Emergency help. Scope purchase.

[calexanian]

We need to buy a Scope this week for Tax purposes. We need 4 channel and lots of memory. 100MHZ will do. If all else fails i will just go with a TEK MDO3014 since nobody ever got fired for buying a Tektronix scope, but before we plunk down 4K I wanted to hear from anybody who has one and see if they wish they had chosen something else. 

[LabSpokane]

At this point, make sure it's available.  I just did the same thing on a DMM and Microlease was the only one who had inventory.  One technical point is that the unit must be *in service* by 12/31. 

[calexanian]

Not a problem. The Tek rep has a finger on one somewhere. Literally I need to decide by this afternoon so they can ship it overnight. Just making sure we are not making a mistake. Siglent is making a pretty good argument with all of their free upgrades right now, but then again why do they need to give that stuff away to sell scopes. I have one of their 1000 series scopes and its fine, but it only sees occasional use and not any of the deep functions.

[Electro Fan]

http://www.tequipment.net/Tektronix/MDO3014/Mixed-Signal-Oscilloscopes/?gclid=CNHUuqL368ICFeY-MgodZ1AAGw

Their web site says they have one in stock - might be worth checking.  They offer a discount to EEV members.

Maybe a good idea to check (from whatever source you choose) on the price of any options you might want now or later (hardware and software accessory prices can add up).

[tautech]

Would the SDS2000 Power analysis option be of interest in your field?
I'm having a play with this option ATM, quite daunting the range of features it provides.
Lots of efficency analysis for PSU's etc.
Current probe required for full functionality.

Teq also handle Siglent.

[Wuerstchenhund]

We need to buy a Scope this week for Tax purposes. We need 4 channel and lots of memory. 100MHZ will do. If all else fails i will just go with a TEK MDO3014 since nobody ever got fired for buying a Tektronix scope, but before we plunk down 4K I wanted to hear from anybody who has one and see if they wish they had chosen something else.

I don't have one but from what I hear from some of the labs I work with the MDO3k firmware is still buggy, and they only got these scopes because some clueless monkey in corporate thought the same about not getting fired for buying Tek and not because it was requested by an engineer. The built-in SA is nice in theory but pretty useless for many applications where you need a scope and the SA at the same time.

It's hard to suggest something when we don't know what you need the scope to do but if you don't need the SA then for $4k you could already get a LeCroy WaveSurfer 3024:
http://teledynelecroy.com/oscilloscope/oscilloscopeseries.aspx?mseries=466&capid=102&mid=504
200MHz, up to 4GSa/s instead of 2.5GSa/s, larger screen, faster processing, touch screen. It also comes with much better support than Tek offers (which has become pretty poor in recent years).

It also seems to be in stock with Digikey and Mouser:

http://www.digikey.com/product-search/en?x=0&y=0&lang=en&site=us&keywords=Wavesurfer+3024

http://www2.mouser.com/ProductDetail/Teledyne-LeCroy/WAVESURFER-3024/?qs=%2fha2pyFadujJWeX%2fwfDD9r1Ni8eBeO6prxYfayRwY%2fOihHBwlCzV8g%3d%3d

[nctnico]

Or look at Keysight's (aka HP or Agilent) offerings.

[calexanian]

We need to buy a Scope this week for Tax purposes. We need 4 channel and lots of memory. 100MHZ will do. If all else fails i will just go with a TEK MDO3014 since nobody ever got fired for buying a Tektronix scope, but before we plunk down 4K I wanted to hear from anybody who has one and see if they wish they had chosen something else.

I don't have one but from what I hear from some of the labs I work with the MDO3k firmware is still buggy, and they only got these scopes because some clueless monkey in corporate thought the same about not getting fired for buying Tek and not because it was requested by an engineer. The built-in SA is nice in theory but pretty useless for many applications where you need a scope and the SA at the same time.

It's hard to suggest something when we don't know what you need the scope to do but if you don't need the SA then for $4k you could already get a LeCroy WaveSurfer 3024:
http://teledynelecroy.com/oscilloscope/oscilloscopeseries.aspx?mseries=466&capid=102&mid=504
200MHz, up to 4GSa/s instead of 2.5GSa/s, larger screen, faster processing, touch screen. It also comes with much better support than Tek offers (which has become pretty poor in recent years).

It also seems to be in stock with Digikey and Mouser:

http://www.digikey.com/product-search/en?x=0&y=0&lang=en&site=us&keywords=Wavesurfer+3024

http://www2.mouser.com/ProductDetail/Teledyne-LeCroy/WAVESURFER-3024/?qs=%2fha2pyFadujJWeX%2fwfDD9r1Ni8eBeO6prxYfayRwY%2fOihHBwlCzV8g%3d%3d

Looking at those now

[calexanian]

Or look at Keysight's (aka HP or Agilent) offerings.

Have been looking at those as well, but the big hangup with Keysight (Ughhhh) is their limited memory. We need deep memory for some of what we do. The models with it that fit our requirements do not fit our budget.

[don]

We need to buy a Scope this week for Tax purposes. We need 4 channel and lots of memory. 100MHZ will do. If all else fails i will just go with a TEK MDO3014 since nobody ever got fired for buying a Tektronix scope, but before we plunk down 4K I wanted to hear from anybody who has one and see if they wish they had chosen something else.

I don't have one but from what I hear from some of the labs I work with the MDO3k firmware is still buggy, and they only got these scopes because some clueless monkey in corporate thought the same about not getting fired for buying Tek and not because it was requested by an engineer. The built-in SA is nice in theory but pretty useless for many applications where you need a scope and the SA at the same time.

It's hard to suggest something when we don't know what you need the scope to do but if you don't need the SA then for $4k you could already get a LeCroy WaveSurfer 3024:
http://teledynelecroy.com/oscilloscope/oscilloscopeseries.aspx?mseries=466&capid=102&mid=504
200MHz, up to 4GSa/s instead of 2.5GSa/s, larger screen, faster processing, touch screen. It also comes with much better support than Tek offers (which has become pretty poor in recent years).

I've owned an MDO3000 since October and it is NOT buggy -- it's extremely solid.  By comparison I tried a Siglent SDS2000 and a Rigol DS4000.  I found about 10 bugs in just a couple hours on each of those.  Sent them back and got the Tek.

I needed a lot of memory as well so I ruled out Agilent. Tek has 10M per channel, all modes (analog + digital).  If you need a lot of memory then agilent is a not a good fit.  Even a 3000x with 4M upgrade is only 2M/channel in single aquisition mode.  Drops to 1M if you use the run/stop. And is further halved if you add the digital channels.

The MDO3000 is the best bang for the buck in my opinion if you want for general debug.  Networking works well, realtime access through a webrowser (control + real time screen updates).

The LeCroy is a rebadged Siglent (or the other way around).  It might be great, but I had such bad luck with Siglent SDS2000 I can't recommend anything that shares the same HW or SW.

Also, I've had good luck with Tek support in the U.S. -- might differ in other countries.   Also rubbish that this is a scope purchased by managers.  It's a mini version of the MDO4000 and it's arguably  one of the best scopes for general debug if you ignore the cost.  And cost is the major issue that MDO3000 solves.   High end LeCroy's are nice but I'm very dubious of their low end offerings (3000 series included) until I can take one for a test drive.  I certainly would not blindly by any scope just because it has a LeCroy name.  We have more broken LeCroy's in our lab than any other scope ( might be why they need to service for 7yrs).  At work I use a large variety of Agilent, tek and lecroy scopes -- I would certainly choose the MDO3000 for daily debug over many higher end models from all three big names. 

Biggest complaint I hear about tek is it's an old and slow architecture.  But it gives you 20x the memory of the Agilent on all channels and only slows down in corner cases when performing math intensive operations.  As do all high end scopes.   Try doing an I2C decode / zoom / search on a Lecroy with 32M memory -- it takes minutes.  And you can't count the speed of an agilent 3000x because it's memory is so small.  Move up to an Agilent 9000 or 90000 and the speed will drop because of the increased memory.

[calexanian]

Thanks everybody.

I put the purchase on hold. I was about ready to pull the trigger on the Tek, BUT...... There are a couple features on the LeCroy wave surfer that may make it the better choice for us. I am going to schedule a demo for one to see if it is the real deal, or just marketing bluster for a barley functional features.

I am going to get a couple Rigol DP832's for our new bench and a couple other non critical things which are in stock and can ship out over night to meet the deadline for corporate purposes and take the time to see if the other scopes are really a good fit.

Thank you everybody

[T3sl4co1l]

I've used modern Tek scopes before.  They do not impress me.  Not at all for the price.  I'd rather suggest an Agilent (as in, lightly used?) or Keysight (new, I guess), they look much more responsive.  I would expect comparable electrical and numerical performance from both.

Tim

[Someone]

Or look at Keysight's (aka HP or Agilent) offerings.

Have been looking at those as well, but the big hangup with Keysight (Ughhhh) is their limited memory. We need deep memory for some of what we do. The models with it that fit our requirements do not fit our budget.

There are a few triggering conveniences that hide the limitations of the smaller memories on the Keysight 2000/3000/4000 series scopes, one often used here is the serial triggers are hardware and will work even when the stored points can't reproduce the data. And of course the segmented memory capture, which is also excellent on Lecroy products.

Every time I've been asked to help someone who "needed" deep memory, they could solve the measurement problem in other ways. What specific need for deep memory do you have?

[don]

Thanks everybody.

I put the purchase on hold. I was about ready to pull the trigger on the Tek, BUT...... There are a couple features on the LeCroy wave surfer that may make it the better choice for us. I am going to schedule a demo for one to see if it is the real deal, or just marketing bluster for a barley functional features.

I am going to get a couple Rigol DP832's for our new bench and a couple other non critical things which are in stock and can ship out over night to meet the deadline for corporate purposes and take the time to see if the other scopes are really a good fit.

Thank you everybody

Good deal, best to try before you buy.  LeCroy should be a nice scope if SW works well.  Buggy scopes are bad news.  For example, the oddest bug I saw on the Siglent SDS2000 is chunks of the waveform would disappear at certain zoom levels as you scroll.  But the chunks that disappeared were replaced by some other part of the waveform so it was not obvious.  There were no discontinuities in waveform either as vectors always connected the two pieces of waveform.  No scope is perfect but blatant waveform errors are really tough to deal with.

[Howardlong]

Every time I've been asked to help someone who "needed" deep memory, they could solve the measurement problem in other ways. What specific need for deep memory do you have?

I'd agree with that, after all, how ever did we cope only a few short years ago? Sure it's a convenience, but you also have to also ask yourself how much is "enough"? One meg? Ten meg? One gig?

[Wuerstchenhund]

I've owned an MDO3000 since October and it is NOT buggy -- it's extremely solid. 

As I said I don't have one so I haven't seen any bugs myself (but I had the chance to witness the scope locking up for several minutes when doing CPU-intensive stuff, which happened quite a lot during the simple demonstration I've seen). The complaints I heard weren't about stability but silly bugs like issues with the new probe calibration system and serial decode, or only dBm as vertical units in SA mode. Most people I talked also complained about the scope's sluggishness, although it is apparently much better than previous Tek scopes.

By comparison I tried a Siglent SDS2000 and a Rigol DS4000.  I found about 10 bugs in just a couple hours on each of those.  Sent them back and got the Tek.

Well, that should have been no surprise, considering the long threads on firmware issues with the SDS2000 or DS4000 in this forum. Siglent is still crap at writing software, and Rigol can't be arsed to fix the scope models (DS4000, DS6000) that no-one buys anyways.

I needed a lot of memory as well so I ruled out Agilent. Tek has 10M per channel, all modes (analog + digital).  If you need a lot of memory then agilent is a not a good fit.  Even a 3000x with 4M upgrade is only 2M/channel in single aquisition mode.  Drops to 1M if you use the run/stop. And is further halved if you add the digital channels.

Yes, Keysight should really update the memory configuration on their lower-end DSOX scopes. But I guess we'll see a successor model in the near future anyways.

I wonder why you haven't considered R&S? The HMO3000 offers 8Mpts, 4GSa/s sample rate and good MSO capabilities. It should have been in a similar price range.

The MDO3000 is the best bang for the buck in my opinion if you want for general debug.  Networking works well, realtime access through a webrowser (control + real time screen updates).

I guess this depends on what you want and how much you paid, but you're actually the first person I met who's opinion is overwhelmingly positive.

The LeCroy is a rebadged Siglent (or the other way around).  It might be great, but I had such bad luck with Siglent SDS2000 I can't recommend anything that shares the same HW or SW.

No, the WaveSurfer 3000 is *not* a rebadged Siglent. Unlike the cheap-ass WaveAce, the WaveSurfer 3000 has been developed in cooperation with Siglent who also acts as ODM for the hardware (and who has retained the right to sell the scope under their name in China). The software is also developed by LeCroy and not by Siglent, so your fear of issues like with the SDS2000 firmware are unfounded.

This is nothing new btw, LeCroy did the same in the past with other vendors (i.e. Iwatsu, who manufactured the WaveRunner/WaveRunner2/WavePro 900 Series which was developed by LeCroy, and sold them under their own label in Japan).

Also, I've had good luck with Tek support in the U.S. -- might differ in other countries.   Also rubbish that this is a scope purchased by managers.  It's a mini version of the MDO4000 and it's arguably  one of the best scopes for general debug if you ignore the cost. 

Well, it's not rubbish that these scopes are mostly bought by managers. I'm working in various corporate environments and in the few occasions where Tek scopes are bought they are so because someone in procurement bought them by basic specs or because they have to buy a certain contingency from them. In the last say 5+ years I've never met an EE who actually preferred a modern-day Tek over any other scope, and due to my work I meet quite a large number of EEs.

And no, the MDO3000 is not just a mini version of the MDO4000. They are completely different designs (i.e. the MDO3k only has four ADCs which is why you can't use the SA at the same time as the scope; the MDO4k has eight).

And cost is the major issue that MDO3000 solves.   High end LeCroy's are nice but I'm very dubious of their low end offerings (3000 series included) until I can take one for a test drive.  I certainly would not blindly by any scope just because it has a LeCroy name. 

I wouldn't buy *any* scope (aside from a really cheap entry level scope like the DS1000z maybe) that I can't test beforehand. No matter who makes it.

We have more broken LeCroy's in our lab than any other scope ( might be why they need to service for 7yrs).

Interesting. Can you elaborate? What scopes are these, and why are they broken? I know that some of the old W2k-based units (WR6k, WP7k, WM8k) had some driver issues but that was in the beginning and has been fixed long since. The also was an issue with the acquisition board connectors on the WaveSurfer 400 and early units of the WaveRunner Xi, but both have been fixed by LeCroy. So I'm curious to hear what problems you had.

Biggest complaint I hear about tek is it's an old and slow architecture.  But it gives you 20x the memory of the Agilent on all channels and only slows down in corner cases when performing math intensive operations.  As do all high end scopes.

   

Not really, at least not to the same extent.

Try doing an I2C decode / zoom / search on a Lecroy with 32M memory -- it takes minutes.

No, it doesn't. Decoding I2C or searching on a sequence on a WavePro 7300A with 48Mpts takes several seconds and feels slow like hell, but not minutes. And this scope is now over 6 years old and uses a slow P4 processor with single PCI bus. It's much faster on a modern scope with a fast CPU and PCIe.

If it takes several minutes on your LeCroy scope (I guess a WavePro 7k or WaveMaster 8k?) then there's something that needs addressing, i.e. an old X-Stream software or a weak CPU. I guess you're running the scope with basic config (i.e. 256MB RAM and 1.3GHz Celeron)?

This aside, you can always increase the speed by lowering the memory to say 10Mpts and the sample rate to 2.5GSa/s (or what the equivalent rate is the MDO3k uses at the required timebase).

[Wuerstchenhund]

I put the purchase on hold. I was about ready to pull the trigger on the Tek, BUT...... There are a couple features on the LeCroy wave surfer that may make it the better choice for us. I am going to schedule a demo for one to see if it is the real deal, or just marketing bluster for a barley functional features.

This isn't a Siglent rebadge, the WaveSurfer 3000 has been developed by LeCroy together with Siglent. The software is also developed by LeCroy.

I'm not sure if you've seen it but KF5OBS has made a short review of this scope:

[nctnico]

Every time I've been asked to help someone who "needed" deep memory, they could solve the measurement problem in other ways. What specific need for deep memory do you have?

I'd agree with that, after all, how ever did we cope only a few short years ago? Sure it's a convenience, but you also have to also ask yourself how much is "enough"? One meg? Ten meg? One gig?

