Author Topic: LNH-5010S - 20MHz Rogowski Coil probe for 80USD  (Read 362 times)

0 Members and 1 Guest are viewing this topic.

Offline TopQuarkTopic starter

  • Frequent Contributor
  • **
  • Posts: 326
  • Country: hk
LNH-5010S - 20MHz Rogowski Coil probe for 80USD
« on: November 30, 2024, 12:00:05 pm »
Recently I've been window shopping for test equipment on Taobao, and I came across this Rogowski coil probe that claims 1kHz to 20MHz BW for 583 RMB (80.5 USD), they also list just the coil element for 187 RMB.

I always had a Micsig Rogowski coil probe in my shopping cart, but that thing cost 8-10 times more, and I can't really justify spending that for something I realistically won't be using that often. This thing is cheap enough for me to be not too upset if it doesn't work well.

I did actually come across this post (https://www.eevblog.com/forum/projects/mini-rogowski-coil-current-probe/) by member Weston before, who generously open sourced his Rogowski probe project here (https://github.com/westonb/rogowski-relief), and am aware of member dpenev's shop (https://rogprobe.com/) who sells a product based on this project. I'm mentioning this as I am almost certain the LNH-5010S is a rip-off of the original design from Weston, and I thought I'd give credit to Weston because the LNH-5010S seller didn't attribute the design to the original source, as mandated by the CC BY-SA 4.0 license.

I requested the 50mV/A version which had run out of stock, so we agreed for them to modify the gain of a 10mV/A probe instead, hence the 50mV/A sticker on the casing.
« Last Edit: November 30, 2024, 12:05:37 pm by TopQuark »
 

Offline TopQuarkTopic starter

  • Frequent Contributor
  • **
  • Posts: 326
  • Country: hk
Re: LNH-5010S - 20MHz Rogowski Coil probe for 80USD
« Reply #1 on: November 30, 2024, 12:28:55 pm »
Powering it on, the output of the probe is quite noisy, with a lot of low frequency noise (<100Hz) that seems quite random and not related to 50Hz mains. It is quite odd because the bandwidth of the probe starts at roughly 1kHz, so the noise isn't really from the current sensor itself. Looking at the profile of the noise, my suspicion is usage of MLCCs somewhere in the signal path are generating microphonics noise which is not filtered out adequately at the output of the signal path.

I made some comparisons with my ZCP30 50MHz DC coupled current probe, measuring current through a buck regulator switching MOSFET. CH1 is the LNH-5010 output, CH2 is the ZCP30 output. The current waveform matches up decently well, though some higher frequency details are missing from the LNH-5010 due to the lower bandwidth spec compared to the ZCP30 probe.
 

Offline TopQuarkTopic starter

  • Frequent Contributor
  • **
  • Posts: 326
  • Country: hk
Re: LNH-5010S - 20MHz Rogowski Coil probe for 80USD
« Reply #2 on: November 30, 2024, 12:56:26 pm »
To measure the frequency response of the LNH-5010, I made a fixture for the current probes to measure the center conductor current of a coax cable. The coax is terminated with the scope's 50ohm input, providing a reference signal for the bode plot. CH1 is the LNH-5010s, CH2 is the ZCP30 current probe for comparison.

The scope struggled to provide a clean trace at lower frequencies due to the noise mentioned before. Still, nasty peaking can be observed at around 100Hz, as well as 18.1MHz, towards the higher end of the probe's bandwidth.

Taking apart the probe reveals a few things:
  • There's no shielding around the amplifier area which is the most sensitive to noise.
  • The arrangement of the components as well as the OPA2822 opamp basically verifies this product is based on Weston's original design.
  • Trim pots needed to trim for a flat gain is missing, compared to Weston's design.
  • Some protection components are missing.

I think this kind of build quality and effort is roughly what I expected for this price, and it should be fun to see what I can do to make this better. I have a few thoughts regarding the component choices and ways of extending the lower frequency. Will play with the circuit and report back.  ;D
 
The following users thanked this post: nctnico, Kean, ch_scr

Offline Weston

  • Regular Contributor
  • *
  • Posts: 221
  • Country: us
Re: LNH-5010S - 20MHz Rogowski Coil probe for 80USD
« Reply #3 on: November 30, 2024, 07:38:24 pm »
Thanks for posting this and including pictures of the PCB. It really does look like it is derived from my design. Sadly they cut out about $2 of components which impact performance and make this a pretty bad implementation. Penny wise pound foolish.

I built mine because I was frustrated there was no low-cost mini probe. So it's nice to see China follow up and make one, I consider the copy a badge of honor  ;D .However, I am not pleased with the quality of this implementation, they cut so many corners. I might just have to buy one though. It would be funny to own a copy of something I designed.

Some Complaints:

All the protection components were removed, even the PCB spark gaps and fusable traces. Given how thin the insulation on the coil is, this is a serious safety risk.

The probe has horrible flatness with peaking at low frequency and at high frequency and a jump in gain on the transition from the active integrator to passive integrator. They removed all the adjustable components and it seems like they changed the coil properties without changing other important component values.


TopQuark, if you are interested in taking any more measurements:

What does the coil look/feel like? I spend hundreds of dollars on trying different heat shrink out for the coil and it looks like they used polyolefin heat shrink (the generic type of heat shrink used in most applications), which I had ruled out. Compared to the FEP heatshrink I used, polyolefin heat has a lower breakdown voltage, is more easily abraded, and is less heat resistant. All important factors when you are running this probe close to high power electronics that can get hot / have high dV/dt.


Whats the interference rejection look like? Try putting your cellphone a few inches from the coil or the amplifier and running an internet speed test while on WiFi or LTE. There will probably be volt+ spikes on the output. On my design the capacitance for the passive integrator on the input is distributed with careful layout to reject EMI coupled into the cable and coil and I used a small ferrite bead common mode choke on the input cable in conjunction with a shielded case. This design has none of those features.


Its hard to tell from your measurements, whats the RMS and P-P noise? As you said, it does look like they are using a X7R or similar capacitor for the servo highpass filter, which could couple in noise. But it also seems like the lower frequency cutoff is higher. Either way I am curious how the noise level compares to my probe.
« Last Edit: December 01, 2024, 05:20:15 am by Weston »
 
The following users thanked this post: Kean


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf