Products > Test Equipment
UNI-T UT-P06 300Mhz passive Probes. Are they any good?
battlecoder:
--- Quote from: KungFuJosh on October 06, 2023, 04:04:45 am ---There are other dealers that sell probemaster probes. welectron.com is one. There might be others.
Personally, I would wait until I could afford the better option. The master kit is also available with only 3 probes if you need to cut the cost.
--- End quote ---
I mean...maybe? I don't want to use bad probes with the most decent scope I've gotten for my lab, so maybe saving for really good ones makes sense.
In the meantime I got a quote from a local distributor. The Rigol probes would cost approximately $50 each, and if I get the exact model that I already have (of which they have stock, despite the probe being discontinued) I would only need 3, and the whole operation would fit neatly in my $150 budget.
Now, I also need a couple of cheap ~200Mhz probes for a FNIRSI portable scope that came with extremely garbage probes, so I'll probably get a couple of P2200 for it.
For what is worth, I found this other post in the forum: https://www.eevblog.com/forum/reviews/cheap-china-p6xxx-probes-good-bad/msg490775/#msg490775 . User "calin" apparently got the UNI-T P06 probes (Reply #12) which I opened this post for, and their review (#21) is that they "work a treat". I mean, I wish there were some numbers in that review, but at least they seem to have a positive experience with them.
battlecoder:
Btw, I tried to perform a more reasonable test of the Rigol RP3300A probe to determine whether biting the bullet and upgrading from the stock 150Mhz Keysight probes to a set of Rigol probes was worth it (which is so far the best plan that fits my budget).
Surprisingly, the Keysight probes (N2862B) despite being only rated for 150Mhz gave me a virtually identical risetime to the one I got using the RP3300A (350Mhz) probe (it was almost a matter of which one flickered more between 1.6 and 1.7ns). Performing a -3dB test also gave me very similar results (~230Mhz bandwidth) with both probes, so, measurement-wise I don't think I would be getting a lot from the upgrade.
Now, I'm surprised that the Keysight probes measured similarly to a higher bandwidth probe. I had previously estimated that they were limiting my bandwidth to approximately 175Mhz, but admittedly it could have been just shoddy probing in my previous attempt.
Of course the 350Mhz probe should be able to get a bit more of the higher >200Mhz frequency content without as much attenuation as the 150Mhz probe gets, and as such gives a slightly better square wave, but I'm not sure if it's enough to justify upgrading.
Here's a comparison of a 20Mhz square wave that should have a risetime of < 1nS (I only know it's < 1nS. I don't have a scope with that kind of bandwidth :-DD ).
The yellow trace (CH1) is the Keysight N2862B 150Mhz probe.
The orange trace (REF) is the Rigol RP3300A 350Mhz probe.
I used averaging to rule out any noise or ripple from the generator itself.
The trace with the Rigol probe is slightly flatter, but that's basically it. Unless I'm missing something here (I'm not an expert, so if there's anything else interesting going on here please let me know).
Performa01:
--- Quote from: battlecoder on October 07, 2023, 12:50:50 am ---Surprisingly, the Keysight probes (N2862B) despite being only rated for 150Mhz gave me a virtually identical risetime to the one I got using the RP3300A (350Mhz) probe (it was almost a matter of which one flickered more between 1.6 and 1.7ns). Performing a -3dB test also gave me very similar results (~230Mhz bandwidth) with both probes, so, measurement-wise I don't think I would be getting a lot from the upgrade.
--- End quote ---
Now you have finally got a vague idea about one simple fact that so many people tenaciously ignore: Any properly designed x10 probe will peform up to several 100 MHz in the industry standard test setup with 25 ohms source impedance, no matter what the specifications say. In practical use though, they all will only perform well at higher frequencies if connected to insensitive low impedance nodes. Furthermore, other than the LF compensation, there's also the HF compensation, which is user adjustable (only) on some more expensive probes.
Most probes are not user adjustable for the HF matching. They are matched to the scopes where they belong to. Consequently, it doesn't sound like a good plan to e.g. use a Rigol probe (without HF adjustment) on a Keysight oscilloscope. It might work okay, but it could also fail miserably. I have once tested a bunch of different probes on a 300 MHz Siglent scope. With the original (300 MHz) probes it reached a bandwidth of 450 MHz (yes, probes can expand the total system bandwidth beyond the bandwidth of the scope alone), but a humble 100 MHz Tektronix probe also came close to 400 MHz AFAIR. By contrast, a fancy 500 MHz Keysight probe performed less favourably, particularly was it worse than the 100 MHz Tektronix one.
