Professional Scope Purchasing Criteria

[coppice]

In instrumentation "flexible and "upgradeable" are generally code words for sucking more money out of customers. Some recent products can be upgraded up simple software changes, and these are probably OK in general. Things requiring chunks of hardware be replaced usually ends up costing a lot more, and only a small percentage of units ever get upgraded.

[rx8pilot]

You might also consider getting a used/remanufactured one from Keysight on their ebay store. If they say 'Premium' it usually comes with a three year warranty, note some don't, so look at the description.

http://stores.ebay.com/Keysight/PREMIUM-Used-/_i.html?_fsub=874614011&_sid=869664151

For example here if a 500MHz MSO for 6K, its possible they might go down to $5500:

http://www.ebay.com/itm/Keysight-Premium-Used-MSOX3054A-Oscilloscope-500-MHz-Agilent-MSOX3054A-/181598244400?pt=BI_Oscilloscopes&hash=item2a48194e30

Its unfortunate the upgrade to the ultimate scope promotion ended today.

They have less expensive models as well.

You are right! 200Mhz 4ch+8digital for $2500! At that price I can spend some money on active and differential probes too. When/if I out grow that model, I would not care so much about any upgrades, I would simply replace it and re-purpose or sell the older one.

Refurbs may be a great option.

[LabSpokane]

Refurbished is an option you'd really be crazy not to consider in the US. The scope will come from Malaysia likely, but we don't have customs issues like other countries do. My experience was amazingly painless and I got a MSOX2024 for a bit less than $2500 all in with sales tax.  It also ships FedEx so there's very little time in transit for something bad to happen to it.

[Someone]

You might also consider getting a used/remanufactured one from Keysight on their ebay store. If they say 'Premium' it usually comes with a three year warranty, note some don't, so look at the description.

http://stores.ebay.com/Keysight/PREMIUM-Used-/_i.html?_fsub=874614011&_sid=869664151

For example here if a 500MHz MSO for 6K, its possible they might go down to $5500:

http://www.ebay.com/itm/Keysight-Premium-Used-MSOX3054A-Oscilloscope-500-MHz-Agilent-MSOX3054A-/181598244400?pt=BI_Oscilloscopes&hash=item2a48194e30

Its unfortunate the upgrade to the ultimate scope promotion ended today.

They have less expensive models as well.

You are right! 200Mhz 4ch+8digital for $2500! At that price I can spend some money on active and differential probes too. When/if I out grow that model, I would not care so much about any upgrades, I would simply replace it and re-purpose or sell the older one.

Refurbs may be a great option.

The 2000 Series do not support active probes, have fewer options available, and have less memory accessible, you get what you pay for.

[rx8pilot]

The 2000 Series do not support active probes, have fewer options available, and have less memory accessible, you get what you pay for.

You saved me from making a poor decision, Than you. I do not want to eliminate active probes right off the bat.

Tek MDO3000 or Keysight X3000 seem to be the top choices so far. I can see and touch both of them in the same place close to where I live. The Tek has spectrum, but I don't need that right now and have no plans to need it. Who know what the future holds - RF is pretty far outside of my comfort zone.

[tautech]

For a 200 MHz scope you don't really need an active probe anyway, ordinary probes will be fine.

A powered differential probe should have alternative means for power supply.
Separate PSU or USB or battery are the usual options.
Failing that you can use 2 normal probes and engage the DSO maths. 

[robrenz]

......
Failing that you can use 2 normal probes and engage the DSO maths. 

Won't CMRR be vastly worse with that method?

[coppice]

......
Failing that you can use 2 normal probes and engage the DSO maths. 

Won't CMRR be vastly worse with that method?

Both CMRR and dynamic range are problems with that method. Sometimes AC coupling removes the dynamic range issue, but if you really need DC coupling to properly see the nature of the differential signal you are stuck. DSO maths as a differential tool is useful, but far less useful than a real differential input.

[rx8pilot]

Both CMRR and dynamic range are problems with that method. Sometimes AC coupling removes the dynamic range issue, but if you really need DC coupling to properly see the nature of the differential signal you are stuck. DSO maths as a differential tool is useful, but far less useful than a real differential input.