More memory helps to avoid having to setup complicated triggers. Recently I used my MSO to debug a project involving driving a TFT display from a CPLD. I just captured 2 frames of video data and zoom in on the problem area. Having over 10Mpoints of data makes that easy.

IMHO the Lecroy Wavesurfer 3000 looks like a good deal if the options aren't driving the price up to insane levels.

[Wuerstchenhund]

I'd agree with that, after all, how ever did we cope only a few short years ago? Sure it's a convenience, but you also have to also ask yourself how much is "enough"? One meg? Ten meg? One gig?

Large memory can be a tremendous help. And since fast memory is cheap these days there's no reason that new scopes should come with 1Mpts or 4Mpts memory only.

However, it's no panacea, and what many people ignore is that larger sample memory means more strain on the scope's processing which usually means things get slow.

[EEVblog]

Siglent is making a pretty good argument with all of their free upgrades right now, but then again why do they need to give that stuff away to sell scopes.

If you are willing to spend $4K then you wouldn't buy a Siglent!

[nctnico]

Well.. you could buy two scopes instead of one.

[EEVblog]

I've used modern Tek scopes before.  They do not impress me.  Not at all for the price.

I agree on the performance aspect (they are sluggish to use), but the MDO with the spectrum analyser adds a lot of value. So I think the MDO3000 is a decent buy if you can make good use of the (limited, but better than FFT on scope) spectrum analyser.

[don]

The LeCroy is a rebadged Siglent (or the other way around).  It might be great, but I had such bad luck with Siglent SDS2000 I can't recommend anything that shares the same HW or SW.

No, the WaveSurfer 3000 is *not* a rebadged Siglent. Unlike the cheap-ass WaveAce, the WaveSurfer 3000 has been developed in cooperation with Siglent who also acts as ODM for the hardware (and who has retained the right to sell the scope under their name in China). The software is also developed by LeCroy and not by Siglent, so your fear of issues like with the SDS2000 firmware are unfounded.

This is nothing new btw, LeCroy did the same in the past with other vendors (i.e. Iwatsu, who manufactured the WaveRunner/WaveRunner2/WavePro 900 Series which was developed by LeCroy, and sold them under their own label in Japan).

It's foolish to assume HW and SW are completely decoupled from one other.   I agree that Siglent can build decent "HW", but I would never assume that LeCroy can work around all HW issues.  I would not put Iwatsu and Siglent in the same class at this point, Siglent is still learning.  I also saw issues with SDS2000 intensity grading as well as the waveform corruption I mentioned earlier.  Siglent has no concept of intensity grading a static waveform through compression as you zoom in /out.   Are these issues all SW?  I doubt it.  Is Siglent designing fpga or LeCroy? I would think Siglent if they own the HW.  In which case cross your fingers.

I saw enough negative things with SDS2000 that a partnership between Siglent and LeCroy means very little at this point other than a cost saving measure.   

Plus the WS 3000 maui interface looks pretty stripped down and somewhat archaic.  Not sure if they are performance limited on HW side or this is just how the existing GUI api's on the embedded windows looks. If you are used to xstream you will be dissapointed with the eye candy (or maybe pleased if you find GUI is excessive / in your face on xstream).

Also, I've had good luck with Tek support in the U.S. -- might differ in other countries.   Also rubbish that this is a scope purchased by managers.  It's a mini version of the MDO4000 and it's arguably  one of the best scopes for general debug if you ignore the cost. 

Well, it's not rubbish that these scopes are mostly bought by managers. I'm working in various corporate environments and in the few occasions where Tek scopes are bought they are so because someone in procurement bought them by basic specs or because they have to buy a certain contingency from them. In the last say 5+ years I've never met an EE who actually preferred a modern-day Tek over any other scope, and due to my work I meet quite a large number of EEs.

And no, the MDO3000 is not just a mini version of the MDO4000. They are completely different designs (i.e. the MDO3k only has four ADCs which is why you can't use the SA at the same time as the scope; the MDO4k has eight).

