EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: Lightages on January 21, 2015, 03:31:44 am
-
I kind of hijacked the Review Summary video #703 with a discussion of the quality of probes supplied with the DS1054Z.
The original post I made with an image of an example of how the waveform appeared differently with the RP2200 probes that came with my DS1054Z as compared to the RP3300 probes that came with my DS1052E.
https://www.eevblog.com/forum/blog/eevblog-703-rigol-ds1054z-oscilloscope-review-summary/msg589804/#msg589804 (https://www.eevblog.com/forum/blog/eevblog-703-rigol-ds1054z-oscilloscope-review-summary/msg589804/#msg589804)
Both scopes were modified to 100MHz bandwidth. The image is linked below.
It was suggested that I needed to change the termination load, measure without the ground leads, maybe the ground lead length difference caused the appearance change, or maybe my signal was not good enough.....
Well I have measured with two scopes, all the available channels, with the cable from the function generator open ended alligator clips, 50 ohm load at the generator, at the end of a BNC terminated cable, with the spring clip ground point on the probes, directly into the function generator, and all the different arrangements that are possible. Yes the observed wave form changed, but there is always the same relative difference of the appearance of the wave form between the RP2200 and the RP3300.
I am either doing something wrong, or there is a visible degradation of the input bandwidth of the scope with the RP2200 model probes.
So two questions:
#1. Are my attempts futile and the apparent change in the displayed waveform is my lack of knowledge. This is a high probability.
#2. Does the performance of the RP2200 actually degrade the total system performance of the DS1054Z ?
Bonus question: If #2 is an affirmative, is this actually a smart move by Rigol to limit the system bandwidth of the scope so that the under sampling with 3 or 4 channels enabled is less of a problem?
Discussion?
-
Wow. Are we saying the RP3300 probes (DS1052) are better than the PR2200 probes (DS1054)?
If so, that is a little surprising.
On the other hand, Rigol's specifications say that RP2200 are good up to 150 MHz, while RP3300 are good up to 350 MHz. So the RP3300 should be better. If I happen to have a pair of the RP3300 I guess I'm keeping them...
-
My video on this is rendering now.
Here are some screen shots of the two probes, as measured across a 50ohm load on a 100MHz DS1054Z with proper coaxial probe top connector.
-
Wow. Are we saying the RP3300 probes (DS1052) are better than the PR2200 probes (DS1054)?
If so, that is a little surprising.
On the other hand, Rigol's specifications say that RP2200 are good up to 150 MHz, while RP3300 are good up to 350 MHz. So the RP3300 should be better. If I happen to have a pair of the RP3300 I guess I'm keeping them...
I can only guess that Rigol ran out of RP2200 probes when the demand for the DS1052E went up and they had no choice but to ship them with the RP3300 probes to keep shipments moving. They are better probes by spec at least so yes, keep them!
-
OK Dave, I will wait and see you video. But it is very surprising to me that I have a real difference in the waveform, and with proper termination. The termination should not matter as long as the impedance of the probes are not significantly low enough to effect the waveform. I did one of my comparisons across a properly terminated, and not terminated cable, and the wave form always changed between the different probes.
-
I can only guess that Rigol ran out of RP2200 probes when the demand for the DS1052E went up and they had no choice but to ship them with the RP3300 probes to keep shipments moving. They are better probes by spec at least so yes, keep them!
Apparently I don't have the RP3300 probes after all, so I guess I can stop dreaming... :(
-
The termination should not matter as long as the impedance of the probes are not significantly low enough to effect the waveform.
Of course it will be. The 16pF input capacitance is about 100ohms @ 100MHz.
-
Hmmm, I know I am treading on deep water here with my lack of experience with some of these things. The RP2200 specs are 15pF and I measure 14.7pF with my DER EE DE-5000 at 100kHz. At the moment I cannot find my data sheet for the RP3300 probes but I measure 17.5pF at 100kHz??!! Shouldn't the higher capacitance provide a higher load and therefore the appearance of the waveform would look like lower bandwidth?
EDIT
Actually I just measured with both probes' compensation adjusted on the same scope and input and their capacitance is pretty much equal shows the RP2200 with lower capacitance than the RP3300 at 100kHz.
-
RP3300 spec is 16pF +/- 5pF
RP2200 spec is 17pF +/- 5pF
-
Sorry I edited my post after you posted. The exact measurement I get is 17.6pF for the RP3300, and 14.9pF for the RP2200. I know 100kHz is a far cry from 100MHz, but are their caps really that dependent on frequency between the two models? This after compensating each probe and immediately measuring the capacitance right after.
Edit:
Not equal sorry
-
Sorry I edited my post after you posted. The exact measurement I get is 17.6pF for the RP3300, and 14.9pF for the RP2200. I know 100kHz is a far cry from 100MHz, but are their caps really that dependent on frequency between the two models? This after compensating each probe and immediately measuring the capacitance right after.
There aren't caps in there, it's the capacitance of the construction of the tip up until the input resistor. It can be quite variable as the spec indicates and yes, will change with frequency.
-
There aren't caps in there, it's the capacitance of the construction of the tip up until the input resistor. It can be quite variable as the spec indicates and yes, will change with frequency.
OK I can understand that caps capacitance ratings can change their value with frequency, but the variable cap for adjusting the compensation is not a cap? :-//
So now that you have asserted that the RP2200 probes have a higher capacitance than the RP3300 probes at 100MHz, I will await your video to see the actual bandwidth measurement for the total system with the DS1054Z to see if the full 100MHz bandwidth is actually available with the supplied probes. ;)
-
I think I need to say that my previous experience in electronics was basically limited to 10MHz or so and I have little experience above 25MHz or so. Please excuse my ignorance on this higher realm but I think it is an opportunity to learn about this "higher" frequency for everyone.
