cheap logging scope for recording and analyzing intermittent faults with VFDs?

[JohnXJohn]

Halp!
 Looking for a affordable oscilloscope to efficiently record and analyze intermittent faults with vfd equipment. My management isn't that bright, they didn't update equipment in time, now we stuck with ancient vfds, that trip for unclear reasons. Im just using this as an opportunity to learn, prepare for my next job .  So paying for equipment out of pocket. I have a Micsig sto1004 but it doesnt seem to have trigger and record function, if thats the right name for it. 
 Not sure if this is the right Idea, but I think I need something that can trigger and record multiple events, hopefully control inputs and outputs to really see what caused the trip.
Thanks  a Bunch!

[ledtester]

An Analog Discovery would probably work for you - it has two 30 MHz scope channel inputs as well as GPIO pins. But you'd have to write some code to save the captures. Fortunately you can use a high level language like Python.

[inse]

The micsig does appear to be a decent tool at first glance.
Your description of the failure condition of your DUT is rather vague, you need to understand what you are looking for to select the right debug tool/approach.
If you are able to define a trigger condition for the failure you can set the scope to single shot and record whatever is necessary and analyze the data afterwards.
I know the limit test function available from the big players (Keysight, Tektronix, Lecroy…), which allows to define conditions and take action on violation.
But I don’t know whether this is available for budget scopes.
Possibly you might also need additional probes like a current probe or a differential probe to measure in different areas or isolation domains of your DUT.

[WattsThat]

Save your money, such a device doesn’t exsist at a budget you’d want to pay for. At a minimum, you’d need a decent 6 to 8 channel device with decent bandwidth along with three differential probes and three appropriately sized current transformers to monitor both current and voltage of each output phase.

Even when you have the ability to see what is happening with the output phases, chances are extremely high the problems will all be related to the IGBT’s and their driver circuits. Old dried out caps in the gate driver circuits result in devices turning on and off incorrectly which leads to the over current events when IGBT’s do not turn on and off within their required windows of operation. Eventually it will lead to a device failure for which the typical solution is to rebuild with all new IGBT’s and gate driver circuits.

Stop waiting for the drives to fail, either rebuild or replace and move on. Otherwise it will be death by a thousand cuts.

I don’t know where you are with the rebuild/replace calculation - but in my world anything at 100hp and less is an outright replacement as it doesn’t pay to put money into at-end-of-life older products.

[Rainwater]

I use a $30 battery powered scope like this and a voltage divider that steps 100V down to 1.
Most of the VFC's i troubleshoot are 208 three phase ~ 240 single phase 50/60hz input with 190v three phase delta configuration(high leg) 40-140hz output, made by fepetro. Very robust units and good as protecting their driver circuits

As wattsthat said, you will almost certainly have to replace the electrolytic capacitors, also be on the lookout for current sence resistors that are no longer in spec. The units give detailed error codes when pluging a serial monitor into them,  but the manufacturer denies this functionality exist and in their newest software revision requires a password to access it. The real starting point will be determined by what problems/alarm codes you are currently experiencing

[JohnXJohn]

The micsig does appear to be a decent tool at first glance.
Your description of the failure condition of your DUT is rather vague, you need to understand what you are looking for to select the right debug tool/approach.
If you are able to define a trigger condition for the failure you can set the scope to single shot and record whatever is necessary and analyze the data afterwards.
I know the limit test function available from the big players (Keysight, Tektronix, Lecroy…), which allows to define conditions and take action on violation.
But I don’t know whether this is available for budget scopes.
Possibly you might also need additional probes like a current probe or a differential probe to measure in different areas or isolation domains of your DUT.

These are really old vfds mounted on old equipment, seem to get a lot of overcurrent faults, not sure, could be the motors/hardware or vfds themselves... I'm learning as I go.  or  I already own three differential probes and current clamps, may be im missing something in the settings, but I haven't found a way to trigger and record for extended periods of time without me present.

[JohnXJohn]

An Analog Discovery would probably work for you - it has two 30 MHz scope channel inputs as well as GPIO pins. But you'd have to write some code to save the captures. Fortunately you can use a high level language like Python.

Super interesting ! Definitely looking into this, I dont know how to code but getting ready to start learning

[JohnXJohn]

Save your money, such a device doesn’t exsist at a budget you’d want to pay for. At a minimum, you’d need a decent 6 to 8 channel device with decent bandwidth along with three differential probes and three appropriately sized current transformers to monitor both current and voltage of each output phase.

Even when you have the ability to see what is happening with the output phases, chances are extremely high the problems will all be related to the IGBT’s and their driver circuits. Old dried out caps in the gate driver circuits result in devices turning on and off incorrectly which leads to the over current events when IGBT’s do not turn on and off within their required windows of operation. Eventually it will lead to a device failure for which the typical solution is to rebuild with all new IGBT’s and gate driver circuits.

Stop waiting for the drives to fail, either rebuild or replace and move on. Otherwise it will be death by a thousand cuts.

I don’t know where you are with the rebuild/replace calculation - but in my world anything at 100hp and less is an outright replacement as it doesn’t pay to put money into at-end-of-life older products.

when I said my management is dumb i meant it. These are 2 -3 hp motors, lots of them , with long cable spans. they do have additional units called "reflection absorbers"  I think. Im told they tried to install modern vfds, they were tripping a lot so they just have been rebuilding old ones 100 times.  No real engineers were involved. Like I said this is a learning project for me.

Do I need current transformers or current clamps will do?
is this the kind of scope you talking about?

https://www.amazon.com/Pico-Scope-4824A-Channel-Oscilloscope/dp/B0B9GGQGHB/ref=asc_df_B0B9GGQGHB/?tag=hyprod-20&linkCode=df0&hvadid=642202684271&hvpos=&hvnetw=g&hvrand=15291947500952572804&hvpone=&hvptwo=&hvqmt=&hvdev=c&hvdvcmdl=&hvlocint=&hvlocphy=9025143&hvtargid=pla-1948406590221&psc=1&mcid=44738e0ab9783404b0e1a91e024914a1

[JohnXJohn]

I posted this a separate thread, is it a silly idea?

https://www.eevblog.com/forum/beginners/vfd-testing-rig-advise/msg5296195/#msg5296195

[WattsThat]

Oh boy, it sounds like an improperly designed system right from the start. Low power drives and long cables don’t mix well at all. Over current faults are expected with that combination due to the output cable capacitance - every time an output transistor switches, it has to change the voltage level in the cable which requires current, potentially requiring more current than the drive can deliver, leaving little available current to actually produce torque in the motor. The filters were an attempt to limit the current but you can’t change the physics of what the cable does to the applied signal.

If you want to better understand what’s really going on, some system details would be helpful like motor nameplate info (really just volts, amps, power, speed and frequency), drive model numbers and then finally the cable specs and lengths.

What is typically done in these situations with replacements is to oversized the drive by 50-100% to have sufficient current to handle the cable losses. But, you do need to understand what is in play and the magnitude of the problem when selecting replacements.

Honestly, any effort to chase this problem with a scope or something similar is really a purely academic exercise as you’ve already found the source of the problems, you just didn’t understand the symptoms. What so many fail to understand is that the output from a VFD is not a sine wave, it’s not even close, it’s just a bunch of high voltage dc pulses which behave quite differently when pushed into a long cable.

[WattsThat]

I posted this a separate thread, is it a silly idea?

https://www.eevblog.com/forum/beginners/vfd-testing-rig-advise/msg5296195/#msg5296195

Not necessarily silly but it’s not a trivial task to design a high voltage PWM switch to load the dc motor.

If you want to test a VFD for functionality, you can use a direct across the line 3 phase ac motor as the load and over speeding the driving motor to create as much load as you need/want. The dol motor regens the excess energy right back to the line, easy peasy.

[Jeroen3]

You can extend the cable length between VFD and motor by adding a sine wave filter. However the drive needs to know how to deal with that because it changes some characteristics of the motors and adds limitation to the switching patterns of the igbt's.
Danfoss has some combinations that allow up to 500 meters of cable.

There generally is not that much to see by measureing the output of the vfd except perhaps missed pulses or stuck igbt's. Measureing it is also a reasonably high frequency, high voltage (for power stuff) endeavour that a simple power analyzer can't see without expensive options. But you don't need a scope for that to firgure it out.
So if you are going this path you would need a scope with at least 3 hv differential probes eg: DP20003 and some DC-capable current clamps, eg: fluke i30s and alike.
This alone would set you back ~€3000.

Don't forget there is also a front side of the controls here, a low voltage boards that receives the go signal and possible analog circuit.

Today you can get a lot more tech, so I would suggest re-evaluating the options. Replace the drives, move the drives, replace the motors?
Unless of course downtime is free, then don't bother.

[JohnXJohn]

You can extend the cable length between VFD and motor by adding a sine wave filter. However the drive needs to know how to deal with that because it changes some characteristics of the motors and adds limitation to the switching patterns of the igbt's.
Danfoss has some combinations that allow up to 500 meters of cable.

There generally is not that much to see by measureing the output of the vfd except perhaps missed pulses or stuck igbt's. Measureing it is also a reasonably high frequency, high voltage (for power stuff) endeavour that a simple power analyzer can't see without expensive options. But you don't need a scope for that to firgure it out.
So if you are going this path you would need a scope with at least 3 hv differential probes eg: DP20003 and some DC-capable current clamps, eg: fluke i30s and alike.
This alone would set you back ~€3000.

Don't forget there is also a front side of the controls here, a low voltage boards that receives the go signal and possible analog circuit.

Today you can get a lot more tech, so I would suggest re-evaluating the options. Replace the drives, move the drives, replace the motors?
Unless of course downtime is free, then don't bother.

Once poor management makes a mistake  they'll be married to it unless something drastic happens. My thought is to learn as much as I can out of this before moving on to the next job.  I already own Micsig DP10013  1300v rated, not enough?  inexpensive current clamps Hantek CC-65 AC/DC seem to work reasonably well. Im thinking to gather a passable kit to measure, understand whats happening, and test and showcase  solutions on a separate rig. They tried to replace drives, but im guessing they were not oversized as WattsThat suggested. Or may be they bolted them on old filters that didnt play nice together not sure..

[JohnXJohn]

I posted this a separate thread, is it a silly idea?

https://www.eevblog.com/forum/beginners/vfd-testing-rig-advise/msg5296195/#msg5296195

Not necessarily silly but it’s not a trivial task to design a high voltage PWM switch to load the dc motor.

If you want to test a VFD for functionality, you can use a direct across the line 3 phase ac motor as the load and over speeding the driving motor to create as much load as you need/want. The dol motor regens the excess energy right back to the line, easy peasy.

Sorry, could you kindly spell it out in dumber terms for thee nooob.   

[BeBuLamar]

I never need a scope to troubleshoot VFD. The VFD itself would tell me what's wrong. Even with long cable it shouldn't be tripping unless there is problem with the load or the motor.