Just so it is clear, EE is my second career so I do not have a lot of experience. I come from the ME world and have owned multiple businesses including soup to nuts manufacturing. What I learned is that asking smart people good questions tends to reveal useful answers. Here is an example of what I am trying to measure, maybe you have some insight....
Measuring dI/dt across a sense resistor. On the low end, its only a few mV while an inrush spike may be a few hundred mV. My thinking is that the dual probe method would be too crude to make this even remotely accurate. This is what I was planning to use a differential probe for. The good news is that this a low speed event that could easily be captured by a low cost BK PR-60 (at least I hope it would work)
http://www.tequipment.net/BKPR60.html?v=0
If that is a workable solution, then the auto-ranging self powered probe ports on the higher-end don't really help for that particular scenario. That little guy is self powered. I am still questioning whether this is the right approach to measuring transient current. One note is that the sense resistors are already in the circuit - not as an external measurement fixture.

For the single-ended active probes, I don't have a specific need and agree that at 200Mhz they won't do much more than burn a hole in my checkbook. I was only considering those if I developed the need for speed and upgraded the box to a 1Ghz - I would appreciate the having the integration. It seems that most of the active probes I have seen on the market are expecting the integrated interface. I did not understand they could be used with external power. So, single ended active probes are merely a consideration for the future.

Side note on my tool purchasing philosophy that has served me well. I have always purchased way more tool than today's job calls for to give me room to grow. For example, when I needed to machine a bunch of simple parts i went out and did some research and came home with a brand new 5 axis CNC mill. It was FAR beyond any skill that I had at the time. In fact, I did not even know how to turn it on. Learning to be a machinist making extremely precise parts advanced my ME skills more than any class could have ever taught. I can now engineer a sophisticated mechanical system that works the first time and has a predictable cost. It's 50% knowledge and 50% dark art. So far EE seems to be a similar experience. I feel that I will be punished if I go low-end on gear and software. My Rigol DS1102E feels like a manual bridgeport mill - good for learning, bad for making money.

[nctnico]

If you want to measure current then you need a current probe. Measuring a few mV across a resistor is not a good idea due to noise.

[tautech]

If you want to measure current then you need a current probe. Measuring a few mV across a resistor is not a good idea due to noise.

Yep.
For the Power electronics the OP is doing a good current probe is invaluable IMO
Just make sure the output (mV/A) will match up with the attenuation settings in the new DSO you might buy for accurate readouts on screen.
If you must have a sense resistor on the PCB, you can change to a wire link type and attach the current probe there.
Otherwise clip over any off PCB cabling for readings.

[rx8pilot]

So I did some tests measuring current, the hard part is knowing if the results are believable. I also looked at DC current probes at little bit and they seem to be a challenge to get into a small space. They will, however be great for measuring current right at a panel mounted connector with wires.

For a test, I used a 1x passive probe [switchable cheap-o, 1x bandwidth limited] and the scope at 2mV/div and a ground spring right at the tip of probe. I isolated the circuit to measure across a 2512 .005ohm resistor with a single probe. The noise did not look crazy at all, relatively flat line with only a tiny offset. I powered the circuit with known external v & i measurements and they matched the scope results at least pretty close.

Does this sound like a reasonable way to measure at the sense resistors? I read a Tek paper about probes saying that the active probes will have better noise rejection with low voltage applications. I think that may be splitting hairs though as they go on to say their general purpose passive probe "is also very capable of providing accurate low voltage measurements".

[robrenz]

You probably already know this but for transients your sense resistor will need to be non inductive.

[rx8pilot]

Yes, that makes 'sense'
I am using these resistors http://rohmfs.rohm.com/en/products/databook/datasheet/passive/resistor/chip_resistor/pmr.pdf which make no mention of inductance although they are specific to current sensing so maybe that is inferred in such a device?

[Vgkid]

For those, It doesnt look like inductance will be much of an issue.

[tautech]

So I did some tests measuring current, the hard part is knowing if the results are believable. I also looked at DC current probes at little bit and they seem to be a challenge to get into a small space. They will, however be great for measuring current right at a panel mounted connector with wires.

