High Voltage Forum

General Chat => Laboratories, Equipment and Tools => Topic started by: davekni on October 08, 2020, 05:23:05 AM

Title: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: davekni on October 08, 2020, 05:23:05 AM
This is an optically-isolated scope probe for use in probing high-side gate drive and other such non-ground-referenced signals.  Also eventually plan to use a pair of these to measure charge transfer on top of my DRSSTC, from secondary to top-load and from secondary to breakout point.

This probe uses standard 1mm-core plastic optical fiber.  A "T-1 3/4" (5mm-diameter) LED feeds the fiber with light proportional to input signal plus a fixed DC offset.  Photodiode receiver amplifies the light signal and subtracts the fixed offset.  Most newer efficient LEDs have surprisingly linear drive-current to light-output.  Here's a picture of the probe, driver and receiver and fiber.  Each end is powered by a flat 4V lithium-ion cell under the ECB.

 [ Invalid Attachment ]

Bandwidth without any compensation for LED capacitance is just over 4MHz.  With an added R+C to peak LED current, bandwidth is roughly 10MHz.  (I tried a couple LEDs packaged specifically for 1mm-core 2.2mm OD optical fiber.  They were higher-bandwidth, but much less efficient.  Light coupled into the fiber was much lower, so would have required higher receiver gain and associated higher noise.)  Below are scope traces of a 12V square wave out of a UCC27525 gate-driver chip.  Channel 4 is with a normal 10x scope probe.  Channel 1 is this isolated probe output.  Both scope channels are set to 20MHz bandwidth limit.  Notice that this optical probe output has roughly twice the rise/fall times, so about half the bandwidth.  It also has more delay due to the coil of fiber.

 [ Invalid Attachment ]

 [ Invalid Attachment ]

 [ Invalid Attachment ]

Input divider resistors for this first unit are designed for gate-drive scoping, +-20V.  Output to scope is +-200mV (100x attenuation).  Input impedance is 30k-ohms.  Low enough to avoid needing to tweak input attenuator compensation capacitors, but high enough to not significantly load gate-drive signals.  Here's the driver schematic including R10+C3 peaking to compensate for LED capacitance:

 [ Invalid Attachment ]

And the receiver using the IF-D91 photociode designed for plastic fiber:

 [ Invalid Attachment ]

I found two successful ways to couple the 5mm OD LED to 2.2mm OD (1.0mm core) fiber.  One is to drill a 2.2mm hole in the LED deep enough to almost hit the bond wire, then glue one end of the fiber into the hole.  Glue makes reasonable optical matching (similar index of refraction to the LED housing and fiber core).  Here's a close-up of the LED with fiber glued into it:

 [ Invalid Attachment ]

The other option is to grind/sand down the end of the LED to make a flat surface, again almost to the bond wire.  Didn't have any good way to polish the surface, so instead glued a small piece of mylar to the end to make a shiny surface.  Then used a short piece of 3/16" ID (just under 5mm) rubber tubing to couple the LED to fiber.  I'd found some fiber with an outer jacket that was conveniently 5mm OD.  The 3/16" ID tubing could either join the LED directly to this fiber, or I could remove a short section of the large fiber's outer jacket, join that section to the LED with rubber tubing, then insert normal 2.2mm OD fiber into that piece of jacket.

Finally, on the outside chance that anyone wants to experiment with this design unmodified, here are the zipped gerber files:

 [ Invalid Attachment ]

Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: rikkitikkitavi on October 08, 2020, 08:46:14 AM
Ingenious!  :)

You have the advantage of transfering the signal galvanically isolated in full without bothering with the CMRR of a differential high voltage divider with all its asociated troubles in high impedance/parastics and high voltage/large transients/sharp edges power circuits yadayada!

Fiber delay is about 5ns/meter 
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: Hydron on October 08, 2020, 05:09:46 PM
Nice work!

You can also get analogue output receivers for plastic fibre (e.g. versalink) and glass (e.g. multimode ST-connector ones).

I have one of the former and a couple of the latter, and did a quick gain vs frequency check quite a while ago on a multimode glass one, see attached (Transmitter was a HFBR1415, Receiver was a HFBR-2414 which doesn't seem to be well documented, but is an analogue output version). I didn't really look at linearity, but the datasheet suggests it will be fairly reasonable.

I have some other methods of getting isolated measurements, but these guys are sitting there in-case I need to build a similar isolated probe.