-
OK I can understand that caps capacitance ratings can change their value with frequency, but the variable cap for adjusting the compensation is not a cap? :-//
That is across the 9M input divider resistor.
https://www.dfad.com.au/links/THE%20SECRET%20WORLD%20OF%20PROBES%20OCt09.pdf (https://www.dfad.com.au/links/THE%20SECRET%20WORLD%20OF%20PROBES%20OCt09.pdf)
So now that you have asserted that the RP2200 probes have a higher capacitance than the RP3300 probes at 100MHz, I will await your video to see the actual bandwidth measurement for the total system with the DS1054Z to see if the full 100MHz bandwidth is actually available with the supplied probes. ;)
I didn't do that in this video, just tried to verify if there was a pulse response difference difference between two probes on the 100MHz scope. There isn't really any difference.
-
how about the pulse response of the scope itself, i.e. terminate the scope intput with 50ohms and connect the sig gen directly?
-
Ahhhh, too bad that you didn't measure the actual response. :(
Yes I know where the cap is situated in the probe circuit. It doesn't mean there isn't a cap! >:D
Oh well, I will await an answer for the system response with the RP2200 and maybe vs the RP3300. Maybe someone with better experience than I have and the appropriate equipment can show this.
-
So the RP3300 are rated at a nominal 16pF and the RP2200 at 17pF. This implies that from the assertion that 16pF is 100ohms that 17pF would be maybe 94ohms. Does this account for the difference I see in the waveform I am measuring? It looks like it might be so from the visual difference I see. I really wish I had something to do this quantitative measurement. >:(
-
i made a few screenshots with a tek dpo3000 scope. The BW is only 300MHz, but the difference in rise time is still measurable.
I probed the terminated output of a 50-ohm square wave source (a VFO for a shortwave RX passive mixer) with the stock 500MHz 8pF probe, the vfo connected directly to the scope using the internal 50ohm termination and finally, using an HP 150MHz 15pf probe.
The input capacitance of the scope is 11.5pF.
The fastest rise time is obtained when the source is coupled directly to the scope. 500MHz probe follows, and the slowest is the 150MHz probe.
The rise time of the source itself is well under 700ps.
obviously, the combined BW of the probe+scope is lower than the BW of the scope alone, eventhough the 8pf tip capacitance should in theory load the source output less than the 11.5pf on the input of the scope.
-
Oh well, I will await an answer for the system response with the RP2200 and maybe vs the RP3300. Maybe someone with better experience than I have and the appropriate equipment can show this.
It doesn't matter what the actual measured response is, the point is that there was essentially little to no measurable difference between the RP2200 and RP3300 when used on the DS1054Z, and I thought that was what was under question here?
And there is no reason to believe that the RP2200 does not meet its claimed 150MHz bandwidth (3dB down), I would be stunned if it didn't.
-
So the RP3300 are rated at a nominal 16pF and the RP200 at 17pF. This implies that from the assertion that 16pF is 100ohms that 17pF would be maybe 94ohms. Does this account for the difference I see in the waveform I am measuring?
No, unlikley that it would, as evidenced by my measurement. Unless you have two probes that are vastly out on on the two edges of the spec or something.
I suspect there is something amiss with your test method.
-
Looks like at least one of the probes might not be compensated correctly. Did you make sure to adjust both probe compensation screws to get the proper flat top with no overshoot or undershoot?
-
One of the questions was:
#2. Does the performance of the RP2200 actually degrade the total system performance of the DS1054Z ?
This was in response to the video review where you stated that the probes would not limit the performance of the scope. When I saw a difference I wondered based on what I had seen with my probes compared to the RP3300s
And yes to the last poster, both of the probes were adjusted as perfectly as I could make them. This was done each time I connected and immediately before I made any test.
-
what if you measure the capacitance of the 2 probes at 1Mhz and 10Mhz ? is there such an option to measure?
if you compensate a probe @ 1kHz, should you compensate for it @ higher frequencies too? or rather, would the compensation be required in a different manner at higher frequencies? so should there be 2 compensation values for 1 lower (below 10khz) and 1 higher (above 1 Mhz)?
-
One of the questions was:
#2. Does the performance of the RP2200 actually degrade the total system performance of the DS1054Z ?
What self-respecting manufacturer would supply a probe that might?
They would be quickly found out and shamed or should be.
-
One of the questions was:
#2. Does the performance of the RP2200 actually degrade the total system performance of the DS1054Z ?
This was in response to the video review where you stated that the probes would not limit the performance of the scope. When I saw a difference I wondered based on what I had seen with my probes compared to the RP3300s
I mentioned this in my video, but did not actually measure it, and there is a reason for that. The answer is yes, in theory it can, because when you cascade two systems with a given bandwidth, the total bandwidth is always lower, to the tune of the inverse root sum of the inverse squares. Even 1GHz probes would lower the bandwidth a bit. But it's not that simple in passive probing systems like this, but I gave that example anyway.
But the reason I did not measure it is because it has been standard practice for eons for manufacturers to supply the same rates bandwidth probes as the scope, or a bit higher. And Rigol does this, so it's a not a problem. So it was sufficient to just to see if there was any difference between the two, and there isn't, hence the 150MHz probes are just fine. Unless of course you think the 350MHz probes aren't up to snuff either.
-
Lightages - I assume you measured the probe capacitance with the probe connected to the scope input. Was it the same input channel? Was the signal from the capacitance meter overloading the input by any chance?