The popular rise time test can be misleading. Firstly we cannot really calculate the scope bandwidth from the risetime as long as we don't know the exact frequency response (lowpass characteristic) of the frontend. Secondly risetime measurements can be flawed for many reasons, e.g. on scopes that do not measure the whole record but decimated data, like the Keysight Megazoom scopes do or also if the pulse form is not perfect but has a lot of overshoot - maybe due to improper HF matching of the probe...
Finally, probes in conjunction with the rather complex input network of a DSO do not usually have a clean lowpass characteristics. Somewhere in this forum I have demonstrated the frequency response of a cheap Siglent PP510 probe up to 1 GHz - and it showed a positive amplitude error at very high frequencies (>900 MHz). In conjunction with fast pulses (e.g. leo Bodnar) This can lead to false good looking rise time measurements, even though the probe isn't actually very useful because of the high ripple in the frequency response. Even just 1 dB amplitude error at certain frequencies will cause visible pulse distortions on the screen.
So wile pulse measurements are nice as an additional piece of information, probe performance can only be accurately measured with a levelled signal generator or AWG. Only if such a measurement with trustworthy equipment confirms that the existing probes are inadequate, I would ever start thinking about alternatives. The only really interesting specification is the tip capacitance. If you can find a probe that has significantly lower capacitance and is still perorming equally well in the industry standard bandwidth test, then it's worth buying. But I doubt that you will easily find one that meets these requirements.
Here's some food for thought (reply #56):
https://www.eevblog.com/forum/testgear/siglent-sds1104x-e-in-depth-review/msg1434665/#msg1434665
also here (reply #191):
https://www.eevblog.com/forum/testgear/siglent-sds1104x-e-in-depth-review/msg3290546/#msg3290546
KungFuJosh:
--- Quote from: Performa01 on October 07, 2023, 12:20:55 pm ---Here's some food for thought (reply #56):
https://www.eevblog.com/forum/testgear/siglent-sds1104x-e-in-depth-review/msg1434665/#msg1434665
also here (reply #191):
https://www.eevblog.com/forum/testgear/siglent-sds1104x-e-in-depth-review/msg3290546/#msg3290546
--- End quote ---
Thanks for sharing those links, I was too lazy to search for them. 😉
A couple other things worth pointing out with those Rigol probes are:
- Only rated to 300V in 10x mode (150V in 1x)
- Looks like they're larger and stiffer than better probes that use silicone like the probemaster probes (which are rated for 600V).
That's why I say don't upgrade unless it's actually an upgrade, and wait until you can afford to.
battlecoder:
--- Quote from: Performa01 on October 07, 2023, 12:20:55 pm ---Now you have finally got a vague idea about one simple fact that so many people tenaciously ignore: Any properly designed x10 probe will peform up to several 100 MHz in the industry standard test setup with 25 ohms source impedance, no matter what the specifications say. In practical use though, they all will only perform well at higher frequencies if connected to insensitive low impedance nodes. Furthermore, other than the LF compensation, there's also the HF compensation, which is user adjustable (only) on some more expensive probes.
Most probes are not user adjustable for the HF matching. They are matched to the scopes where they belong to. Consequently, it doesn't sound like a good plan to e.g. use a Rigol probe (without HF adjustment) on a Keysight oscilloscope. It might work okay, but it could also fail miserably. I have once tested a bunch of different probes on a 300 MHz Siglent scope. With the original (300 MHz) probes it reached a bandwidth of 450 MHz (yes, probes can expand the total system bandwidth beyond the bandwidth of the scope alone), but a humble 100 MHz Tektronix probe also came close to 400 MHz AFAIR. By contrast, a fancy 500 MHz Keysight probe performed less favourably, particularly was it worse than the 100 MHz Tektronix one.
--- End quote ---
Thanks for the detailed response. This is my first dive into the depths of probe performance, so all of the information you've shared and detailed comparisons are extremely useful to me.
I think the Texas 250 Mhz probes have two adjustable capacitors, so I assume they are for LF and HF adjustment. Would that be a better option if I were to pick a set of probes that can eventually work with other scopes?
--- Quote from: KungFuJosh on October 07, 2023, 03:15:09 pm ---A couple other things worth pointing out with those Rigol probes are:
- Only rated to 300V in 10x mode (150V in 1x)
- Looks like they're larger and stiffer than better probes that use silicone like the probemaster probes (which are rated for 600V).
That's why I say don't upgrade unless it's actually an upgrade, and wait until you can afford to.
--- End quote ---
You have absolutely no idea how tempted I am by the probemaster kit, so waiting is definitely one of the top options right now (especially after discovering that the included Keysight probes are performing a lot better than I expected).
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version