EEVblog 1557 - Micsig SigOFIT Optical Fibre Probe + GaN Experiment

[EEVblog]

Review of the new Micsig SigOFIT Optical Fibre oscilloscope probe, including a practical demonstation of the incredible CMRR range of opticallly isolated probing of a 300V GaN transistor driver circuit.
And a direct comparison with the HVP-70 high voltage differential probe to show you the incredible game changing difference an optical fibre probes makes.

https://www.micsig.com/SigOFIT/

CMRR Explained video:

Links:
https://www.infineon.com/dgdl/Infineon-1EDF5673K-DataSheet-v02_04-EN.pdf
https://www.infineon.com/dgdl/Infineon-ApplicationNote_CoolGaN_600V_emode_HEMTs_-Driving_CoolGaN_high_electron_mobility_transistors_with_EiceDRIVER_%201EDI_Compact-AN-v01_00-EN.pdf
https://www.infineon.com/dgdl/Infineon-ApplicationNote_CoolGaN_600V_emode_HEMTs-AN-v01_00-EN.pdf
https://www.infineon.com/dgdl/Infineon-ApplicationNote_EvaluationBoard_EVAL_1EDF_G1_HB_GAN-ApplicationNotes-v01_03-EN.pdf
https://www.infineon.com/dgdl/Infineon-GalliumNitride_CoolGaN_emode_HEMTs-ProductSelectionGuide-v01_01-EN.pdf
https://www.infineon.com/dgdl/Infineon-IGOT60R070D1-DataSheet-v02_14-EN.pdf
https://www.infineon.com/dgdl/Infineon-ProductBrief_GalliumNitride_CoolGaN_600V_e-mode_HEMTs-ProductBrief-v03_00-EN.pdf
https://www.infineon.com/dgdl/Infineon-SolutionBrief_CoolGaN_and_GaN_EiceDRIVER-ProductBrief-v01_00-EN.pdf
Demo board: https://www.digikey.com/en/products/detail/infineon-technologies/EVAL1EDFG1BHBGANTOBO1/13548558

00:00 - Micsig SigOFIT Optical Fibre Probe vs Active FET
02:25 - High Voltage Differential Limitation and why you might need an Optical Fibre probe
03:25 - Galvanically isolated, and power-over-fibre
04:39 - Comparison with the Tek IsoVu & Lecroy DL-ISO
05:55 - The CMRR is AMAZING!
07:15 - A look at the SigOFIT probe
10:04 - An optical fibre probe is NOT a high voltage differential probe!
10:40 - Practical example using a 600V Gallium Nitride GaN circuit demo
11:39 - What is a Gallium Nitride (GaN) Transistor?
13:23 - GaN / HEMT, and Silicon Carbide SiC Transistors
14:21 - IGOT60R070D1 600V GaN datasheet & Half Bridge switching configuration
15:30 - This is IMPOSSIBLE with a regular probe!
19:07 - GaN EiceDRIVER chip
19:54 - The practical GaN test setup
23:10 - Remote MXO4 scope control to take measurements
24:36 - High side GaN power transistor measurement with the SigOFIT
26:26 - Here is where the CMRR MAGIC happens!
27:41 - Comparison with HVP-70 High Voltage Differential Probe
28:45 - Look how the High Voltage Differential Probe is just horrible here!
31:56 - Baseline High Voltage Differential probe ringing vs SigOFIT
32:21 - Noise measurement
36:27 - Conclusion

[dobsonr741]

Amazing! How would anyone go about making a diy version of an optically isolated probe, both on signal and power side? If not the full cable length opto isolation, then perhaps with optoisolators?

[BrianHG]

Amazing! How would anyone go about making a diy version of an optically isolated probe, both on signal and power side? If not the full cable length opto isolation, then perhaps with optoisolators?

I was searching for it, but, the thread is gone.
There existed here at eevblog a 10$ LED with photodiode based isolated probe where 10MHz bandwidth was achieved.  Cant find it anymore.  The probe side could have been powered by battery.

A laserdiode would probably give far better bandwidth.
<2.5ns photodiodes for the RX are cheap today.
It's the voltage drift calibration which would be most difficult to contain with an analog design.

[SilverSolder]

It's amazing what's possible today.  The performance of the EEVblog differential probe stood up quite well, I thought.  I guess the capacitance on the input to the probe could quickly become a real problem when measuring gate voltages?

Amazing! How would anyone go about making a diy version of an optically isolated probe, both on signal and power side? If not the full cable length opto isolation, then perhaps with optoisolators?

I was searching for it, but, the thread is gone.
[...]

Is this the thread you were thinking of?  https://www.eevblog.com/forum/beginners/floating-probe!-for-$2-50/?all

[Weston]

Was it a loaner and they did not want you taking it apart? It would be cool to see the insides where the magic happens.

I was also interested in creating a DIY fiber isolated probe a while ago.

I forgot where I saw them, but there were internal photos of the saker fiber isolated probes somewhere https://saker-mv.com/fiber-isolated-voltage-probe/ .The optics package does not look too complicated, and at least for the Saker probe it looks like they are directly modulating a led / laser diode. I think the tek fiber probe uses some sort of electro-optical modulator. I assume that has better linearity / less drift.

They are not stocked on Digikey, but you can buy some pretty fast fiber coupled laser diodes and photodiodes with high bandwidth for <$100. I ran some quick figures on the noise figures a while ago and it seemed like you could get at least 70dB of dynamic range with 100MHz+ bandwidth. The biggest challenge is compensating for drift and any non-linearity (or biasing it to reduce non-linearity). I was curious if you could use the back-face monitor photodiode that some laser diodes have to linearize the system but I never found any values for bandwidth of it. 

[EEVblog]

Was it a loaner and they did not want you taking it apart? It would be cool to see the insides where the magic happens.

They advised me not taking it apart as I wouldn't see anything. They implied it was potted or some such. Whether that is a ruse or not I don't know. I'll have a crack at it at least, but I don't want to destroy it.
So I didn't risk it before I finished the video playing with it.

[EEVblog]

Screw holes look hard potted 
Probe end I can't see any way in short of ripping off all the rubber surround.

[tszaboo]

Are we going to get a teardown? The probe is quite interesting.

[EEVblog]

Are we going to get a teardown? The probe is quite interesting.

Doesn't look like it. As it seems I likely have to damage the case to get it apart.

[Marco]

Amazing! How would anyone go about making a diy version of an optically isolated probe, both on signal and power side? If not the full cable length opto isolation, then perhaps with optoisolators?

For 100 MHz range, I'd look for some pigtailed laser diodes and pig tailed PIN diodes and work from there. As for power, just use a battery.

[Fungus]

How does it power the probe part? Laser? How many watts? How do they convert the light to power at the other end?

[SilverSolder]

[...]  How do they convert the light to power at the other end?

Solar roadway?   

[EEVblog]

How do they convert the light to power at the other end?

A photo diode can do that.

[BrianHG]

How do they convert the light to power at the other end?

A photo diode can do that.

Especially one fed by a laser, if the photodiode is tuned to the optical wavelength, you can feed through a significant amount of power.

[switchabl]

I forgot where I saw them, but there were internal photos of the saker fiber isolated probes somewhere https://saker-mv.com/fiber-isolated-voltage-probe/ .The optics package does not look too complicated, and at least for the Saker probe it looks like they are directly modulating a led / laser diode..

Shariar did a teardown of the Saker probe:

[bsdphk]

How do they convert the light to power at the other end?

A photo diode can do that.

Bell Labs prototyped an optically connected telephone in 1979:

https://archive.org/details/bstj58-7-1735/mode/2up

[tszaboo]

Amazing! How would anyone go about making a diy version of an optically isolated probe, both on signal and power side? If not the full cable length opto isolation, then perhaps with optoisolators?

I vaguely remember seeing a Linear technology application note. It was about making linear characteristics for an optocouple. So they had an equal optocouple in the feedback, and it was used on the transmit side to correct for any nonlinear characteristics. I don't know how linear optical cables and fiber optic transmitters are, but that's one way of doing it. Or maybe they are doing it completely differently, since 10Gb and even 100Gb optical networks existed or a while, they might just send it all digital.

[nctnico]

Amazing! How would anyone go about making a diy version of an optically isolated probe, both on signal and power side? If not the full cable length opto isolation, then perhaps with optoisolators?

I vaguely remember seeing a Linear technology application note. It was about making linear characteristics for an optocouple. So they had an equal optocouple in the feedback, and it was used on the transmit side to correct for any nonlinear characteristics. I don't know how linear optical cables and fiber optic transmitters are, but that's one way of doing it. Or maybe they are doing it completely differently, since 10Gb and even 100Gb optical networks existed or a while, they might just send it all digital.

I've worked on such a device a decade ago. The entire acquisition system (battery powered) is connected using an optical (ethernet) network link and the data is send pre-formatted over the network. But this is for use in high voltage labs where they use it to measure into the MVolt / tens of kAmpere range and float the units to tens of kVolt.

The biggest trick is to get this techology to become cheap. You can probably get a long way using a battery and SFP + FPGA + ADC / DAC

[Fungus]

You can probably get a long way using a battery and SFP + FPGA + ADC / DAC

I was thinking that, too. If it's low-power enough to be powered by light then surely a PP9 battery would go a long way. 

[tszaboo]

Amazing! How would anyone go about making a diy version of an optically isolated probe, both on signal and power side? If not the full cable length opto isolation, then perhaps with optoisolators?

I vaguely remember seeing a Linear technology application note. It was about making linear characteristics for an optocouple. So they had an equal optocouple in the feedback, and it was used on the transmit side to correct for any nonlinear characteristics. I don't know how linear optical cables and fiber optic transmitters are, but that's one way of doing it. Or maybe they are doing it completely differently, since 10Gb and even 100Gb optical networks existed or a while, they might just send it all digital.

I've worked on such a device a decade ago. The entire acquisition system (battery powered) is connected using an optical (ethernet) network link and the data is send pre-formatted over the network. But this is for use in high voltage labs where they use it to measure into the MVolt / tens of kAmpere range and float the units to tens of kVolt.

The biggest trick is to get this techology to become cheap. You can probably get a long way using a battery and SFP + FPGA + ADC / DAC

For a "low" frequency system in the digital domain, there are Delta-Sigma modulators that output bitstream, that practically can directly be sent to a SPF and on the RX side it can be turned into an analog signal with an 1 bit DAC. How well this would work in practice, I don't know, probably DC accuracy needs calibration, and Delta-Sigma typically stops being practical at 10MHz. Maybe it's possible to push this a bit with very high speed opamps and comparators. In theory an optical probe with this would be super cheap, but also not very userful due to the limited frequecy.
For high speed, maybe it's possible to skip the FPGA, if one has a good match of interfaces. There needs to be an ADC, probably Flash ADC, generating data at that very high 100MSPS-1GSPS speed paired up with a SERDES. With the RX side, having a similar SERDES and a DAC. I think the difficulty could be finding such a SERDES, as most of them seem to be only targeting video interfaces.
I think if someone could put together a 50-100MHz probe without going crazy in the BOM and development costs, there is a market to grab.

[Marco]

I don't see how going entirely digital would ever allow a cheap hobbyist grade product at 100 MHz. Two path is an option and would make calibration easy, digital/PWM for LF (across a separate fiber) and direct amplitude modulation of laser current for HF.

[tszaboo]

I don't see how going entirely digital would ever allow a cheap hobbyist grade product at 100 MHz. Two path is an option and would make calibration easy, digital/PWM for LF (across a separate fiber) and direct amplitude modulation of laser current for HF.

I know what you mean, but I always get surprised on some part costs. Like the digital radio and TV receivers brought down the hobby SDR and VNA costs from thousands to dozens.
I had a brief look, and TI is selling laser drivers for 2 dollars, that do 10 gigabit. Maybe some video ADC and DAC can be repurposed for it, or for an oscilloscope even an 8 bit R-2R DAC could provide enough resolution. Since the most scopes we have don't have more than 8 bit anyway.
Something came to me. I think we can tell a lot about it just by looking at the delay of the probe. Micsig datasheet is claiming 15ns delay, which, for a 1GHz probe is a lot, but for actual propagation delay in a POF cable is nothing. I think it's safe to say they do it digital.

[Weston]

I really wonder what the optics package in the Micsig probe consists of. Just the setup for transferring power over fiber is pretty costly. I bet they are using that to transmit data to the probe head though (if needed).

Dave: Could you please take some linearity / FFT measurements with the probe? As a function of input amplitude? It would be interesting to see. Any setup modulating a laser diode directly (possibly this probe, likely the saker one) will be more non-linear than something using a electro-optical modulator (like the Tek probe). So any harmonics should be indicative of how it works.

Close up waveforms at ms+ time constants of the settling behavior (like the last few fractions of a percent of vertical settling) for a large voltage step would also be interesting. I am a bit curious if there are any thermal-scale time constants due to heating of the optical TX or RX.

[BreakingOhmsLaw]

Hey Dave, if you're measuring GaN or SiC, you should look into Double Pulse Testing.

https://www.infineon.com/dgdl/Infineon-Double_pulse_testing-Bodos_power_systems-Article-v01_00-EN.pdf?fileId=5546d46271bf4f920171ee81ad6c4a1f