-
For those who really really need to see it ;D
EEVblog #707 - Rigol Oscilloscope Probe Performance (https://www.youtube.com/watch?v=pHbooMWS0bU#ws)
-
what if you measure the capacitance of the 2 probes at 1Mhz and 10Mhz ? is there such an option to measure?
if you compensate a probe @ 1kHz, should you compensate for it @ higher frequencies too? or rather, would the compensation be required in a different manner at higher frequencies? so should there be 2 compensation values for 1 lower (below 10khz) and 1 higher (above 1 Mhz)?
Slow probes do not generally have high frequency compensation pots. My TEXAS 250MHz ones has 1 low and 2 high frequency pots.
-
First things first: A big :-+ to Dave for doing this video. I certainly believe your results. I guess the differences I see are accounted for by the bandwidth rating differences.
My RP3300 probes do not look like Dave's. Mine have an identical appearance to the RP2200 and have the compensation adjust on the probes and only have the BNC connector at the scope end, no box. I believe Dave's are now called RP3300A and mine are the old version.
nfmax: I measured with the probes connected and disconnected just now and the RP2200 always show a lower capacitance but as we know capacitance can change with frequency.
I also remembered as my head hit the pillow that both probes were connected at the same time for the image I posted so there is no difference in the loading on the signal between the two probes. This means the difference I see on the waveform in that image is solely due to the probes and not a loading issue.
In conclusion the differences I see are real, and due to the bandwidth differences of the probes.
I hope someone feels like doing a bandwidth test of common probes supplied with the lower end scopes. It would be interesting to see the real performance.
P.S. Oh yes, I will continue to use the RP3300 for single and dual channel use for obvious reasons.
-
My experience is that the RP2200's is that they were not at all in the same league when compared to RP3300/RP1300's.
-
My main problem with Rigol's probes right now: I received my scope last week and one of the probe tips came off when I removed the hook attachment for the second or third time.
-
My RP3300 probes do not look like Dave's. Mine have an identical appearance to the RP2200 and have the compensation adjust on the probes and only have the BNC connector at the scope end, no box. I believe Dave's are now called RP3300A and mine are the old version.
Nope, RP3300A are x10 only and look completely different.
-
I wonder if Rigol is rushing out the probes to meet demand and the quality control is lacking for this reason.
So based on the bandwidth calculation Dave showed in the video, the RP2200 probes with the scope give theoretical system bandwidth of 83.2MHz. Yes he corrected his assertion and qualified it, but that is a limit on the bandwidth. The RP3300 probes give a calculated system bandwidth of 96.2MHz. That is much closer the to rated bandwidth of the scope alone.
-
For those who really really need to see it ;D
EEVblog #707 - Rigol Oscilloscope Probe Performance
I noticed the probe settings on all of the scopes were set to 1X. Would this have any effect on the results?
-
No, the probe setting only changes the indicated values on the screen. It does not affect the acquired signal at all.
-
My main problem with Rigol's probes right now: I received my scope last week and one of the probe tips came off when I removed the hook attachment for the second or third time.
Can second that. The supplied Rigol probes are of mediocre quality mechanically. After half a year of occasionally use, the contact of the gripper to the probe tip of one of my RP3300 probes became loose and horrible unreliable. Rigol must make their money somewhere :D But I didn't see the overshoot Lightages showed up in his OP. More the opposite. For comparison the RP33000 compared to a TesTec HF212. Both were connected at the same time to a T-splitter and a 50 ohm feed through terminator. So total source resistance is 25 ohm loaded by 29.5 pF, which gives a LP roll off at 215 MHz. Channel 1 (yellow) = RP3300, Channel 2 (blue) = TesTec HF212. B.t.w. the HF212 has two HF adjusters.
-
To be clear, it is not the probe that is ringing or overshooting, it is the actual waveform and the different bandwidths of the different probes show different traces of the same waveform.
-
Ok, then it will be difficult to judge the probes right and adjust them right. What I am missing in this thread: Did you adjust the HF adjustment of the RP3300 as well (the trimmer in the probe itself)?
With a proper square wave I get an almost perfect Gaussian response with a rise time of 3.5 ns on my DS2102 at 100 MHz bandwidth. And with almost no ripples and no overshoot at all.
-
Yes, I adjusted the probes correctly on the compensation test point on the scope. The waveform as generated contains the ringing, not the probes. The ringing is the high frequency content I am inspecting and seeing the difference with the different probes.
Coincidentally, 4 new Chinese 300MHz probes arrived today. I will give them a test now.
-
Yes, I adjusted the probes correctly on the compensation test point on the scope.
The RP3300 has 2 adjustment points: One at the BNC connector (LF) and one at the probe tip (HF). The one at the probe tip is hidden and sealed with a small grey rubber plug. That rubber plug is almost invisible and you have to remove it to get to the HF adjustment (at least at the RP3300's I have).
-
The RP3300 has 2 adjustment points: One at the BNC connector (LF) and one at the probe tip (HF). The one at the probe tip is hidden and sealed with a small grey rubber plug. That rubber plug is almost invisible and you have to remove it to get to the HF adjustment (at least at the RP3300's I have).
Wow, you are right. It just looks like part of the case molding on mine, but sure enough, remove it and there is the trimmer.
And sure enough there it is in the manual too ;D
-
I already said, before in this thread, that my RP3300 only have the compensation trimmer on the probe. They are not the ones with the small box at the BNC connector.
I just tested the "P6300" probes I purchased on ebay. They are built OK, but they show almost exactly the same appearance of the waveform as the Rigol RP2200, a little higher peaks on the ringing, but nothing to say they are really better in bandwidth than the RP2200. I paid $15 each for them, and for the price they aren't bad, but certainly not 300MHz by my estimation.
For everyone interested, I consider the question on the probes a closed matter. From the video that Dave did and the differences seen between his probes, what I see seems to match an expectation of a drop from 96.2MHz to 84.2MHz bandwidth, as calculated. It also does answer my question about the probes limiting the bandwidth of the scope. Yes they do theoretically. How much I can only guess by the calculation but it appears somewhat logical from the appearance of the waveforms.
-
Please note that there are RP3300's (grey) and the newer RP3300A's (black). They are different. The non-A's are 16pF and 1/10x and the A's are 13.5pF and 10x only.
-
You're right, looked up the manual: http://www.rigol.com/download/Oversea/DS/User_guide/RP3300A_UserGuide_EN.pdf (http://www.rigol.com/download/Oversea/DS/User_guide/RP3300A_UserGuide_EN.pdf) And those have no HF trimmer :o
-
Just read some of those replies on Dave's youtube channel for the video.... Everyone thinks I am seeing things, including Dave! :-//
I have concluded that I am actually seeing the difference in the probes. I am not probing incorrectly. I have tried all methods suggested and the results are the same. |O
-
DS2072A says scope input is 16pF. I measured almost 18.5pF @70MHz.
-
My DS2000 came with the newer RP3300A probes with only compensation at the BNC connector and they are 10:1 fixed attenuation, maybe Lightages probes are really the RP3300A.
Here are the specs
Input capacitance 13pF +- 3pF
Bandwidth 300MHz, the non A is 350MHz
No rise time in the spec.
(https://www.eevblog.com/forum/testgear/scope-probes-as-supplied-by-rigol-for-the-low-end-scopes-a-discussion/?action=dlattach;attach=131467;image)
-
DS2072A says scope input is 16pF. I measured almost 18.5pF @70MHz.
16pf +- 5pf
And that is the probe input, don't know about the scope input if it adds more.
Edit: and not a big deal, but I'm not sure why you are rounding up. 18.447pf rounded to one decimal point should be 18.4pf or 18.45pf if using 2 decimal places, yeah picky but we all know how you feel about Rigol :)
Anyways, well within specs.
-
For those who really really need to see it ;D
EEVblog #707 - Rigol Oscilloscope Probe Performance (https://www.youtube.com/watch?v=pHbooMWS0bU#ws)
Question is about the rise time of the function generator source. If the rise time of the source is limited (which I see in the AWG datasheet as 8ns) then of course the performance of the two probes will be identical - they have 2.3ns and 0.9ns respectively. You have to test them with rise time that is close to the rise time of the scope (which is 3.5ns) in order to see a difference.
The test with 8ns rise time pulse under these conditions is meaningless.?
-
My DS2072A (mid 2014) came with RP3300A's. I measured the input (probe-tip) cap and it is 13.8pF which is nice within 13+/-3pF. The RP3300A's are smaller and only 10x, what I like. They have a nice rubberized feel and a thinner hook. The ground clip is very stiff and the clip-on connection with the probe is exposed. They are also only 300V. Don't expect them to be very durable. I bought one extra quality probe for daily use and keep the RP3300A's for 2nd channel and external trigger.
-
For those who really really need to see it ;D
EEVblog #707 - Rigol Oscilloscope Probe Performance (https://www.youtube.com/watch?v=pHbooMWS0bU#ws)
Question is about the rise time of the function generator source. If the rise time of the source is limited (which I see in the AWG datasheet as 8ns) then of course the performance of the two probes will be identical - they have 2.3ns and 0.9ns respectively. You have to test them with rise time that is close to the rise time of the scope (which is 3.5ns) in order to see a difference.
The test with 8ns rise time pulse under these conditions is meaningless.?
also, nobody really cared to hook their funct generators directly to the scope bnc, terminate everything properly and then assess the contribution of the probe vs what the scope shows w/o it.
A simple circuit that can give <1ns rise time with low jitter could be a x-tal oscillator built around 74AC04 dead bug style on copper foil with ample decoupling of the VDD pin to ground. The spare inverters can be paralleled together to buffer the oscillator and provide really low output resistance.
-
Question is about the rise time of the function generator source. If the rise time of the source is limited (which I see in the AWG datasheet as 8ns) then of course the performance of the two probes will be identical - they have 2.3ns and 0.9ns respectively. You have to test them with rise time that is close to the rise time of the scope (which is 3.5ns) in order to see a difference.
The test with 8ns rise time pulse under these conditions is meaningless.?
No, it's not. The whole idea was to try and verify the gross irregularity that Lightages saw.
(https://www.eevblog.com/forum/testgear/scope-probes-as-supplied-by-rigol-for-the-low-end-scopes-a-discussion/?action=dlattach;attach=131203;image)
Note the 50ns/div timebase.
So 8ns risetime was plenty to confirm the problem. In fact it's about 3ns or so, take a look at my screen shots.
-
I think Lightages signal source might be at fault? That overshoot seems to be too low frequency to be directly caused by the scope/probes.
My main problem with Rigol's probes right now: I received my scope last week and one of the probe tips came off when I removed the hook attachment for the second or third time.
Wow, never had that before and i've owned 6 of the RP2200 probes. The ones that were supplied with the DS1054Z do seem to be slightly lower quality than the ones that came with a DS1052E - the plastic casing just feels cheaper and the hooks are a little more rounded so they fall off (unclip) easier (though, the older ones weren't much better). I can't understand why they can't make the probe hooks as sharp as the cheap P6100 chinese probes you can get from ebay/etc.
-
I will say it again. The overshoot is not the probes. The waveform actually has this ringing. The loading of the source by the probes have nothing to do with the difference because they are both connected at the same time.
I have a Siglent SDG1025 almost in my hands. I will do another test with a more proper source properly loaded and we will see if I can detect any difference.
-
DS2072A says scope input is 16pF. I measured almost 18.5pF @70MHz.
What instrument did you use for this test?
-
I'm not sure why you are rounding up. 18.447pf rounded to one decimal point should be 18.4pf or 18.45pf if using 2 decimal places
First, i said "almost 18.5". Second, i am not a programmer and i live in a real world. Try paying for you morning coffee using your rules and see how much that guy/girl at the cash counter agree with you.
, yeah picky but we all know how you feel about Rigol
Oh thank you for the recognition of my humble efforts. Yes i am not OK with companies that have no idea what they are doing.
-
I will say it again. The overshoot is not the probes. The waveform actually has this ringing. The loading of the source by the probes have nothing to do with the difference because they are both connected at the same time.
I have a Siglent SDG1025 almost in my hands. I will do another test with a more proper source properly loaded and we will see if I can detect any difference.
Check this thread for a REAL comparison:https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg590683/#msg590683 (https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg590683/#msg590683)
Siglent SDG1025 & SDS1072CML.
-
I will say it again. The overshoot is not the probes. The waveform actually has this ringing. The loading of the source by the probes have nothing to do with the difference because they are both connected at the same time.
I have a Siglent SDG1025 almost in my hands. I will do another test with a more proper source properly loaded and we will see if I can detect any difference.
Check this thread for a REAL comparison:https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg590683/#msg590683 (https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg590683/#msg590683)
Siglent SDG1025 & SDS1072CML.
Very entertaining thread. https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg592378/#msg592378 (https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg592378/#msg592378)
-
I will say it again. The overshoot is not the probes. The waveform actually has this ringing. The loading of the source by the probes have nothing to do with the difference because they are both connected at the same time.
I have a Siglent SDG1025 almost in my hands. I will do another test with a more proper source properly loaded and we will see if I can detect any difference.
Check this thread for a REAL comparison:https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg590683/#msg590683 (https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg590683/#msg590683)
Siglent SDG1025 & SDS1072CML.
Very entertaining thread. https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg592378/#msg592378 (https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg592378/#msg592378)
And it's getting better: https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg592751/#msg592751 (https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg592751/#msg592751)
-
I will say it again. The overshoot is not the probes. The waveform actually has this ringing. The loading of the source by the probes have nothing to do with the difference because they are both connected at the same time.
I have a Siglent SDG1025 almost in my hands. I will do another test with a more proper source properly loaded and we will see if I can detect any difference.
Check this thread for a REAL comparison:https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg590683/#msg590683 (https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg590683/#msg590683)
Siglent SDG1025 & SDS1072CML.
Comparison of using a BNC instead of probes?
But the subject at hand is all about using probes, so not sure what you mean by a real comparison.
-
I will say it again. The overshoot is not the probes. The waveform actually has this ringing. The loading of the source by the probes have nothing to do with the difference because they are both connected at the same time.
I have a Siglent SDG1025 almost in my hands. I will do another test with a more proper source properly loaded and we will see if I can detect any difference.
Check this thread for a REAL comparison:https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg590683/#msg590683 (https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg590683/#msg590683)
Siglent SDG1025 & SDS1072CML.
Comparison of using a BNC instead of probes?
But the subject at hand is all about using probes, so not sure what you mean by a real comparison.
Ok, poorly put. :palm:
But the fact of the matter is probes alone are not appropriate for accurate representaion of a waveform with a fast leading edge. Impedance matching is also very important.
The "show us your square wave" thread has many examples of connection (some on purpose) that display similar ringing to Lightages.
@Lightages
Are you using the SDG1025 with High Z or 50 Ohm output setting?
Probe to BNC adapter?
-
It was described before but people seem to not be understanding the situation. Perhaps this is my fault so here it is again, in excruciating detail.
I am using an analog function generator that has a dedicated TTL output, not rated at 50ohm. I have a 50ohm terminator on one side of the T and a BNC coax on the other end. This gives a ringing square wave as the output. It is not meant to be a proper square wave. I was looking for the fastest rise time I could get from the output. It was just by accident that I arrived at this setup.
I am still waiting for my SDG1025 to arrive.
I have a DS1052E upgraded to 100MHz that came with RP3300 probes, old versions, that do not have a compensation adjustment at the BNC.
I have a DS1054Z upgraded to 100MHz that came with RP2200 probes.
I was casually comparing the performance of the two systems. When I started looking at the waveforms that were displayed on both scopes while probing the same signal I noticed that the DS1052E seemed to show more detail on the waveform than the DS1054Z. So thinking that perhaps the upgrade had not worked on the DS1054Z I started to investigate. After some poking around I swapped the probes from one scope to the next (after compensating of course) and the difference I observed moved with the probes.
Both RP3300 probes showed more detail on the waveform than the 4 RP2200 probes I have, and on either scope and on any of the channels. One RP2200 and one RP3300 probe were always connected at the same time to the signal so any difference in loading of the signal from the different probe model is a moot point. I have tried probing the source BNC directly with ground spring clips instead of the alligator clip leads, the same on the end of the original arrangement, at the T adapter, etc....
In each and every case, the same difference in the intentionally ringing waveform appeared. The RP3300 probes show more detail and higher peaks on the ringing than the RP2200, regardless of scope, channel and test method.
When calculated according to the formula presented on Dave's video, a theoretical 100MHz bandwidth scope with theoretical 150MHZ probe will have system bandwidth of around 84.2MHz. The same scope with a 350MHz probe will have a bandwidth of 96.2MHZ. With this in mind, it is a small leap to justify the difference I see with my different probes. It also shows that theoretically the probes limit the performance of the scope, whether it is normal practice to supply the same bandwidth probes for a scope or not.
It has now been demonstrated, at least in one person's case, that the DS1054Z upgraded to 100MHz actually seems to have a bandwidth of between 130MHz and 150MHz. With this in mind the difference I see in the waveform is even easier to rationalize. It is because this affects what effect the probes have on the bandwidth. It is a wider spread in the system bandwidth, namely 106MHZ for the RP2200 vs 138MHz for the RP3300. This is assuming that the scope has 150MHz bandwidth.
So as far as I am concerned the difference that I see is easily explained by these last theoretical calculations and it answers my question about whether or not the probes as supplied affect the overall scope's performance. If the 100MHz bandwidth of the scope is assumed, or lower, then the probes will bring the overall system performance down below the rating of the scope. In the case of the 50MHz bandwidth model, down to a theoretical 44.7MHz. If the bandwidth of the 100MHz model scope is really closer to the 150MHz suggested by another person then the RP2200 probes bring the system bandwidth down to 106MHz and the probes do not affect the rated performance of the scope as a system with the probes.
Another conclusion can be reached to I think. If you do not want your scope measurement system to be degraded by the probes, they should have at least 3.5 times the bandwidth of the scope. At the very least you should be aware of the effects of the bandwidth of the probes in your overall system performance. The last sentence holds true for any system that all components need to be taken into account of course but I just wanted to complete the logical conclusion.
-
....................................
But the fact of the matter is probes alone are not appropriate for accurate representaion of a waveform with a fast leading edge. Impedance matching is also very important.
The "show us your square wave" thread has many examples of connection (some on purpose) that display similar ringing to Lightages.
Brings up also the question: What for are high impedance probes used? Certainly not for perfect pulse response at hundreds of MHz. For that you need other measures. Probing with high impedance probes is mostly done at places with strongly varying source impedances that are, most time, not taken into account. So you need to know not to expect the "perfect". And sometimes how the probe loading affects the DUT if that counts. It has its limits. Doesn't mean they may limit the bandwidth of scope or may have uneven frequency/phase response by itself. But most time the circuitry does it due to high impedances, especially with analogue circuitry "seen" by the probe.
-
It was described before but people seem to not be understanding the situation. Perhaps this is my fault so here it is again, in excruciating detail.
I am using an analog function generator that has a dedicated TTL output, not rated at 50ohm. I have a 50ohm terminator on one side of the T and a BNC coax on the other end. This gives a ringing square wave as the output. It is not meant to be a proper square wave. I was looking for the fastest rise time I could get from the output. It was just by accident that I arrived at this setup.
I am still waiting for my SDG1025 to arrive.
I have a DS1052E upgraded to 100MHz that came with RP3300 probes, old versions, that do not have a compensation adjustment at the BNC.
I have a DS1054Z upgraded to 100MHz that came with RP2200 probes.
I was casually comparing the performance of the two systems. When I started looking at the waveforms that were displayed on both scopes while probing the same signal I noticed that the DS1052E seemed to show more detail on the waveform than the DS1054Z. So thinking that perhaps the upgrade had not worked on the DS1054Z I started to investigate. After some poking around I swapped the probes from one scope to the next (after compensating of course) and the difference I observed moved with the probes.
Both RP3300 probes showed more detail on the waveform than the 4 RP2200 probes I have, and on either scope and on any of the channels. One RP2200 and one RP3300 probe were always connected at the same time to the signal so any difference in loading of the signal from the different probe model is a moot point. I have tried probing the source BNC directly with ground spring clips instead of the alligator clip leads, the same on the end of the original arrangement, at the T adapter, etc....
In each and every case, the same difference in the intentionally ringing waveform appeared. The RP3300 probes show more detail and higher peaks on the ringing than the RP2200, regardless of scope, channel and test method.
When calculated according to the formula presented on Dave's video, a theoretical 100MHz bandwidth scope with theoretical 150MHZ probe will have system bandwidth of around 84.2MHz. The same scope with a 350MHz probe will have a bandwidth of 96.2MHZ. With this in mind, it is a small leap to justify the difference I see with my different probes. It also shows that theoretically the probes limit the performance of the scope, whether it is normal practice to supply the same bandwidth probes for a scope or not.
It has now been demonstrated, at least in one person's case, that the DS1054Z upgraded to 100MHz actually seems to have a bandwidth of between 130MHz and 150MHz. With this in mind the difference I see in the waveform is even easier to rationalize. It is because this affects what effect the probes have on the bandwidth. It is a wider spread in the system bandwidth, namely 106MHZ for the RP2200 vs 138MHz for the RP3300. This is assuming that the scope has 150MHz bandwidth.
So as far as I am concerned the difference that I see is easily explained by these last theoretical calculations and it answers my question about whether or not the probes as supplied affect the overall scope's performance.
The thing is, the 'ringing' in your signal is low enough in frequency that neither probe should change it's shape.
I've replicated your signal and compared it to a square wave (w/ similar rise time) and it's already pretty far down by 100MHz, so why are we seeing a difference between 150MHz and 350MHz probes, which should both be more or less completely flat to 100MHz? There should be very little change visible on an 8bit 'scope display when frequency components that are already 60dB down are further attenuated.
Time domain:
(https://farm9.staticflickr.com/8648/15724649653_12dc2f1e74_b.jpg)
Freq domain (800KHz to 200MHz):
(https://farm8.staticflickr.com/7518/15724649603_d07c467601_b.jpg)
And here is a 2.5MHz square wave generated by a DG4062 and probed with a DS1054Z(100MHz) and RP2200 probe using alligator clip on the BNC of the function gen. As expected the ringing is much higher in frequency and significantly smaller in amplitude, consistent with frequency components >100MHz being attenuated and components <100MHz being more or less untouched.
(https://farm8.staticflickr.com/7558/16154525047_be675f4ca3_b.jpg)
Imo what you are observing is an effect of the difference in capacitive loading by each of the probes, not the difference in their inherent bandwidth (frequency response). You'd probably see the same <100MHz effect with a 500MHz probe or by adding a small capacitor.
-
Imo what you are observing is an effect of the difference in capacitive loading by each of the probes, not the difference in their inherent bandwidth (frequency response). You'd probably see the same <100MHz effect with a 500MHz probe or by adding a small capacitor.
One RP2200 and one RP3300 probe were always connected at the same time to the signal so any difference in loading of the signal from the different probe model is a moot point.
|O
I understand what you are trying to say and what you have tried to simulate. What appears in the waveform looks like 50MHz ringing. The fact is that there is a difference with the 350MHz probes vs the 150MHz probes with the same conditions. Please understand that both probes were connected at the same time, therefore no different loading seen, they were loading the source in parallel. The same difference in the displayed waveform exists if they are connected one at a time too.
I have no other explanation as of yet except the conclusion I have reached. I wish I had a better way to test the conclusion, but I don't.
I am letting this go until someone who has the right equipment to measure my probes definitively steps up and offers to test them. That would be getting a bit more anal than even I am being about this right now. :-DD
-
Remember that just because the voltage is the same at the probe tips doesn't mean the current (and phase of the current) going into each probe is the same. If one of the harmonic components is shifted in phase slightly it can cause a dramatic visual difference in the waveform.
-
Yes, I understand that probes and their interactions are complex. The thing is that they display the same difference in the waveform respectively whether they are connected at the same time, separately, ground leads or ground clips, and connected directly to the source or through cables.
Without the right equipment there is no determining if what I see is the real performance of the probes or something else.
-
In the case of the 50MHz bandwidth model, [the RP2200 brings it] down to a theoretical 44.7MHz.
Unless the 50MHz model is slightly over 50MHz and with the 150MHz probe ends up at the specified 50MHz. In other words, maybe the bandwidth specification is the system bandwidth with the supplied probes.
-
I just came to a stupidly obvious realization...... :palm:
Perhaps all my other probes are good and the two RP3300 probes I have are the incorrect ones. Perhaps this is why Rigol changed the probe design to the way they are now instead of the design that I have. Perhaps they realized that 300MHz was pushing envelope of the old design. Remember that my RP3300 probes look identical to the RP2200 probes.
Sorry if this might have been an obvious thought to others, but when you get tunnel vision it can get really bad.
This makes more sense in light of the peakiness of the ringing at around 50MHz because this would be well within the bandwidth of any setup but faulty or badly designed probes could be peaky at that frequency.
-
Well I have measured with two scopes, all the available channels, with the cable from the function generator open ended alligator clips, 50 ohm load at the generator, at the end of a BNC terminated cable, with the spring clip ground point on the probes, directly into the function generator, and all the different arrangements that are possible. Yes the observed wave form changed, but there is always the same relative difference of the appearance of the wave form between the RP2200 and the RP3300.
I am either doing something wrong, ...
Yes, you are doing something wrong. There's absolutely no point in trying to measure bandwidth/risetimes with the ringing shown on your screen capture, since the parasitic components causing the ringing will define the circuit's response - and the probe+scope response will be irrelevant. So, what are those parasitic components?
In this and, as far as I can see subsequent notes, you have ignored one very important feature of your measurement: the wire associated with "cable from the function generator open ended alligator clips" and the wire associated with "spring clip ground point on the probes". Those wires are inductors, and they resonate with the probe tip capacitance to give the ringing overshoot in your screen capture.
I can't estimate the inductance introduced since you haven't specified how long the wires are, but online calculators will show that 10cm of wire has an inductance of around 100nH. Couple that with the probe tip capacitance and it is capable of explaining the waveforms seen.
The first step, as others have pointed out, is to get rid of the ringing by getting rid of the inductance, which means getting rid of the ground wires. Either connect directly, or use a short ground such as the bayonet ground illustrated below (a 6.5pF 500MHz probe).
Note that the probe's high frequency compensation is irrelevant to those effects.
-
Thanks for the response tggzzz, but I think you missed that I probed directly at the BNC of the output on the generator too, and the the apparent difference between the probes was still there. I also said that I was trying to explain the difference I saw in the waveform, it was not to measure the rise time.
As I said in my last post, it seems more likely now that the RP3300 probes are either faulty or poorly designed. They have been the only probes that show this extra peakiness to the signal. All my other probes, RP2200 and some P6300? probes I just received, show the same waveform. This points to the RP3300 probes being the inaccurate ones. I made the assumption that the RP3300 probes would be better than the others and my logic or "illogic" flowed from that.
-
Perhaps they realized that 300MHz was pushing envelope of the old design.
I've seen this done before. I've another probe where they stretched bandwidth from 75 to 100MHz which causes a slight peak in the response, but they documented it. As fa as I know, Rigol does not specify response flatness.
-
Hi, ;D
I dit some testing with probes on my Rigol DS1104Z.
Yellow probe: Rogol RP2200, 150Mhz ~17pF
Pink probe: Picotech Type?, 60Mhz
Blue probe: Philips PM8935, 250Mhz ~11pF (0.8nS)
1Khz Generator = Hameg HMF2525, about 1.2V
First the pictures of the generator en the probe connectors :D
(http://www.bramcam.nl/Rigol/1Khz-Gen-02.png)
Frequentie setting.
(http://www.bramcam.nl/Rigol/1Khz-Gen-01.png)
First, remenber...
