Fast optocoupler shutoff

[matthuszagh]

You still haven't said how many amps it needs to pass.

When the input blocking capacitor is charged and the bidirectional MOSFET switch is closed the current will be miniscule. nV or uV into the high JFET input impedance. When a power supply is turned on or off (or hotplugged), the current could be quite large while the blocking capacitor charges or discharges. I don't know exactly how much, but this should be able to handle any level of current while the 10u blocking capacitor charges. When the MOSFET switch opens, the capacitor will (dis)charge much more slowly through a 100k load resistor to minimize loading on the DUT. The low MOSFET Rds(on) is more about limiting thermal noise contribution from the switch (though I could probably go up to about 1ohm total and that would be fine). It's also a byproduct of using high power / voltage MOSFETs.

How about an SCR in series with a depletion mode MOSFET? The SCR and MOSFET can be coupled to the drivers via pulse transformers. The SCR will need a short pulse to turn on and the MOSFET a pulse to turn it off for long enough for the SCR to latch off.

Hm, I'm not sure I totally follow. Is this an SCR across the line or in series with it? Either way, the input can positive or negative and the switch needs to conduct both ways. Also, I don't want anything shorting out the input for any appreciable amount of time.

[matthuszagh]

If 60 Ohm is way to high this means a really low noise amplifier.
One may consider having something like MOVs at the input to catch ESD or other high voltage peaks well higher than the normal DUT and than rely on an inductor (possibly physically relatively large, like a flyback transformer) to slow down the rise in current, before the protection can kick in.  For a short time the LED in the OK should be OK for maybe 100 mA. So with lets say 100 V and 1 mH one would have 100 A/ms and thus 1 µs to reach 100 mA. With higher current to the LED the turn on time is expected to go down. So even "normal speed" optocouplers like TLP290 can turn on in less than 500 ns.
A fast rising voltage would also need extra current to discharge the gate. So one may want fast switching fets for not too high a voltage.
A problem could be that the clamp voltage from the LEDs may reach 1.5 or maybe even 2 V at higher current.

I was planning on using an SP00R6 or similar as high speed protection until the switch kicks in. If the switch is fast enough, I think this combination should be enough to avoid damaging the preamp. Though this is purely speculation still and I plan to test it. I'll reconsider inductors too, to limit current into the preamp.

But, with an isolated supply and the isolators that permits, it should be possible to kick in the switch very quickly and I wonder how much additional protection I'll need (though I'll definitely still keep the TVS). For example, I'm looking at the ISO7710. That gets me back into the 10ns range, though a bit more when I factor in the comparators, charging and discharging the MOSFET gates, etc. The TVS clamps around 2V. It seems unlikely that a <50ns pulse at this voltage would damage a JFET.

[T3sl4co1l]

The question of how much speed I actually need is a bit tough to answer. I'm trying to protect a sensitive, downstream preamplifier. The first stage is some n-channel JFETs in parallel biased to 0V. I need to be able to handle hotplugging power supplies (DUT) up to +/- 100V to the input. There are at least two parts to this challenge: protecting the JFETs and not shorting out the DUT for too long. Eventually I'll do some surge testing of these JFETs, but otherwise its difficult to get data on how much power they can actually dissipate for very short periods (ns to us durations). Anyway, generating an isolated supply should make this whole thing a lot easier. Thanks for the suggestion!

Finally some information!

Please show the system connection (how, and what, they are hotplugged to, in whatever combinations are relevant) and the amplifier schematics.

Tim

[matthuszagh]

Please show the system connection (how, and what, they are hotplugged to, in whatever combinations are relevant) and the amplifier schematics.

I've attached a block diagram. The input will not always be a high voltage power supply, but it can be and that's the worst case in terms of damage protection. The purpose of this preamp is to amplify very weak low-frequency noise (<1Hz to 1MHz) in the presence of a potentially large DC offset.

I've shown an LED+photodiode in the diagram but I'll actually use something like the ISO7710 and an isolated supply.

The switched in load resistor is 100k, as already mentioned. The exact value isn't super critical, but it needs to be large enough not to load down low current, high voltage power supplies.

Both blocking caps are 10u high-voltage WIMA MKP 4.

I have not finished designing the preamp, but for the sake of damage protection the input stage will look like that. The series resistor may or may not be used, but if used it will be very low valued (a few ohms, max).

[T3sl4co1l]

Zero source resistance?

No filtering?

What about a startup delay on hot-plugging to do the discharge automatically?

Tim

[matthuszagh]

Zero source resistance?

No filtering?

From the power supply? Yes, it's a normal (real) power supply. But there's no stipulation on what power supply can be plugged in other than the peak value stay within the voltage range. Typically it will have bypass caps at the output and all power supplies have some source resistance... But, since any supply in that voltage range can be plugged in I don't want to rely on source resistance as part of the protection circuit.

What about a startup delay on hot-plugging to do the discharge automatically?

I'm not sure how this would work. Can you elaborate? This is not a fixed circuit. This is not always plugged into a power supply. This is a protection circuit for a preamplifier that can be used for many different things as long as the application stays within the voltage limits. One example is measuring the noise of a power supply. Another is measuring the noise of voltage references. Hot plugging and unplugging are supported. So is turning on the supply after it's plugged in. I don't know how a startup delay would address all of these. Maybe this modified diagram makes it more clear.

[matthuszagh]

Zero source resistance?

No filtering?

From the power supply? Yes, it's a normal (real) power supply. But there's no stipulation on what power supply can be plugged in other than the peak value stay within the voltage range. Typically it will have bypass caps at the output and all power supplies have some source resistance... But, since any supply in that voltage range can be plugged in I don't want to rely on source resistance as part of the protection circuit.

Oh, I think I misunderstood your question. I now think you were asking whether I will add any series resistance or filtering to the input. Probably not. I'm not eager to increase the equivalent input noise of the preamp unless I need to. As it is, even a few ohms will surpass the equivalent input noise. I also don't want to filter the noise of the DUT since that's precisely what I'm trying to measure.

[T3sl4co1l]

From the power supply? Yes, it's a normal (real) power supply. But there's no stipulation on what power supply can be plugged in other than the peak value stay within the voltage range. Typically it will have bypass caps at the output and all power supplies have some source resistance... But, since any supply in that voltage range can be plugged in I don't want to rely on source resistance as part of the protection circuit.

Between the supply and preamp I mean.

1MHz bandwidth seems to imply you can put a filter in there, and not need to clamp things for at least some hundreds of nanoseconds.  And the filter will necessarily introduce impedance, making the inrush current well-defined.  But it also needs to be terminated and a zero-impedance source doesn't exactly work with that (termination can be on the amp side, but the source impedance still preferably needs equalization -- a constant resistance filter is best).

Also if the source is always zero ohms, a JFET doesn't seem the lowest noise option; a nice thick BJT should be able to do better than a JFET array.  Lowest noise may not be the top priority however, so that's fine.

I'm not sure how this would work. Can you elaborate? This is not a fixed circuit. This is not always plugged into a power supply. This is a protection circuit for a preamplifier that can be used for many different things as long as the application stays within the voltage limits. One example is measuring the noise of a power supply. Another is measuring the noise of voltage references. Hot plugging and unplugging are supported. So is turning on the supply after it's plugged in. I don't know how a startup delay would address all of these. Maybe this modified diagram makes it more clear.

Well your first claim was hotplugging so if that's the only thing you're worried about, that could be worked around in many ways.  You didn't say it needed to do anything else, so we're free to assume anything and everything until said otherwise...

If it's not the only thing, maybe you could make a hotpluggable interface module and only use it to test supplies with that?  More steps, but makes things more flexible / foolproof.

No idea; we're not telepathic here, after all.

Tim

[matthuszagh]

Between the supply and preamp I mean.

1MHz bandwidth seems to imply you can put a filter in there, and not need to clamp things for at least some hundreds of nanoseconds.  And the filter will necessarily introduce impedance, making the inrush current well-defined.  But it also needs to be terminated and a zero-impedance source doesn't exactly work with that (termination can be on the amp side, but the source impedance still preferably needs equalization -- a constant resistance filter is best).

Yeah good point. The preamp should achieve a noise floor of about 300pV/rtHz and needs to measure noise down to about 2nV/rtHz for power supplies and lower for e.g., transistor noise measurements. That doesn't leave a ton of margin for additional noise sources as I'd like to stay far enough below the DUT noise to maintain accuracy. There are other tricks I'll emply here; mainly, two preamps and cross correlation but that imposes its own challenges and has its own drawbacks so I'd like to get the preamp noise as low as possible first.

The protection circuit will be able to be switched out for certain measurements such as for 0-DC sources where it's really not needed. Such a protection circuit is probably also overkill for biasing arrangements used in transistor noise measurements. Anyway, maybe there's some headroom for adding a filter with a small series resistance if this protection circuit will be exclusively used for power supply and voltage reference measurements. I'll investigate; thanks for the idea.

Also if the source is always zero ohms, a JFET doesn't seem the lowest noise option; a nice thick BJT should be able to do better than a JFET array.  Lowest noise may not be the top priority however, so that's fine.

This isn't necessarily true. While the DUT might be low impedance, the blocking cap at the lower frequency range of operation isn't. I specified <1Hz as the cutoff, which I appreciate is a bit vague (this should go down to at least 100mHz, preferably all the way down to 1mHz). But, using 1Hz as the frequency that 10u blocking cap looks like a 16k impedance. The BJT current noise through that cap will swamp the equivalent input noise. JFETs do much better in this regard. And, while you're right that JFETs don't do quite as well as BJTs for true low impedance sources, they can do remarkably well and are really not all that far off the performance of BJTs from what I've seen. I could, of course, increase the blocking capacitance, but this quickly becomes impractical.

Well your first claim was hotplugging so if that's the only thing you're worried about, that could be worked around in many ways.  You didn't say it needed to do anything else, so we're free to assume anything and everything until said otherwise...

If it's not the only thing, maybe you could make a hotpluggable interface module and only use it to test supplies with that?  More steps, but makes things more flexible / foolproof.

No idea; we're not telepathic here, after all.

I apologize for being vague - that really wasn't my goal. And I do appreciate the suggestions I've received.

As for hotplugging modules, I guess I just don't know enough about those so I'll spend some more time looking into it. The commercial hotplugging modules I've seen have all been for a single polarity supply and for a limited voltage range (makes sense, given the typical application). I suppose the principles underlying these modules could be used to make a custom one. But, looking at the ADM1177 for instance, doesn't make me think that my current topology is wrong. They use a single external MOSFET whereas I need a bidirectional switch. I monitor the voltage and trigger the shutoff with high-speed comparators, whereas they use a sense resistor and ADC. I like my approach better, since it's simpler, should be faster and doesn't require adding a sense resistor.

[peter-h]

A photo diode optoisolator will be fast. A photo transistor isolator has a much higher gain but is much slower - largely due to Miller Effect (which is why the better optos have the base coming out on a pin so you can decouple it.

Hot plugging needs to be done carefully because even a 100nF capacitor will short out the supply rail for some nanoseconds or more and crash everything upstream

[Zero999]

You still haven't said how many amps it needs to pass.

When the input blocking capacitor is charged and the bidirectional MOSFET switch is closed the current will be miniscule. nV or uV into the high JFET input impedance. When a power supply is turned on or off (or hotplugged), the current could be quite large while the blocking capacitor charges or discharges. I don't know exactly how much, but this should be able to handle any level of current while the 10u blocking capacitor charges. When the MOSFET switch opens, the capacitor will (dis)charge much more slowly through a 100k load resistor to minimize loading on the DUT. The low MOSFET Rds(on) is more about limiting thermal noise contribution from the switch (though I could probably go up to about 1ohm total and that would be fine). It's also a byproduct of using high power / voltage MOSFETs.

How about an SCR in series with a depletion mode MOSFET? The SCR and MOSFET can be coupled to the drivers via pulse transformers. The SCR will need a short pulse to turn on and the MOSFET a pulse to turn it off for long enough for the SCR to latch off.

Hm, I'm not sure I totally follow. Is this an SCR across the line or in series with it? Either way, the input can positive or negative and the switch needs to conduct both ways. Also, I don't want anything shorting out the input for any appreciable amount of time.

What input blocking capacitor?

What are you protecting?

My post assumed you're protecting a circuit against over-voltage on the DC supply rail, but now it appears you're talking about an AC signal.

Post a full schematic.

[matthuszagh]

What input blocking capacitor?

What are you protecting?

My post assumed you're protecting a circuit against over-voltage on the DC supply rail, but now it appears you're talking about an AC signal.

Post a full schematic.

This information has already been provided. Please see my recent explanations to T3sl4co1l and the associated block diagram. I haven't posted a full schematic yet since I'm changing the isolated MOSFET drive portion.

