Oscilliscope Purchase Help

Thread Starter

Darbstar

Joined Apr 21, 2016
6
Hi all. Newbie here, but love electronics. I work in the Heating and Cooling industry, and my point in this thread is based on the fact that we replace many Variable Frequency Drives. I have taken some apart, and feel confident in testing individual components, when off the PCB. The next step in my learning is to test them while integrated, which I would like to use a scope for. I am asking the forum what would be the specifications required for a new Oscilliscope purchase? I feel my goal is to be able to test the rectification circuit, the dc filtering circuit (IE caps), and the switching circuit (usually IGBTs). Below are my working conditions:
VFDS at 600 vac, 3 phase, so bus voltage at ~870 vdc.
IGBT's on VFD's fire between 5 khz and 15 khz (from my understanding). So would you guys have any suggestions as to what kind of Bandwidth, sampling rate, # channels, etc I would need? And would anyone like to suggest analog vs. digital vs. USB?
If the above is sufficient data for an opinion, feel free the disregard the following. As stated above, my goals will be to test the main three components that I feel comfortable to replace. If it were possible, I might like to be able to see the firing of the IGBT's at the gate, but I wouldn't be able to replace a single component that would affect the IGBT firing, mostly it's too complicated for me at this point, and really, if it's tied to the logic of the board, I wouldn't know what to do besides replace the board. Again, I feel limited at this time, just because I'm a newbie!
Further, would I be able to see any distortion, harmonics, current leakage through bearings? My hopes would be for a simple device that wouldn't be too expensive, but that would cover the minimal requirements are listed If I could get more for a small amount of cost increase, I would probably like to do so. I hope I have been clear, and I really appreciate any input you guys can give! Thanks for your time.
 

Thread Starter

Darbstar

Joined Apr 21, 2016
6
Hmm. Very thoughtful reply and gives much to think about. Well, what if we limited the scope to 208v 3 phase? That would reduce the voltage issue, but not the source grounding issue- requiring the purchase of the differential probes I guess. Besides the purchase of the differential probes, any further recommendations on scopes? And would reducing the working voltage reduce the cost significantly?
As a point of thought though, is a differential probe required? We rectify AC to DC, then we filter it (giving a DC BUS a +V reference and a -V), then we pulse the +V, which then goes back to -V on the BUS. I don't honestly know this, but could you not always use -V as your ground? Even when we pulse the IGBT's on the 'negative' portion of the 'sine wave' (really PWM square wave signal), we go 'positive to negative'. Wouldn't negative be an 'zero' VDC?
 

shteii01

Joined Feb 19, 2010
4,644
1. Forget USB. For professional/industrial use they are crap. It is pretty much purely hobby level stuff.
2. I have been looking into getting probes for used o-scope that I recently bought, and it seems good idea to get single attenuation probes instead of dual attenuation probes that have little switch to change attenuation from 1x to 10x, so that is on my list to get. While doing a bit of research I came across this article, I am going to attach it to this post. Talks about a lot of o-scope probe stuff, bandwidth, voltage rating, etc.
 

Attachments

#12

Joined Nov 30, 2010
18,224
VFDS at 600 vac, 3 phase, so bus voltage at ~870 vdc.
Well, what if we limited the scope to 208v 3 phase? That would reduce the voltage issue,
This worries me. It sounds like you are saying that buying a lower voltage rating on the scope will reduce the voltage on the motors.
Another thing that worries me is the fact that 440V will vaporize a screwdriver and 600 VAC is even scarier.
I'm an old fart, so I would use my 20 MHz analog scope as if I expect a motor drive has no need to get to even 1 MHz.
What I wouldn't do is stick a scope probe into 600 to 900 volts.
I'm really afraid you are going to be dead before the year is finished.
 
Here's a low end differential probe: https://www.quantasylum.com/content/Products/QA190.aspx but really not suitable.