I think there are regional differences driving our opinions of what is common and what isn't.  I can't imagine a manager choosing equipment where I work.  It's all driven by engineers.  Managers simply approve the equipment need, make sure resources are managed accordingly and approve the req (amazing, right?).   Engineers meet the reps, get demos, and know what they need to complete their jobs followed by ordering the equipment.  I'm not the only engineer that like's Tek where I work.  I'd say population is about 50/50 in terms of agilent or Tek.  And then some love LeCroy, some hate, some are indifferent.  The only LeCroy's that get used are less than 2 years old and weigh about 80lbs and cost $$$$.  Older ones are too slow with a steeper learning curve for general debug.  For general debug (low to mid range scopes) I'd say LeCroy is a distant number three with Agilent / Tek tied for first.  We have plenty of LeCroy scopes, they just are not used for whatever reason.    At the high end it's predominately Agilent but LeCroy is up there as well.   Regarding support, we have Tek and Agilent on site all the time for all equipment needs.   Issue with Tek or Agilent? Call the rep and they generally resolve. Their success rate is nothing to complain about at least.   Not a single Hameg scope by the way.  Plenty of R&S equipment, but no scopes.

I understand the difference in SA's on the MDO3000 and MDO4000.  By baby MDO4000, I mean the GUI is identical for the scope as well as basic operation.  The triggers are the same, acquisition modes are the same, same probe interface, cursor modes the same, same excellent intensity grading.  What you lose is higher screen res.  You can only display 4 measurements instead of 8. You have 10M of memory instead of 20M.  And no time corrlelated SA and scope.  And SA is lower performance, scope has lower DC accuracy at less than 50mV (but in practice it blows away the specs anyway so who cares).   That's what I mean by a baby version. 

Just because you can't use simultaneous SA and scope does not make the feature silly or a gimmick.  How many setups do you see that have time correlated SA's and scopes to begin with like the MDO4000?  MDO3000 loses that, but press a button and you have an SA.  Press a different button and you have a scope.  Easy to go back and forth and still very useful even though not time correlated.   On a side note, people complain that it's SA is slow in reviews and can lock up, but keep in mind it has a 3GHz capture BW (absolutely huge) and reviewer probably running with a span to RBW ratio of 200000:1.  Reduce the ratio to something you normally use and speeds way up.   

We have more broken LeCroy's in our lab than any other scope ( might be why they need to service for 7yrs).

Interesting. Can you elaborate? What scopes are these, and why are they broken? I know that some of the old W2k-based units (WR6k, WP7k, WM8k) had some driver issues but that was in the beginning and has been fixed long since. The also was an issue with the acquisition board connectors on the WaveSurfer 400 and early units of the WaveRunner Xi, but both have been fixed by LeCroy. So I'm curious to hear what problems you had.

I would have to take inventory, but at least a few with OS problems (one says license is bogus and chooses to ignore anything we input), and a couple that just hang when the scope is up.  And plenty with missing knobs of course.   

Try doing an I2C decode / zoom / search on a Lecroy with 32M memory -- it takes minutes.

No, it doesn't. Decoding I2C or searching on a sequence on a WavePro 7300A with 48Mpts takes several seconds and feels slow like hell, but not minutes. And this scope is now over 6 years old and uses a slow P4 processor with single PCI bus. It's much faster on a modern scope with a fast CPU and PCIe.

If it takes several minutes on your LeCroy scope (I guess a WavePro 7k or WaveMaster 8k?) then there's something that needs addressing, i.e. an old X-Stream software or a weak CPU. I guess you're running the scope with basic config (i.e. 256MB RAM and 1.3GHz Celeron)?

This aside, you can always increase the speed by lowering the memory to say 10Mpts and the sample rate to 2.5GSa/s (or what the equivalent rate is the MDO3k uses at the required timebase).

It definitley did.  I used I2C on an 8zi I believe, with segmented memory and some search operations, cursors, zoom and maybe a math.  It was so slow.  I basically had to set it up and tell the person I was working with to let it do it's thing.  Zooming was painful and cursoring was jumpy.  I'm not saying this is the norm - likely is not.  The contrary actually, it's a nice scope.  But Tek gets dinged for being slow when it's no different than high end scopes with deep memory. I think Agilent 3000x series with limited memory brain-washed the general population into thinking all scopes should be as responsive even though it's doing most of the calculations on screen vs. waveform with minimal memory.

[calexanian]

Wrong but sound.. I just had to dig this up..

[n45048]

Quick, Call 911 No Call E14 instead.

I thought the number was 0118 999 881 999 119 7253 

And for those who don't get the reference: