Products > Test Equipment

Dongerkeji DR100N1KV - digitally tunable x50/x500 100MHz HV diff probe

(1/1)

TopQuark:
I was shopping for a HV diff probe on Taobao for work, and I spotted this "Programmable" diff probe for around 230USD, from a Chinese company I never heard of before, called Dongerkeji (yea the english name is a bit dodgy  :-DD).

Company Website: http://www.dongerkeji.com/web-page/index.html#/
Taobao link: https://item.taobao.com/item.htm?id=610273138460
Manual: http://www.dongerkeji.com/web-server/files/%E4%B8%9C%E5%84%BF%E6%95%B0%E6%8E%A7%E5%B7%AE%E5%88%86%E6%8E%A2%E5%A4%B4%E8%AF%B4%E6%98%8E%E4%B9%A6V1.1%E7%89%882020-10-19.pdf

What makes this stand out, is it is supposed to be digitally tunable with a Windows PC through the probe's USB port. Things you can adjust include offset, and perhaps more interestingly, CMRR balance across a three ranges of frequency, for both attenuation ratios.

They claim this features allows the probe to have high CMRR by digitally fine tuning away the component tolerances, and this method does not suffer from drift as much as trim pots and trim caps (likely it will still drift, but at least you can tune it back to spec with a few clicks instead of using screwdrivers :-/O)

They claim CMRR of -70dB at 50Hz, -35dB at 50MHz, and actually include a typical CMRR graph in the manual, I don't think I've seen that anywhere else. BW wise, they claim 70MHz for 50x, and 100MHz for 500x.

So I went ahead and purchased one for work, and after testing it a bit, found it to be quite impressive for its cost, and got a couple more for home use.

I found the quality of the banana plugs, spring grabbers, BNC cable, USB cable to be usable, but a bit meh in quality. So what you see in the picture are the two original probes, but all accessories are custom.

Posts of test results will follow.

TopQuark:
As they made big claims about CMRR, it was the first item I tested.

Right off the bat, I found one of the two probes to meet CMRR spec, while the other one was off by a bit. Slightly disappointed, but hey, you are supposed to be able to tune the CMRR spec, right?

Tuning the CMRR performance requires their dodgy no-name software from their equally dodgy website, and the software was windows only while I run Linux on my machine. So I ran the software within a VM, and it worked fine.

The manual actually does not contain the instructions for tuning the probe, but I found a PPT on their website that does describe the process of tuning the probe.

The software talks to the probe through UART, and there are 4 sliders for each of the x50/x500 ranges. One is for offset, one for low frequency CMRR, and two for high frequency CMRR.

The described process does not require any special equipment other than a scope and a SMPS.
1. Short the inputs and adjust the offset
2. Connect both input to mains live wire  ::) and adjust LF CMRR until you don't see the signal
3. Connect both inputs to the Drain of a square wave-ish driven, hard switched SMPS MOSFET  ::) ::), adjust the HF slider so that the signal is minimised.  The bottom right slider is to reduce the spurs on the probe output caused by the switching edge, the top right slider is then tuned so that the top and bottom edge of the switching waveform lines up to be perfectly flat.

I used method 1 and 2 to satisfaction, but method 3 was a bit inconsistent (and too dodgy) for my taste.

I tried hooking both inputs to the center pin of a function gen outputting a fast switching 10V p-p 1MHz square wave and tune it that way. I do see the signal change as I adjust the sliders, but for the bottom right slider where you are supposed to minimise the switching edge residual, it was hard to determine if one ringing response was better than the other. As my goal is to try and match the two probes I have (rather than trying to hit certain performance numbers), I ended up using my scope's bode plot function, and adjust the sliders until both probes matched.

TopQuark:
After a bit of fiddling, I managed to get both probe's CMRR response to match quite closely, and the result is quite impressive for the price. Sure, it is not a PMK Bumblebee, and in 500x it doesn't quite meet spec, but at 230USD, I think this one of the better HV diff probe in terms of CMRR.

All CMRR graphs are referred to input, i.e. the channel attenuation ratio is set to match the probe's attenuation setting, and that is carried over into the bode plots, they are already scaled correctly.

I also measured the BW and gain flatness of the probe in both attenuation ratio. A bit of a bump towards the BW limit, but within reason IMO.

Navigation

[0] Message Index

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod