Here from Martin's screenshot we see some small offset with the MICSIG current probe WRT the voltage probe. Negative pulse inaccuracy ?
Also Deskew is not set correctly as the rising edge of the current should match the voltage rising edge better.
If both channels are set on 0V as these are and both similar amplitude the trigger needs be set to 50% and at that point the deskew adjusted so both signals cross through the trigger point at exactly the same instant. ps accuracy is possible.
For most measurement needs channel deskew does not matter very much but for accurates results from Power analysis it does. Practice makes perfect.
Oh and another trick when one trace is less noisy than the other.....select it so it overlays the noisy trace and they both become visible.
Also Deskew is not set correctly
As I said, I never used this before so any "hints" like yours are welcome.
Negative pulse inaccuracy ?
No, forget to press "zero" before...
Also Deskew is not set correctly
As I said, I never used this before so any "hints" like yours are welcome.
Well, what is maximum deskew time on SDS2000+ ?
On KS MSOX3000T it is ± 100ns... So on KS MSOX3000T you have to deskew current probe channel for max 100 ns , and then deskew rest of the channels in opposite direction (but all for same value to be in phase) to achieve delta of 170ns..
Maximum is +/- 100ns.
So on KS MSOX3000T you have to deskew current probe channel for max 100 ns , and then deskew rest of the channels in opposite direction (but all for same value to be in phase)
As I did the measurements, I also did one where both channels were deskewed to the max 100ns:
Negative pulse inaccuracy ?
No, forget to press "zero" before...
(Attachment Link)
OK on this one ch3 is the current probe and trigger so now ch4 needs deskew adjusted to have it also at the 0s, 0V trigger crossing position. For really fine adjustments increase the timebase as required.
As some example of propagation delays and the need for channel deskew to align edges for precise measurements here is 100 KHz squarewave connected to ch4 via BNC and ch3 via its 10x probe and a Tek P6021 current probe measuring the squarewave current into a 50 Ohm load.
Please disregard ringing from this setup as it's not relevant to the deskew discussion.
SDS5104X, source SDG6022X
Triggering is on ch4 @ 50% and all channels are aligned to some graticule and amplitudes adjusted so that any propagation delays are plain so deskew can be performed on channels 2 and 3 to align the rising edge 50% point and the required amount is shown in deskew and cursor measurements.
Cursors left at their settings but deskew adjustments for channels 2 and 3 entered so 50% points all cross 0V close together in time.
Does the 5k also have the 100ns limit for deskew?
Does the 5k also have the 100ns limit for deskew?
+100ns
So we'd select neither our fastest or slowest propagation channel but a midrange channel to trigger on and set deskew to line them all in time.
The difference between these 2 channels is ~7ns but each channel has been set to max and min deskew as an example.
Today I take it to work for further testing, on the desk the lecroy hdo6034a.
And yes it got deskew also (
), the adjustable range is +/- 990µs - WTF....
That's what I call deskew. I was a little disappointed with the ±500ns but if ±100ns is standard then I guess I'm happy with it.
I wish I had 500ns...
Although I think, 100ns are "enough" for compensating e.g different cable-lenghts.
Fortunately, I got a Power Inverter for testing for a few days.
It´s understood I want to test something with the Micsig, DC-Currents I´ve already done, so I can test with the Inverter AC-Currents up to 100A/400hz.
Further pwm power current with a frequency of 10khz.
I´m looking forward the micsig will do a good job on this also.
Update:
I send the micsig back to batronix for an exchange, reason:
Specsheet stated a delay under 150ns in 10A range, mine got an averaging delay of 170ns - Told them, they did a measure with another one and it was 125ns, so they offer me to exchange mine.
So AC current tests and my final verdict about will be delayed until I got the other one.
Let's hope the next one really is in spec! So perhaps they won't always meet all specs which is a bit disappointing.
Well...
If they hadn´t offered me the exchange, it wouldn´t be such a big showstopper for me.
The much more expensive examples I´ve linked in the first post do not even give this value in their specs - Maybe for a good reason...
The much more expensive examples I´ve linked in the first post do not even give this value in their specs - Maybe for a good reason..
The AP011 from lecroy for example...measure it today on work, appx 338ns delay.
So I don´t bother about 170ns, especially since you will use deskewing when taking power measurements.
New probe doesn´t have arrive yet, as batronix got no more on stock and offer me to return my former one to me plus the money, until they have new (and checked before) in stock.
I´ve agreed to it, as I want to make further measures asap to close the "case", giving a final verdict.
Update, the second.
Still got no new probe as they aren´t arrived to batronix.
Now I got a problem, in ebay there was an offer I could not longer withstand it and asked for the last price.
A used combo of tektronix A6302 and mainframe/amplifier.
If I could get it to nearly the price of the micsig, I´ll buy this, sorry.
And the last verdict will be therefore my final:
- Surprisingly low noise
- Accuracy always in the specs
- For it´s price, I think you´ll never get a better one, even if they would cost double or more ( because of the measuring principle - only currentcompensating models should be better in accuracy, these ones you won´t find under 3k€)
If I don´t get the tek, I´ll continue testing.
Martin
Update, the third...
Couldn´t get the tek probe, therefore looking further to get the micsig back.
It´s still out of stock - But the cp2100A with "only" 800khz bandwith are in stock and therefore I´ve order one now.
Let´s see if this perform as well as the B version - If yes, it would be the best bang for bucks, as it only costs 180€ (excl. VAT)....
Should arrive inbetween this week.
Martin
Update, the third...
Couldn´t get the tek probe, therefore looking further to get the micsig back.
It´s still out of stock - But the cp2100A with "only" 800khz bandwith are in stock and therefore I´ve order one now.
Let´s see if this perform as well as the B version - If yes, it would be the best bang for bucks, as it only costs 180€ (excl. VAT)....
Should arrive inbetween this week.
Martin
Always in stock if you are willing to wait on the boat lol.. lowest i've seen for the B is $360 usd or so
I'd think they would be very similar just the 3db point is 0.8 MHz vs 2.5 MHz
Saw a recent review here of the B as well
Nice little review..
The cp2100A does arrive tomorrow
Batronix are fast as lightening.
And as it was on it´s way, another guy from them mailed me, ah yes, today we got the cp2100B on stock...Thank you.
But first I want test the A model, if I don´t find any disadvantages against the B except the bandwith, I think it could stay here.
Tadaa, it´s arrived...
Looks, smells, feels the same, the clamp itself was 100% the same as on the "B" version.
Makes me curious to know what´s inside the box....
And at last I could test switching frequency - I got my deskew fixture....
Tests will follow.
Martin