Thats very easy to check your setup using a DC circuit and a DMM and the scope as well.
Convert the turn on waveform "step" to the measured current.

Never needed a DC current probe myself, but i'm sure there is good argument for certain applications.
Others should chime in with advise on the need for DC current probes.

Does this sound like a reasonable way to measure at the sense resistors?

Sure it is for low frequency stuff, the only issue is isolation, forgetting to do it or another person whose skills are lacking.
I think for production you should have a differential probe to minimise F*** ups. 

[nctnico]

So I did some tests measuring current, the hard part is knowing if the results are believable. I also looked at DC current probes at little bit and they seem to be a challenge to get into a small space.

There is an easy solution for that: bring out a wire made of litz wire to make a signal accessible. Litz wire has a very low self inductance and skin effect so the effect on the signal is neglectible. I myself use a current probe in a box which uses banana sockets to feed the signal through the current sensors. The beauty of current is that it flows no matter what.

[rx8pilot]

In the interest of completely eliminating all options - I took a look as some USB scopes. For a brief moment, the Pico scopes seemed compelling. They have demo software that can be downloaded and played with to get an idea of what can be done. The electrical specs are great in some respects like memory, but hobby level on others like sampling rate - 1GS/s. Not hideous for the modest speed work that I am doing so I continued to look. Obviously screen size is whatever you attach to the computer and that is pretty great. Easy to save data to network without an add-on interface and proprietary software. Serial decoding, waveform generator, spectrum, mask limits, are all included. leveraging the enormous memory, you can zoom in a lot, but it saves the data in windowed segments - you cannot scroll left/right as far as I can tell.

The part that I could not deal with is not having buttons and knobs to manipulate the interface. Mousing around and clicking up/down arrows to change timebase or v/div is a show stopper. For professional audio and video, many companies have control surfaces that provide a more intuitive interface for software based recording and editing. The surfaces have LEDs, displays, knobs, buttons, etc that mimic how it was done in the old days. You can map these functions to the keyboard which is helpful, but still too cumbersome.

Does anything like that exist for USB scopes? Although they are not perfect, they will likely be contenders in the near future and the human interface seems to be the biggest limitations for me.

I am just about to declare the Tek MDO3000 the winner for me. It seems to be a great balance of price, performance, and features for daily use.

[tautech]

https://www.eevblog.com/forum/testgear/mdo3000-is-have-some-hardware-bug-in-large-signal-test-warning/msg561411/#msg561411

[rx8pilot]

Well, I guess that I should look at those issues during the demo period. I have owned several of the high-end Tek video measurement products and never had any challenges with service and after-sale support. There could be more to the story as well that we do not know about. I will not shift my opinion based on a first post by someone, but will ask the Tek people about the observation.

Easy enough to connect it to a square wave right?

[tautech]

Well, I guess that I should look at those issues during the demo period. I have owned several of the high-end Tek video measurement products and never had any challenges with service and after-sale support. There could be more to the story as well that we do not know about. I will not shift my opinion based on a first post by someone, but will ask the Tek people about the observation.

As you should. 

It seemed by how I read that post, he was using it in an unusal way, maybe just as he had always done and the latest 3000 HW & FW were giving him a different result.
Usage patterns DO vary based on taught procedures or habit or even misunderstandings.
But the fact that the other EE's didn't want to use them either should make one wonder.

[rx8pilot]

I have worked with [signed their paycheck] MANY EE's that make snap technical judgments that they are so sure of they fail to look at the problem anymore. I saw this so much that I decided to be an EE myself so that I would know when to call BS.

So many simple problems get inflated to the point that the simple solutions are ruled out, including the possibility that there is no problem at all - just a perceived problem.

[TunerSandwich]

I have worked with [signed their paycheck] MANY EE's that make snap technical judgments that they are so sure of they fail to look at the problem anymore. I saw this so much that I decided to be an EE myself so that I would know when to call BS.

So many simple problems get inflated to the point that the simple solutions are ruled out, including the possibility that there is no problem at all - just a perceived problem.

That is most certainly a double edged sword. 

On one hand you want to make informed decisions, that go beyond the opinion of one individual. 

On the other hand, you don't want to get so bogged down in discovery, that you wind up becoming a stunt double, for the person's paycheck you are signing. 

The other issue is that the pace of technology is changing so rapidly, that if you waffle too long in endless marketing and datasheets, you might miss an opportunity to buy something that was "right for the job". 

I have never found one tool (in any business I have run) that was right for everything.  Granted some are more versatile, but I would gladly follow lessons learned, and focus on what you need right now.....if it happens to amortize/depreciate across multiple projects....and outlives it's projected service life....that is gravy 

I would personally look at Agilent/Keysight direct page on ebay.  They have some amazing deals on refub. "high end" scopes there.  Shockingly good in some cases.  That is if you are sold on Keysight/Agilent.

I personally decided on a used LeCroy (recently), as I had much the same needs you do, although I am not sure what your technical requirements are.  There simply wasn't a better value than the unit I purchased, and I am quite happy with the tool.  Despite being 2009 vintage, it's already replaced my 2 year old Agilent on the bench. 

Tek represents some of the poorest value (IMO) in "high end" scopes.  The offerings in the last year or so are quite interesting, but IMO anything from the prior 4-5 years was WAY behind the curve, as far as processing power vs $ spent. vs competition. 

I think for MSO under $5-8k Hameg is offering some extremely nice hardware.  I always like their UI and the way they drop information onto the screen.  they make fantastic use of real estate IMO. 

If I was looking for another scope right now, for some "serious" work....and I wanted to buy factory direct...I would look at a refurb Agilent MSO9000, a new LeCroy HDO8000 or if my budget was more inline with yours, a top of the range (optioned out) Hameg MSO

IMO the Hameg represents the best value in a new scope, in the mid-range category. 

If you want to run a lot of math/operators and want a ton of flexibility in the way signal is routed, plus want to do some custom UI or app development....I have never touched a tool that does it, quite as good/same way as an upper range LeCroy.  I am personally happy enough with my used acquisition, to go further and investigate a brand new HDO8000.  Which is saying a lot, considering Agilent filled up my bench space up until very recently. 

Some more technical insight would be helpful in developing a purchasing criteria.....I.E. how much flexibility do you really want, and what exactly do you envision using the scope for?  What is day to day for me, might be considered "minority" for others...

[cncjerry]

I was in your same position about a year ago wanting to move up from an analog scope into the DSO age.   I bought an older used Tektronix TDS744a then a TDS3054b.  I don't regret it at all.   Yes they don't have all the bells and whistles of the newer scopes but the value for the dollar was exceptional.

[Wuerstchenhund]

Tek represents some of the poorest value (IMO) in "high end" scopes.  The offerings in the last year or so are quite interesting, but IMO anything from the prior 4-5 years was WAY behind the curve, as far as processing power vs $ spent. vs competition.

They are indeed. The DPO/MDO3k is interesting but that aside they're so much behind the curve it's not even funny, especially on the high end (where they seem to be still stuck at 33GHz anyways). 

Their sales staff also seems to be really desperate.

Tek clearly has passed its zenith. None of the places I know are considering Tek scopes anymore. The majority is now solidly Agilent/Keysight.

I think for MSO under $5-8k Hameg is offering some extremely nice hardware.  I always like their UI and the way they drop information onto the screen.  they make fantastic use of real estate IMO.

Let's say they're definitely some of the better ones, although they could make much better use of their VGA resolution displays.

Their MSOs are attractive if you can live with the very limited choice of decode options.
 

IMO the Hameg represents the best value in a new scope, in the mid-range category.

Hameg isn't really mid-range (which starts with the R&S RTM), all the Hameg kit is solidly placed in the entry-level sector. Even their top-of-the-range scope (HMO3000) is still a entry-level scope. 

Some more technical insight would be helpful in developing a purchasing criteria.....I.E. how much flexibility do you really want, and what exactly do you envision using the scope for?  What is day to day for me, might be considered "minority" for others...

Indeed, a bit more information would certainly be helpful.