of the Rigol DS1000E/D latest firmware v2.06 (00.02.06.00.01) decompilation list, which does not seem to exist in the previous FW releases --at least not unencrypted:

Code: [Select]

SDRAM:6C6128 aDso1000:        ascii "DSO1000",0
SDRAM:6C6130 aAgilent:        ascii "Agilent",0
SDRAM:6C6138 aAgilentTechnol: ascii "Agilent Technologies, Inc.",0
SDRAM:6C6153                  db    0
SDRAM:6C6154 aScreenOscillog: ascii "Screen Oscillograph",0
SDRAM:6C6168 aCopyrightC2009: ascii "Copyright (c) 2009 Agilent.",0
These plaintext, null terminated C strings, are loaded to the system SDRAM space during booting up but they seem to be inactive in the DS1000E/D models.

This confirms the previous assertions that (at least) the Rigol DS1000E/D design was/is assisted by Agilent.

Madworm, I do not know if this helps but responding to Saturation's request I confirmed at the following message that the DS1000E pulse triggering seems to be working as advertised.


-George

[madworm]

The pulse triggering thing is a tricky business. It seems to work for purely periodic signals, but that is quite pointless for me. I use edge triggering for that. I need it to sift out the right pulse when giving the scope messy aperiodic or multi-periodic signals.

Maybe I should ask Agilent to give Lecroy a hand

Might lead to an interesting publicity war.

[drieg]

...
I know that this is the bottom-end of the Lecroy scopes and just a re-branded Chinese something, but I would have hoped they had done some more testing. It's pretty annoying that there will probably be no firmware fixes for this. Not enough money in it for them... only reputation. Does that still matter these days? Who knows.

I bet they are made by Siglent, try to ask them for a firmware update