Yellow - Rigol
Pink - Picotech
Blue - Philips
1Khz 100usec settings on the Rigol DS1104 scoop
(http://www.bramcam.nl/Rigol/1Khz-100uSec.png)
1Khz 5usec settings on the Rigol DS1104 scoop
(http://www.bramcam.nl/Rigol/1Khz-5uSec.png)
1Khz 1usec settings on the Rigol DS1104 scoop
(http://www.bramcam.nl/Rigol/1Khz-1uSec.png)
1Khz 100nsec settings on the Rigol DS1104 scoop
(http://www.bramcam.nl/Rigol/1Khz-100nSec.png)
1Khz 20nsec settings on the Rigol DS1104 scoop
(http://www.bramcam.nl/Rigol/1Khz-20nSec.png)
1Khz 5nsec settings on the Rigol DS1104 scoop
(http://www.bramcam.nl/Rigol/1Khz-5nSec.png)
Anyone tried the Rigol RP2200 Probe? "Thats not so good Al..." (Weezer)
The difference in time, is due to the length of the probe cable.
I am not to happy about the RP2200 Probes...
Kind regarts,
Blackdog
-
You can't expect top notch performance from cheap probes that often use simple of the shelf coax. Normal coax will ring anyway when not properly terminated with its characteristic impedance. The better probes use "resistive" coax to damp the ringing.
B.t.w. you loaded the generator with 3 probes, that means a loading of about 45 - 50 pF. Given the Ri of the generator of 50 ohms, that limits your BW anyway to app. 70 MHZ or a rise time of about 5 - 6 nS. Using a 50 ohms feed-trough terminator reduces the source resistance to 25 ohms, speeding up the source 2x. With a faster rise time the RP2200 will probably ring a lot more.
-
Hi Pjotr,
Do you realy think i'am not aware about the loading of the generator?
I dit a test whit 3 different probes, one slower than de Rigol, one faster than the Rigol.
Both beat the crap out of the Rigol Probe, look at the pictures, there is a lot of oveshoot whit the Rigol probe...
Its all about the difference between the probes...
And the probe's were all measurd at the same time.
Kind regarts,
Blackdog
-
Yes, we are proving that very cheap probes are crap anyway :D
My point was that with a faster source you will see even more ringing from the poor thing.
-
Anyone tried the Rigol RP2200 Probe? "Thats not so good Al..." (Weezer)
The difference in time, is due to the length of the probe cable.
I am not to happy about the RP2200 Probes...
I had the same result in my test. The RP2200 overshoot.
https://www.eevblog.com/forum/testgear/the-possible-hidden-cost-of-the-ds1054z-bandwidth-upgrade-hack/msg573856/#msg573856 (https://www.eevblog.com/forum/testgear/the-possible-hidden-cost-of-the-ds1054z-bandwidth-upgrade-hack/msg573856/#msg573856)
-
That picotech 60MHz probe beats them both, I know you don't have the model number posted, but if you find out what is it please share.
-
Has anybody mentioned the rubber/plastic on the ground alligator clip? The dang thing is stiffer than a wedding dick, very difficult to squeeze and release. :--
-
First the pictures of the generator en the probe connectors :D
(http://www.bramcam.nl/Rigol/1Khz-Gen-02.png)
Blackdog, what adapters did you use to connect scope probe tips to the BNC? Do you know a source in EU or eBay?
I searched for such adapter pieces but could only find very expensive HP/Agilent/Tek adapters.
-
Farnell has some sold separately from Tenma for about €6,-. But those adaptors are not universal because the size of probe tips differs. They fit or they are too big: Loose gnd contact, but still somewhat usable. Or they are too small and don't work at all.
-
Blackdog, what adapters did you use to connect scope probe tips to the BNC? Do you know a source in EU or eBay?
I searched for such adapter pieces but could only find very expensive HP/Agilent/Tek adapters.
All the probe tip to BNC adapters I have have come with the better quality probes I have.
They are not normally included with probes that come with entry level DSO's.
And yes they can be different sizes for some probe tips where the GND sleeve connects to the adapter.
-
Has anybody mentioned the rubber/plastic on the ground alligator clip? The dang thing is stiffer than a wedding dick, very difficult to squeeze and release. :--
I agree, the alligator clip is as stiff as dead dingos donger.
-
All the probe tip to BNC adapters I have have come with the better quality probes I have.
Mine have all come bundled with expensive probes, designed to fit those probes, they are quite loose on the Rigol probes, often having to hold them in place by hand.
-
The Farnell order code for the Tenma adaptor is 2414908. Given the price you can give it a try.
(http://lt.farnell.com/productimages/standard/en_US/5242229.jpg)
-
All the probe tip to BNC adapters I have have come with the better quality probes I have.
Mine have all come bundled with expensive probes, designed to fit those probes, they are quite loose on the Rigol probes, often having to hold them in place by hand.
Exactly.
The Farnell order code for the Tenma adaptor is 2414908. Given the price you can give it a try.
(http://lt.farnell.com/productimages/standard/en_US/5242229.jpg)
Do they give an probe tip diameter that it fits?
Anybody mic'ed up the Rigol probes?
-
The data sheet states 5mm: http://www.farnell.com/datasheets/1801350.pdf (http://www.farnell.com/datasheets/1801350.pdf) The ground sleeve of the Rigol probes is about 4.6 mm. But since it looks it has "springy" ground contacts it can be adjusted somewhat I guess.
-
Testec probes also come with a BNC to probe tip adapter.
Their probes start at EUR 15.70, so that's another possible source for an adapter:
http://www.reichelt.de/Probes-and-Accessories/2/index.html?&ACTION=2&LA=3&GROUPID=4043&LANGUAGE=EN (http://www.reichelt.de/Probes-and-Accessories/2/index.html?&ACTION=2&LA=3&GROUPID=4043&LANGUAGE=EN)
-
I have a few of them and those look exactly the same as the Tenma. The Testec adaptor is also a few tenths of a mm too wide for the Rigol probes and doesn't make a good contact. Better invest into the Testec HF series or similar then directly :D
-
Those Testec probes look like good value, IF they live up to their specs and are good quality.