EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: tsmith35 on December 02, 2013, 01:51:15 am
-
Hi, EEVblog folks. I'm looking for some probes to test 480VAC drives (max output voltages around 1.4kV). I'm looking at the
Agilent Tektronix P5100A, but I'm not sure if that's the right choice. Scope would be either Agilent DSOX4304A or Rigol DS4024.
Anyone have experience with these combos?
-
You mean Tektronix P5100A? In any case I would not recommend using a passive probe in this application unless you have an isolated channel 'scope designed for the expected voltages. Get a high voltage differential probe. For AC drives 20MHz bandwidth is plenty and probes with suitable voltage ranges are available for reasonable prices.
-
You mean Tektronix P5100A? In any case I would not recommend using a passive probe in this application unless you have an isolated channel 'scope designed for the expected voltages. Get a high voltage differential probe. For AC drives 20MHz bandwidth is plenty and probes with suitable voltage ranges are available for reasonable prices.
Oops, Tek P5100A. Thanks for your help. I'll look into your recommendation and see what I can find. :)
-
To directly answer your question, yes, passive probes with BNC connectors can generally be used with all scopes with an input capacitance within its compensation range (7-30 pF for this probe), at least up to 200 MHz or so. Above that probes tend to be designed to match the non-linear input impedance of the scope front-end, so don't expect to get the full 500 MHz bandwidth out of this probe with a non-Tek scope.
Obviously this probe can only do ground-referenced measurements, and I would argue that the 1000 V CAT II rating is too low for what sounds like at least a 480 V CAT III circuit. Many HV differential probes are fed using a wall-wart or battery and will work with any scope. Again, pay attention to its overvoltage rating for both common-mode and differential-mode signals. Something like a Fluke DP120 might work.
-
If these are electronically controlled motor drives with controlled voltage output then I'd probably call them CAT I since their output shouldn't be subject to substantial mains overvoltage which is the reasons for the CAT II thru IV.
You should make yourself VERY aware of what the maximum transient voltage on the ouput of those drives might be. Inductive kick from a motor can produce substantial spikes.
In any case, you don't want to push the probe past it's limit, and you should be very aware of what the real maximum voltage transients might be.
I concur that generally a probe is a probe and will work with any scope unless it is drawing power from the scope and using additional connections besides the BNC. In those cases the probe is almost always manf and often model specific.
-
I know that 480V inverter-duty motors must be able to withstand 1488V, but transient peaks may reach nearly 2100V. Nothing to play around with. Probably a good idea to hand this off to a contractor with more specialized equipment. :)
-
OK, on the topic of scope probes for Agilent DSOX:
I have a DSOX-3034A and the probes that came with this scope are the most unwieldy pieces of garbage that I have ever encountered. Even my cheapo Tek TDS2024B probes are nicer and easier to use than this.
What's the best probe for this scopes? I am looking for something with 500MHz bandwidth minimum, and with a nice, small body like Tek's P6139A probe. I am currently using P6139A's for my DSOX but I have not compared high frequency waveform measurements with these probes vs. the crap agilent probes.
Any suggestions are welcome.
Also, the T2A adapter for these scopes that adapts TekVPI probes to the agilent connector is ridiculously priced, but nice. I am using a 100MHz Tek current probe with my DSOX and it is working nicely. Anyone interested in the schematic of this adapter?
-
Hi Pinkman,
I’m very interested in the schematics of N2744A T2A Probe Interface Adapter – I’ve a few Tek P6243 and it would be nice to be able to use them on my MSO-X 3034A.
Reproducing this adapter’s mechanical parts will surely be quite challenging, but considering the cost (400 Euros), it is worth trying…
Regards,
Pedro Couto
-
For the P6243/P6245 probes you would need a Tektronix 1103 power supply. They're expensive but so are the active probes you have. With the power supply you can use the probes on other scopes. :)
-
OK, on the topic of scope probes for Agilent DSOX:
I have a DSOX-3034A and the probes that came with this scope are the most unwieldy pieces of garbage that I have ever encountered. Even my cheapo Tek TDS2024B probes are nicer and easier to use than this.
What's the best probe for this scopes? I am looking for something with 500MHz bandwidth minimum, and with a nice, small body like Tek's P6139A probe. I am currently using P6139A's for my DSOX but I have not compared high frequency waveform measurements with these probes vs. the crap agilent probes.
Any suggestions are welcome.
I concur. I won't use the N2890A 500MHz probes that came with my X3104. The cable is stiff, the probe body with sprung hook is ridiculously long, and the probe tip is gigantic. Totally inappropriate for fine probing. They're still in the bags they came in.
I replaced all four with N2894A 700MHz probes. These are much better suited for fine probing with its 2.5mm tip. Another option, which is physically the same size, is the N2873A 500MHz model. I thought I would get a little more BW with the N2894A 700MHz probe, but it turns out not to make a difference. (The N2894A really only gets its extra BW with the better front end in the 4000X series). You can get boat load of different nice accessories for the N287xA/N2894A series, at of course, Agilent boat-sized prices.
I also have a set of Tek P6139A 500MHz probes which I used while waiting for the N2894A probes. I like the P6139A probes. They are small, have a great sprung hook made out of a flat piece of metal (so it doesn't slip off as easily), and a very flexible cable. Even the alligator clips are much better than Agilent, although the ground doughnut is annoying. But Tek sure knows (or did know) how to make probes.