Another thought if linearity proves to be troublesome would be to do something similar to analogue optocoupler circuits - put a second receiver in the feedback loop for the LED driver. If this is fed from a splitter half-way down the fibre (sending back half the signal to the driver) then in theory it should linearise the output (assuming that the receiver characteristics are well matched). This is a lot more hassle and probably unnecessary for most applications (e.g. you just want to see the waveform shape, a few % distortion is fine) but maybe useful for more critical signals?


Red is gain, Blue is phase. Filename suggests the gain is measured as voltage output vs current input (this was done as a quick hack a while ago!) I wouldn't put too much faith in this, but thought it may be of interest.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: klugesmith on October 08, 2020, 06:31:59 PM
Great work there, Dave.  I like the DIY methods of launching light into fiber.  In 1990's, working with singlemode fiber, I found that some gigabit laser transmitters didn't like the return loss from an open fiber end (in SC connector).  Got a satisfactory termination by rubbing fingertip on oily nose & then pressing it onto the SC ferrule tip.

How stable is your zero-signal point ( transmitted as about half of maximum optical power ? )
over temperature, battery age, and maybe wiggling of optical joints?
I guess it's a non-issue for GDT output signals that are inherently zero-average, and otherwise a simple adjustment of vertical position or offset knob.

Applause for posting schematics, and having a name on every net.   Did I miss the place in sender circuit for connecting the reference side of input signal voltage?

Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: davekni on October 09, 2020, 12:03:44 AM
Hydron,

Before starting this project, I'd looked for integrated analog fiber receivers without success.  Since you listed HFBR-2414, I just checked into that a bit.  Mouser lists it as an obsolete part.  Looking through the data sheet, it might be usable, but looks to have three diode-junctions worth of DC voltage temperature coefficient (3 x Vbe) on its output.  Do you have any part numbers for currently-available analog fiber receivers?  Such could be simpler and higher bandwidth than my version.

Klugesmith,

Good point - I'd forgot to label which transmitter node is the input reference.  It is node "vtgnd", the node that the input voltage divider returns to.  There are extra holes/pads on the ECB layout for vtgnd that don't have corresponding schematic symbols.  These can be used for input return and/or for connecting a foil shield around the circuit if that becomes necessary.

The idea for gain/offset is as follows.  The transmitter has a fixed LED current of 11.6mA at 0 input voltage.  That 11.6mA comes from 0.50V across 43 ohms.  The exact current doesn't matter as long as it corresponds to exactly 0.5V.  Whatever amount of light is generated and coupled all the way to the receiver is by-definition 0.5V of input voltage.  (That's 0.5V after attenuator, so 25V with my 50:1 input attenuation.)  At the receiving end there's a fixed -0.25V offset to the scope (scope ground is 0.25V above local receiver ground).  With no light to the receiver, the scope sees -0.25V.  With light from the transmitter, I adjust gain until the scope sees exactly 0.0V (with no input voltage to the transmitter).  Then I know that 0.5V at the transmitter (25V into attenuator) corresponds to 0.25V to the scope, for a total 100:1 attenuation.  Any time I reconnect fibers or the LED temperature changes too much, I can adjust gain to get zero volts out.  Drift hasn't been much of a problem with the TINY bit of experimenting I've done so far.

The one issue with all the above theory is it assumes zero opamp offset.  If there's a next-version, I'll add receiver opamp offset adjustment.  For this version, I had to add a hand-selected resistor on the back-side to adjust for opamp offset.

I'd tried cheap laser-pointer diodes for the transmitter.  Much higher optical power into the fiber, but two issues.  One is less stable light output vs. temperature.  The other (related) issue is a large thermal-transient in generated power after a step change in input current.  There's a tiny bit of that issue with the LED.  Actually I don't know for sure that it's thermal-related.  Just that the few microsecond time-constant is reminiscent of thermal transients often seen in analog power amplifiers, such as video drive for old CRT monitors.  If you look very closely at the overview plot of this fiber probe's response, the step goes quickly to ~98%, then takes a few microseconds to go the remaining ~2%.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: flyglas on October 09, 2020, 05:16:19 PM
This thread is very inspiring. I like the concept to measure small voltages without common mode influence.

I have searched for commercial units and other DIY projects. During my search I found a similar project:
 https://hackaday.io/project/12231-fiber-optic-isolated-voltage-probe (https://hackaday.io/project/12231-fiber-optic-isolated-voltage-probe)

The approach is similary, but they use other circuits to realise the transmitter (fet input stage) and a transimpedance amplifier as light to voltage conversion.
Layout and schematic is available.

Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: Hydron on October 13, 2020, 11:27:14 PM
I had a quick look, and the analogue-output receivers I have are the following:
2x HFBR-2526, POF:
https://docs.broadcom.com/doc/AV02-1502EN

1x HFBR-2416, glass fibre:
https://docs.broadcom.com/doc/AV02-0176EN

2x HFBR-2414, glass fibre, as discussed before this doesn't seem to be a current product or have much info available.

All were purchased from CPC's bargain section at greatly reduced prices, most likely due to CPC stopping stocking the parts.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: davekni on October 14, 2020, 12:40:06 AM
Those two receiver parts appear to still be reasonably available.  Both have very loose DC offset specifications, 0.8 to 2.6V with no light input.  They have the same internal simplified schematic which suggests 3 x Vbe temperature coefficient.  They are much higher bandwidth than what I built.  Would be useful if the offset can be managed.  Thank you for listing the part numbers.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: klugesmith on October 14, 2020, 05:10:27 AM
Hey, the higher speed Tx/Rx set in Hydron's link is recommended for 20 to 160 Mbaud.  Speed limitations according to distance and type of fiber are just from fiber dispersion and optical power budgeting.

How about transmitting a frequency modulated square wave, say 70 +/- 10 MHz?
Or a much smaller deviation, for smaller signal bandwidth.
Not sure about voltage-to-frequency IC's for the Tx end, but the Rx end might employ one of many low-cost FM or FSK receiver building block IC's. 

This is not intended to promote FM as better than simple analog optical power modulation with baseband signal.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: Hydron on October 14, 2020, 09:53:14 AM
Those two receiver parts appear to still be reasonably available.  Both have very loose DC offset specifications, 0.8 to 2.6V with no light input.  They have the same internal simplified schematic which suggests 3 x Vbe temperature coefficient.  They are much higher bandwidth than what I built.  Would be useful if the offset can be managed.  Thank you for listing the part numbers.
I think the anticipated use for the analogue output parts is with an AC coupling capacitor, so the DC offset is not a critical specification. When I'm a bit less busy I can probably do a quick check to see how much they actually vary with no optical input, though it would be at best over a modest indoor temperature range (i could blow some hot air over them I guess). In use you'd have to adjust gain/offset to account for loss in the fibre etc anyway (and try not to move the fibre once setup either!), so a fixed offset is less concerning than a heavily temperature dependent one.

You may find the ST connector (glass fibre) analogue output parts in old networking kit, though I believe that Mads had some issues desoldering them without damage (albeit without a vacuum desoldering iron). The ST stuff is easy to get pre-made patch leads for too (and for absurdly low prices). For POF you can use cheap optical audio cables (the thin ones, I think they're 2.2 or 2mm diameter) and put connectors on yourself, but duplex is more annoying as the duplex POF isn't readily available.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: davekni on October 14, 2020, 07:44:42 PM
"How about transmitting a frequency modulated square wave, say 70 +/- 10 MHz?"

Yes, I've thought about modulating a digital signal, with pulse-width or frequency or sigma-delta.  Digital links are much more available and to higher frequencies, 1Gbaud and up.  Might try such some day.  This was simpler.

"In use you'd have to adjust gain/offset to account for loss in the fibre etc anyway (and try not to move the fibre once setup either!), so a fixed offset is less concerning than a heavily temperature dependent one."

Yes, if offset is stable enough, it can be adjusted out.  The advantage of stable offset, as in my simple opamp receiver, is that only a single adjustment for gain is needed to compensate for fiber connection losses.  With accurate base-line LED power at the transmitter, that single gain adjustment can be made using zero signal input at the transmitting end.  That way, for measuring charge on top of my DRSSTC, I don't need to connect a reference signal generator to calibrate, then remove it without disturbing the fiber connections.  I can set everything up, then make a simple gain adjustment at the receiving end before running the coil.  (That will be next spring or summer.  Put my coil away for the winter.)
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: Mads Barnkob on October 18, 2020, 06:50:25 AM
Very nice project and thank you so much for sharing this!

Could the input divider have a voltage range selector between two sets of resistors without ruining the fixed offset? Could be a cheap alternative to differential probes on the inverter output, something that many people do not own.

Is there a reason beyond too-expensive or did-not-have-the-part-at-hand for using the IF-D95T receiver, but not the IF-E96 transmitter?

For you topload charge measurements I assume you set up resistors or CTs with a output in the same region of the gate drive voltage it was designed for? Leads back to my first question.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: davekni on October 19, 2020, 05:12:05 AM
"Could the input divider have a voltage range selector between two sets of resistors without ruining the fixed offset? Could be a cheap alternative to differential probes on the inverter output, something that many people do not own."

Yes, switched input attenuation is certainly possible, and might be reasonable.  I didn't do that because it can be tricky to compensate high divide ratios.  100:1 is likely fine without tweaking compensation, given the low 10MHz bandwidth.  By 1000:1 (for 100V/div), even 0.1pF of stray capacitance across the input resistor needs compensating.  Gets harder if there's significant capacitance to the middle of the input resistor.  So, for now, my plan is to build another board with different input resistance if I need different attenuation.  Then I can learn more about compensation without having switch capacitance to complicate issues.  Scope probes usually contain tiny ceramic circuits with resistance elements built (deposited) onto the ceramic with shaped electrodes on the opposite side to distribute capacitance along the resistance.

"Is there a reason beyond too-expensive or did-not-have-the-part-at-hand for using the IF-D95T receiver, but not the IF-E96 transmitter?"

Yes, I bought a couple IF-E96 transmitters.  They are higher bandwidth, but send much less light down the fiber.  LED technology has advanced rapidly, and IF-E96 appears to use old low-efficiency LEDs.  The trade-off is needing more amplification at the receiving end.  If I stayed with the same TLV3542 (100MHz gain-bandwidth), then higher receiver gain lowers bandwidth more than IF-E96 increases it.  Could use a higher-bandwidth opamp or more stages, but wanted simplicity and low noise.

"For you topload charge measurements I assume you set up resistors or CTs with a output in the same region of the gate drive voltage it was designed for? Leads back to my first question."

My plan for topload charge measurement is to use capacitors instead of resistors for sensing, so voltage will be proportional to charge rather than current.  Planning 20 paralleled C0G 100nF SMD caps to keep inductance low.  2uF total should keep the voltage within my +-20V range.  Measuring charge avoids issues with integrating scope traces with DC offset, and avoids issues of high peak breakout point current during ground-strikes.  Charge is great for topload voltage.  Down-side is that deriving current requires differentiating the signal, so amplifying high-frequency noise.  That's one reason I want to keep noise low, so use a bright LED.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: Mads Barnkob on November 05, 2020, 01:58:52 PM
Just came by a similar DIY optical transmitter / receiver on Youtube and thought it would fine into this thread

Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: Uspring on November 05, 2020, 05:21:49 PM
The idea to use capacitors for sensing is pretty, since Hydron and myself had some problems with non linearities of the scopes. Integration of current to get the voltages didn't work too well.
You have to be careful about DC components in the arc current, though. I don't expect them to be large, but if you e.g. measure hundreds of bursts in a row, they might accumulate. That might also happen in a single QCW burst. I guess, you'll use some input protection anyway.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: davekni on November 05, 2020, 08:04:50 PM
"You have to be careful about DC components in the arc current, though."

Yes, I expect the arc could strike ground more often at one polarity.  My plan is to have resistors across the capacitors, so they are really low-pass filters, not integrators.  For low repetition frequencies, the R/C time constant could have minimal affect during a burst, but mostly discharge the capacitors between bursts.  For higher repetition frequencies, I'll need to filter the resulting scope waveforms to undo the R/C time constant.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: davekni on February 21, 2022, 05:51:30 AM
Update:  Finished three sets of fiber scope probe transmitters and receivers.  Three transmitters are connected to three low-inductance 2.00uF capacitor arrays.  Measured voltage represents cumulative charge.  Initial test used my old bottom-fed SSTC.

Key point is that I need to add shielding to receivers.  Ground plane on ECB is not sufficient.  Receivers are close enough to coil to pick up electrostatic fields from the top-load.  I suspect the issue is mostly the photodiode part.  It is the highest-impedance receiver node, and the fiber receiver has a plastic housing extend farther above the ECB than other parts.

I'll add another post with scope data after adding better shields.  Quickie fix of grounded aluminum foil over receivers helped, but was not sufficient.

SSTC test setup in my dining room.  Three optical scope probe transmitters on top-load.  Three receivers at back right corner of the table are somewhat visible:
 [ Invalid Attachment ]

Zoom in to the three transmitters before covering with shield.  (Transmitter shielding appears sufficient, though I may improve that too.)
 [ Invalid Attachment ]

Transmitters shielded and covered with polycarbonate to further reduce risk of arcs to the wrong place:
 [ Invalid Attachment ]

Quickie shield (grounded foil) over receivers:
 [ Invalid Attachment ]

Arcs:
 [ Invalid Attachment ]
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: davekni on February 28, 2022, 05:47:41 AM
Required lots of shielding to work around TC voltages with low-voltage optical receiver circuitry.  Transmit side wasn't much of a problem, though I did add small inner shields on each transmitter:
 [ Invalid Attachment ]

Receiver shielding included a small section of grounded copper tape over each photodiode, along with a bit-larger rectangular shield that slides over each receiver and clips to ground of the coax cable to the scope.  Bit of a nuisance, as the rectangular shields need to be unfolded or slid off to access the on/off switches:
 [ Invalid Attachment ]

I tried initially without the above-shown grounded aluminum foil, and was surprised by the remaining issue.  One channel was fine and the other two still had some coupling from top-load.  Turned out to be the slightly-thinner coax used between receiver and scope on those two channels, roughly 60cm long soldered to receivers with BNC at scope end.  The thinner coax must have less shield coverage, even though it looks reasonable where stripped for connection.  Eventually I'll fix these cables and hopefully no longer need grounded foil around receivers and down to scope.

With the added shielding, an open input connection to one transmitter channel produces only about 2% of the real top-load charge.  Here's a plot with the breakout channel open:
 [ Invalid Attachment ]

Next are a couple real measurements of my old uninterupted bottom-fed SSTC.  CSV files for these two are zipped and attached at the end.  First is normal short wire breakout as shown in images of previous post:
 [ Invalid Attachment ]

This SSTC tracks rectified 60Hz line power.  Relay closes slightly before line zero-crossing.  Initial slightly-higher voltage (higher peak-to-peak charge) spot is when breakout starts.  Capture ends part-way into the second line half-cycle.

Other capture included here is with a short rounded breakout, copper pipe end-cap shown at left edge of picture:
 [ Invalid Attachment ]

As can be seen, initial voltage rises much farther before arc starts:
 [ Invalid Attachment ]

Secondary inductance is 75mH, which implies 31.8pF top-load capacitance.  Thus every micro-coulomb shown in the graph represents 1u/31.8p=31.4kV.  (Secondary coil capacitance lives within the field gradient of the top-load, which roughly matches coil field gradient, so contributes relatively little to total capacitance.)

One interesting feature that is more obvious here is the initial net negative breakout charge at arc start.  This is positive charge leaving the breakout, or negative charge entering the breakout.  Here's a zoom in to the initial breakout:
 [ Invalid Attachment ]

This same slight negative charge at arc start shows up subtly in the wire breakout captures too.  Polarity is consistent.  I think this is real and not some artifact of my optical probe setup.

I'm still not completely satisfied with probe performance.  With a low voltage sine-wave current source between any two channels, sum of the two is zero within <+-0.5%.  However, the high-voltage captures sum to ~7% of top-load charge.  That can be seen in the attached CSV files.  A fourth charge column is included, the sum of the previous 3 columns.  I've spent all weekend trying to figure out where that excess charge could be escaping, or where the high-voltage is inducing a 7% error.  As shown previously, direct coupling of HV to probe signals is now down to 2%.  I'll revisit this error some day when I get to measuring my DRSSTC.  Scope will be farther away then, and charge will be closer to full-scale.  (I'd sized capacitors and voltage division for +-40uC.  This SSTC is much less, especially breakout.  The tiny signal may be part of this issue.)

I'll make a new thread in the DRSSTC topic once I get to those measurements.  Decided to keep this post here, as it relates to required shielding for these optical probes to be useful at TC voltages.

Zipped CSV files:
 [ Invalid Attachment ]
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: Uspring on February 28, 2022, 06:50:39 PM
Congratulations for getting your probes into a useful state and thank you for the data set. One question: Are the voltages measured between secondary top and toroid, toroid and breakout and secondary top and breakout?

The difference in behaviour between a rounded and a sharp breakout point is somewhat expected but it is certainly nice to see this experimentally validated.
The fast initial rise in secondary voltage is a pretty good indicator of where secondary voltage would end up if the arc wouldn't break out. The arc almost seems to clamp secondary voltage. Also nicely visible is the difference between the top load voltages of the first and second line voltage half cycles, indicating quite a bit of heat and/or ions left over from the first half cycle at the beginning of the next.

Quote
(Secondary coil capacitance lives within the field gradient of the top-load, which roughly matches coil field gradient, so contributes relatively little to total capacitance.)

A good indicator for this is the ratio between secondary base current and secondary top current. Hydron has measured that for one of his coils and if IRC, the ratio was between 1.05 and 1.1. Similar quantities can be calculated with JavaTC. The actual ratio depends much on the size of the toroid.

Quote
One interesting feature that is more obvious here is the initial net negative breakout charge at arc start.  This is positive charge leaving the breakout, or negative charge entering the breakout.

The method of measuring charges instead of currents is a clever way of detecting this. TC arc currents aren't always sinusoidal, particularly during fast growth phases. This can be seen at the onset of arc current in the diagram of this post:
https://highvoltageforum.net/index.php?topic=117.msg827#msg827

A lot of high frequency currents during positive voltages can also be seen on this measurement: http://www.lod.org/gallery/electrum/techdata/waveforms.htm

Possibly that is due to the fact, that electron avalanches, which start near the arc or breakout point, are headed toward the breakout point and connect to it, if the polarity of it is positive. Avalanches near a negatively charged breakout point head away from it and don't connect to it, thus not allowing a charge flow to the point. In a fully grown arc, there is no need for conduction by avalanches, since there is enough heat to generate charge carriers. So the polarity dependency isn't seen there as much.
This is just speculation on what I know about the underlying processes. Corona, which happens before breakout and leads to arcs, has mostly been studied for DC or line frequency voltages. Not much is known for the RF range.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: Twospoons on February 28, 2022, 09:18:02 PM
This might be one of those cases where it would be advantageous to build your receiver into a small metal box  with a BNC on the side so it plugs directly into the scope BNC - no cables.
For power switching how about a small magnet and a reed switch? No need to open your shielding then.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: davekni on March 01, 2022, 05:48:25 AM
Quote
The difference in behaviour between a rounded and a sharp breakout point is somewhat expected but it is certainly nice to see this experimentally validated.
Yes, roughly as expected, but I didn't know ahead of time what actual voltage would be for either breakout.  Also as expected, voltage after breakout is similar.

1.05 to 1.1 sounds reasonable for a current ratio.  Effect on resonant frequency will be a bit less since current into coil capacitance doesn't pass through all the coil's inductance.  I haven't tried to model or measure it more accurately.

Thank you for the thoughts about negative arc charge.  I presumed it would be something more about the breakout point surface, electron emission work function or secondary electrons emitted by electron or ion impact from adjacent arc air.

Quote
This might be one of those cases where it would be advantageous to build your receiver into a small metal box  with a BNC on the side so it plugs directly into the scope BNC - no cables.
Yes, that would be a clean solution.  My mechanical abilities are minuscule and slow.  That's why I build with milk crates, foil, etc.  I'm planning to just replace coax cables on the two problematic receivers.

Quote
For power switching how about a small magnet and a reed switch? No need to open your shielding then.
Interesting idea.  Given how infrequently these optical probes are likely to get actual use, I'll probably not bother with any cleaner solution.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: Uspring on March 04, 2022, 05:40:42 PM
Quote
I presumed it would be something more about the breakout point surface, electron emission work function or secondary electrons emitted by electron or ion impact from adjacent arc air.
Thinking about this a bit more, that might very well be. Negative corona seems to have an advantage over positive one due to the possibility of photoeffect on the cathode surface, which isn't possible for positive corona. A negative charge cloud might build up during lower voltages due to corona and then vanish, when electrons are generated by hot plasma. That is much more efficient in freeing electrons and differences in polarity due to surface photo effects become insignificant.
So the jump in charge might not be due to the onset of corona but due to its end, i.e. replacement by plasma. I've plotted a sliding average of 10 data values from your rounded point breakout charge data. 10 data points correspond to roughly a period, so this will suppress the TC frequency.
 [ Invalid Attachment ]
It's quite an amazing jump, considerably more than a full cycle worth of charge going into and out of the breakpoint. For the sharp wire breakout, the charge jump is about 10 times smaller. It would be interesting to know the value of your bleeding resistor.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: davekni on March 04, 2022, 07:41:11 PM
Quote
It's quite an amazing jump, considerably more than a full cycle worth of charge going into and out of the breakpoint. For the sharp wire breakout, the charge jump is about 10 times smaller. It would be interesting to know the value of your bleeding resistor.
Yes, I was surprised at the jump, which is why I posted about it.  The bleed resistor (probe input divider resistance) is 30.7k.  With 2uF charge accumulation capacitors, time constant is 61.4ms.  Due to the small amplitude, I had the scope on AC-coupled to remove final opamp DC offsets.  That time constant is 16ms, so the more significant one here.  Still longer than the apparent time-constant in charge data.  (Designed this setup for my DRSSTC, so these SSTC charge values are comparatively small, requiring 1mV/div for the breakout channel scope setting.)
Thank you for the possible explanations.  BTW, as you can see in your low-pass-filtered data, total charge was consistently reversed to slightly positive by the end of the first half-line-cycle 8.3ms later.
Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: Hydron on March 05, 2022, 12:18:25 PM
Great to see this going - analogue optical fibre based is definitely a more accessible setup than what I did before!

I'm interested in why you're using 3 channels - are you measuring secondary-probe, probe-toroid, and probe-breakout currents? I only had 2 channels available (though I could actually put all 4 in the topload, but without any left for primary/secondary-base currents), so I had to find out toroid current by subtracting breakout current from the secondary current flowing in.

Good idea measuring current across a capacitor btw - will need to have a go with this myself (I'm finally gonna be in the same country as my big coil again in a few weeks, and have been prepping a UD3 controller etc to be ready for some testing of my own ;D )

Title: Re: DIY DC-10MHz optical-fiber-isolated scope probe
Post by: davekni on March 06, 2022, 12:48:15 AM
I've started a new thread for continued discussion of top-load scope measurements:
https://highvoltageforum.net/index.php?topic=1950.msg14588#msg14588
New more-accurate data are posted there.

I was going to be done with this, but 7% error was too troubling.  So I spent another 1.5 days or so experimenting.  First, changing to RG59B/U coax for two probes (to match the original one) solved the receiver issue.  No need for aluminum foil there - just the little copper-foil boxes around fiber receivers.  Still 7%, though.  Just eliminated foil.

Grasping for straws a bit, I even checked for any possible issue with electric fields modulating attenuation of the plastic optical fibers.  Nothing there.  Considered that perhaps the steel housing for the three transmitters had such a thin skin depth at 103kHz that it wasn't a reasonable shield.  Covered it with copper tape joined with solder.  No difference.  Tried again placing aluminum foil over the entire top-load center connected to top-load.  That makes it worse!  That bit is still a puzzle.

What did help was bending the little foil shields over each probe a couple mm to better cover the ECB.  The breakout channel (channel 3) has always been relatively clean compared to the other two (secondary coil and top-load).  Noticed that channel 3 shield placement was slightly different.  Bending the other two to match reduced total error from 7% to 3.5%.  Haven't been able to make any further improvements, however.  Error appears to all be transmitter shielding related, but adding more shielding often makes no difference or makes it worse.  Still a bit confused about this.  Trying really hard to fight my perfectionist tendencies and be happy with 3.5%.

Quote
    I'm interested in why you're using 3 channels - are you measuring secondary-probe, probe-toroid, and probe-breakout currents? I only had 2 channels available (though I could actually put all 4 in the topload, but without any left for primary/secondary-base currents), so I had to find out toroid current by subtracting breakout current from the secondary current flowing in.
Yes, I spent some time studying and analyzing your results.  Thank you for posting.  Your work is what inspired this project.  The reason for 3 channels is to verify accuracy.  3 provides redundancy, allowing me to sum them and see how close they are to zero.  That is where I have the remaining 3.5% error.  Without this redundancy, I might not have noticed any error and not known shielding improvement was needed.  My very first test (before any shielding improvements) summed to 15% of top-load charge.  Definitely would not have wanted to continue presuming accuracy when reality was 15% different.  (Error appears to be roughly split between secondary and top-load charge, both off by ~1.5-2% now.  Error is at the same phase, so not canceling as they should, yielding ~3.5% total error.)
SimplePortal 2.3.6 © 2008-2014, SimplePortal