Generally the deals are:
1. How far you can float ground. Even a battery scope has limits.
2. All measurements are usually relative to ground.
3. With A-B and Invert, you can do Pseudo-differential measurements, but both sides must have a ground reference.
4. Line operated switchmode power supplies are a real mess to troubleshoot. The primary side isn't referenced to ground.
So, even a 120/240 supply rectifies to somewhat less than 400 V and then goes from there.
5. I have used a scope to measure RF (1000 W into 50 ohms).

BTW: I've worked on/repaired 15 kV at 1A regulators at 100 kV at 0.1 A systems.
 

tindel

Joined Sep 16, 2012
936

shteii01

Joined Feb 19, 2010
4,644

Kermit2

Joined Feb 5, 2010
4,162
I've floated a scope with a 10x probe to look at 250 DC and for 120 AC.

I've shorted out circuits and destroyed a probe as well. If I had done that in a 480 volt circuit or with 600 + VDC, I probably would have been injured or worse

Power off to connect and disconnect. No rings on fingers. Only one hand in equipment. Try to keep the other in your pocket so you don't grab a metal pole or rest it on a metal chassis while working.
 

tindel

Joined Sep 16, 2012
936
I've floated a scope with a 10x probe to look at 250 DC and for 120 AC.
This is one of the DUMBEST things a person can do. The ground wire is connected what ever voltage you connect it to - which means that all of the the other ground (safety) points that protect the user are bypassed leaving the user and anyone that just happens to walk by in EXTREME DANGER. You should NEVER, EVER do this. If you are doing this then you're making your measurements completely wrong. I have yet to find an exception to the rule.

Not to mention that you're still capacitively coupled to earth... ruining your measurements to boot!

The best app note I've seen on proper differential voltage measurements.
http://in.tek.com/dl/3AW_19134_2_MR_Letter.pdf
 

Thread Starter

Darbstar

Joined Apr 21, 2016
6
This worries me. It sounds like you are saying that buying a lower voltage rating on the scope will reduce the voltage on the motors.
Another thing that worries me is the fact that 440V will vaporize a screwdriver and 600 VAC is even scarier.
I'm an old fart, so I would use my 20 MHz analog scope as if I expect a motor drive has no need to get to even 1 MHz.
What I wouldn't do is stick a scope probe into 600 to 900 volts.
I'm really afraid you are going to be dead before the year is finished.
HA. I appreciate the concern, but no, I don't think using a scope rated for 208v rms means I'm reducing the voltage supplied from a power supply to a motor. I think an oscilliscope is a test instrument. I meant that I would only limit myself to 208/3/60, if it meant not needing to buy a 5000$ hobby instrument.
Not to be terribly rude, but what are you doing with your screwdriver at 440 to blow it up? I've never had that happen to me, and definitely not at 600vac. At 15 KV I start to get anxious ;)
This worries me. It sounds like you are saying that buying a lower voltage rating on the scope will reduce the voltage on the motors.
Another thing that worries me is the fact that 440V will vaporize a screwdriver and 600 VAC is even scarier.
I'm an old fart, so I would use my 20 MHz analog scope as if I expect a motor drive has no need to get to even 1 MHz.
What I wouldn't do is stick a scope probe into 600 to 900 volts.
I'm really afraid you are going to be dead before the year is finished.
 

#12

Joined Nov 30, 2010
18,224
(safety) leaving the user and anyone that just happens to walk by in EXTREME DANGER.
Just thought I say again, you should have a person watching to re-start you if you need it, but be sure your safety person knows to not touch you while you're working.
 

Thread Starter

Darbstar

Joined Apr 21, 2016
6
Thanks Tindel and shteii01.
shteii01
Here's a low end differential probe: https://www.quantasylum.com/content/Products/QA190.aspx but really not suitable.

Generally the deals are:
1. How far you can float ground. Even a battery scope has limits.
2. All measurements are usually relative to ground.
3. With A-B and Invert, you can do Pseudo-differential measurements, but both sides must have a ground reference.
4. Line operated switchmode power supplies are a real mess to troubleshoot. The primary side isn't referenced to ground.
So, even a 120/240 supply rectifies to somewhat less than 400 V and then goes from there.
5. I have used a scope to measure RF (1000 W into 50 ohms).

BTW: I've worked on/repaired 15 kV at 1A regulators at 100 kV at 0.1 A systems.
Thanks a ton for the help. I appreciate the later comments from the other members regarding the Fluke meter. I don't know how you feel, but it seems there aren't too many low grade alternatives than what they are suggesting. This reply is specific for you KeepItSimple becuase I found your points very specific and helpful. And I agree, that line operated switches are a mess, but do you have any suggestions? Even if, just for arguments sake, you wanted to see the output signal of the igbt's- would you expect to need a high end device like the Fluke mentionned?
 

#12

Joined Nov 30, 2010
18,224
Not to be terribly rude, but what are you doing with your screwdriver at 440 to blow it up?
A 3 phase disconnect box didn't disconnect one of the phases.
I started to remove a power terminal screw from a 3 phase contactor and the end of the screwdriver disappeared.
Fortunately, I learned my trade working with 30KV so I had six layers of safety including 3 men on the crew who knew not to touch me while I was working. After the blast happened, we all just looked at each other to see if anybody got hurt.
Nobody even felt a tingle.

The layers:
Turn off the 3 phase disconnect box.
Close the contactor with an insulated tool to be sure we turned off the right disconnect box.
Use an insulated screwdriver.
Wear rubber soled shoes.
Place a dry rubber mat on the roof to keep my feet away from any rain water that might still be in the gravel.
Have a safety man watching from a safe distance.

I might have measured each phase to the case, but I didn't have a Cat II meter with me.
 

Thread Starter

Darbstar

Joined Apr 21, 2016
6
I might have measured each phase to the case, but I didn't have a Cat II meter with me.[/QUOTE]
I might have misunderstood this. You worked on 3 phase 440 3 phase system and didn't test the line side before working?
 

#12

Joined Nov 30, 2010
18,224
You worked on 3 phase 440 3 phase system and didn't test the line side before working?
That is exactly correct if you discount the idea that I closed the contactor to see if it would energize the motor.
What? Totally incredulous that I only created 6 layers of safety to compensate for the fact that there wasn't a proper meter available that day?

I'm 65 years old and I'm still alive, so I figure I must be doing something right.
Two of my layers failed. That's why I never have less than 3 layers in place.
In this case, I still had 3 active layers and 1 passive layer in place after two layers failed.
 
Last edited:

Lestraveled

Joined May 19, 2014
1,946
Other than a scope, I think you should look into an "isolator", like a Tektronix A6902. This is a add on front end to any scope that isolates you and the scope from very high voltages. (I have a A6902A). It also works well as a differential probe.


They are not being produced anymore but they are still available. The isolation spec is around 6,000 volts for both channels. It normally comes with a 300 volt probe and a 3000 volt probe. It sounds like this isolator would be more important than the kind of scope you get.
 

Attachments

#12

Joined Nov 30, 2010
18,224
The isolation spec is around 6,000 volts for both channels.
Now you're talking!
The mind boggle of this whole concept is, "How to get a machine designed to measure millivolts to measure the kind of power line that can turn it into a ball of fire."
I believe you presented the answer!
 

#12

Joined Nov 30, 2010
18,224
Here's the way I figure active safety layers:
Any one layer can fail, and you can bet it will fail within your lifetime, at which point your lifetime is over if you only have one layer.
Figuring a smart man can devise a safety layer that will fail once in a thousand activities, that's one in a thousand failure rate for each layer.
A second layer cuts your odds from one per thousand to one in a million. That's pretty good.
A third layer cuts your odds to one in a billion. That's my standard for 208/230 or less.
Being aware that a 440 volt mistake always causes damage or death, I add at least a 4th layer.
In this case, I added a fifth layer.
That is in addition to the passive layer...a person who can re-start your heart, disconnect you from the electricity, and/or call for help after the damage is done.
The passive layer is a significant layer for anything over 50 volts. One-twenty VAC rarely kills people, but it can knock you off a ladder.
 
Top