Preferring the mechanics of the P6139A, I did compare the BW performance to the N2894A, but the N2894A had a flatter response. I didn't do any measurements on the original N2890A probes since there was no way I was going to use them.
Since you're only using your P6139A up to 300MHz, they're probably fine. I was testing up to 700MHz and I think most of the issues were at the high end (didn't save any screen shots - doh!).
Also, the T2A adapter for these scopes that adapts TekVPI probes to the agilent connector is ridiculously priced, but nice. I am using a 100MHz Tek current probe with my DSOX and it is working nicely. Anyone interested in the schematic of this adapter?
The T2A adapts TekProbe II to the Agilent's AutoProbe interface. TekVPI is their new proprietary connector and is a different animal. (The patent ran out on TekProbe II... gotta lock'em in with something else.) If you have a schematic, please send it along...
I have two of the T2A. They are convenient but not as automatic as I would like them to be. You still need to select via a menu which model Tek probe is plugged in. I would guess Agilent has not implemented or decoded Tek's I2C protocol which is part of the TekProbe interface.
-
Preferring the mechanics of the P6139A, I did compare the BW performance to the N2894A, but the N2894A had a flatter response. I didn't do any measurements on the original N2890A probes since there was no way I was going to use them.
The P6139A should have a flatter response on Tek scopes and the N2890A should have a flatter response on the DSOX, since both should be tuned to the front-end used in those particular scopes. Does Agilent state anything about the compatibility of the N2894A with DSOX-3000 scopes? I would expect them to use a common front-end design across all of the (non-rebadged) 1 Mohm inputs, but I haven't checked this. Tek used to use a different front-end in each scope series, which is why the TDS300, TDS400 and TDS500 series all needed different probes for optimum performance. The P6139A/B can be used with most if not all 500 MHz scopes with BNC inputs that Tek released since.
The T2A adapts TekProbe II to the Agilent's AutoProbe interface. TekVPI is their new proprietary connector and is a different animal. (The patent ran out on TekProbe II... gotta lock'em in with something else.) If you have a schematic, please send it along...
I have two of the T2A. They are convenient but not as automatic as I would like them to be. You still need to select via a menu which model Tek probe is plugged in. I would guess Agilent has not implemented or decoded Tek's I2C protocol which is part of the TekProbe interface.
I don't think I2C is used for anything important in TekProbe II probes, just for scaling factors and input impedance. They didn't use I2C for DSP correction filters back then. The same doesn't apply to TekVPI, so extracting optimum performance from (some) TekVPI probes would be much harder.
-
Hi Pinkman,
I’m very interested in the schematics of N2744A T2A Probe Interface Adapter – I’ve a few Tek P6243 and it would be nice to be able to use them on my MSO-X 3034A.
Reproducing this adapter’s mechanical parts will surely be quite challenging, but considering the cost (400 Euros), it is worth trying…
Regards,
Pedro Couto
I have family around for the next week and won't get a lot done, but I'll try to pull the thing apart next weekend and sketch a schematic.
Thanks to all for the comments regarding the probes. I really like the P6139A's and will look at the 700MHz Agilent probes mentioned.
-
The P6139A should have a flatter response on Tek scopes and the N2890A should have a flatter response on the DSOX, since both should be tuned to the front-end used in those particular scopes.
I re-did my testing and indeed the P6139A does have flatter response on a Tek scope (using a TDS3054) than either the N2890A or N2894A. And the converse is also true that the P6139A doesn't do quite as well on the 3000X as the Agilent probes.
Does Agilent state anything about the compatibility of the N2894A with DSOX-3000 scopes? I would expect them to use a common front-end design across all of the (non-rebadged) 1 Mohm inputs, but I haven't checked this.
They indicate the N2894A is compatible with the 3000X (as opposed to recommended) in the selection guide, doc #5968-8153EN.
But at the time I purchased them, there was very little information on the N2894A and I asked about it on Agilent's scope forum. The reply from Agilent was that I might see slightly more BW, but the 4000X has 700MHz on the 1Mohm input path which is why it would really benefit from the N2894A. Since it was only $10 more for the N2894A over the N2873A, and all the accessories were the same, I went for it.
I don't think I2C is used for anything important in TekProbe II probes, just for scaling factors and input impedance. They didn't use I2C for DSP correction filters back then. The same doesn't apply to TekVPI, so extracting optimum performance from (some) TekVPI probes would be much harder.
Right, but for almost US$500 per T2A, I think Agilent should have done a little more reverse engineering, or even obtained the right to use Tek's protocol, so that the scale and impedance would get set up correctly. Looking at a T2A from the outside, it doesn't even appear that the I2C pads are even connected, so no FW upgrade will enable it in the future. Oh well.
-
Tek may very well have been unwilling to share / license the I2C protocol; they have no interest in allowing their probes to be used with scopes from other vendors. Which is part of the reason why they first made the external TekProbe power supply insanely expensive and then discontinued it. They'd rather sell you a scope to go with that probe. They can't very well protect a few pads with DC voltages, but the I2C protocol may either be protected by patents or Agilent may have been unwilling to reverse-engineer it.