This is not a DC supply rail over-voltage protection circuit. It's a protection circuit for the input of a very low noise preamp that can be exposed to inputs with DC voltage anywhere in the range of -100 to +100V. The preamp 1st stage is a 0-bias n-channel JFET in a common source configuration (or a variation on that, but for damage protection that's accurate).

[matthuszagh]

A photo diode optoisolator will be fast. A photo transistor isolator has a much higher gain but is much slower - largely due to Miller Effect (which is why the better optos have the base coming out on a pin so you can decouple it.

Noted, thanks Peter. I'm currently looking at the ISO7710, which is not actually an opto - it uses isolation capacitors instead. But it's quite fast: around 10ns.

Hot plugging needs to be done carefully because even a 100nF capacitor will short out the supply rail for some nanoseconds or more and crash everything upstream

Is a 10-50ns short really that bad? I figured the bypass caps would be ok supplying this. But if that's really very bad for the power supply attached to the input I'll need to rethink things.

[matthuszagh]

There's an additional difficulty with the isolated power supply in that these switching power supplies are really noisy. For example, I've been looking at the DCP020505 to power the output side of the ISO7710. I'm worried about that making its way into the signal path.

[T3sl4co1l]

So what's the isolation about then? It's all common ground...

Tim

[shapirus]

There's an additional difficulty with the isolated power supply in that these switching power supplies are really noisy. For example, I've been looking at the DCP020505 to power the output side of the ISO7710. I'm worried about that making its way into the signal path.

What is the supply current of the (sub)circuit that the isolated power supply will be powering? If it's not too high, it might be quite possible to utilize heavy RC filtering of its output with a large value resistor and still get sufficient voltage after the filter to power the circuit. Or, use a 5V->12V isolated converter, put an RC filter (multi-stage if necessary) at its output, then use a 5V linear regulator to stabilize voltage. Use proper board layout to minimize possible noise coupling.

[matthuszagh]

So what's the isolation about then? It's all common ground...

The drain of the first MOSFET in the bidirectional switch (Q5 in the schematic attached, or first post) can go to +/- 100V. If I don't isolate the gate drive circuitry, the drain-gate voltage can greatly exceed the rating (typically 20 or 30V).

It seems I've caused a lot of confusion by not showing the full protection circuit schematics. I've attached a preliminary version of this schematic. This still uses the old optocoupler method, whereas I'm now considering an ISO7710 and DCP02 isolated supply. Remember, too, that "out" is connected to the preamp, which looks like an n-channel JFET in a common source configuration with 0-bias (low clamp voltage TVS diode across it too). "in" is connected to whatever DUT is being tested. That may or may not be a high (or very negative) power supply, but that's the worst case from a damage protection scenario. I'm sorry for the confusion I've caused and I appreciate all the assistance I've gotten.

[matthuszagh]

What is the supply current of the (sub)circuit that the isolated power supply will be powering? If it's not too high, it might be quite possible to utilize heavy RC filtering of its output with a large value resistor and still get sufficient voltage after the filter to power the circuit. Or, use a 5V->12V isolated converter, put an RC filter (multi-stage if necessary) at its output, then use a 5V linear regulator to stabilize voltage. Use proper board layout to minimize possible noise coupling.

Good idea! The ISO7710 specifies a typical 1.7 mA current draw, so very minimal and there's definitely headroom for RC filtering. I'd been planning to use a high-PSRR linear regulator (probably an LT3042), but adding RC filtering seems like an easy way to get much more noise rejection.

To further reduce noise, I was going to use the pi filter mentioned here: https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/143862/dcp020515d-output-cap-to-reduce-output-voltage-ripple and the Y-cap across the negative supplies. If I go with an LT3042, the datasheet is pretty particular about proximity to the input switcher. Basically, they say to keep it about an inch or a little more away and get rid of the input bypass cap in order to maximize PSRR. This may take a little trial and error to get right. I was also planning to put these as far away from everything else as possible on the PCB and keep other conductors away to minimize capacitive coupling.

I'm a little worried about noise being injected back into the input of the switcher and making its way into more sensitive circuitry that way. I don't know how much of a concern that normally is. I suppose if that is a problem I could add a linear regulator right at the input too. Unfortunately the DCP02 has a minimum load current and so RC filtering the input might be a little tricker.

I'd love to use the original optocoupler (VO1263) as the power supply, but unfortunately I'd need way too many to provide enough current.

[matthuszagh]

Looking at this more, it seems there are better isolator choices. In particular, dedicated isolated gate drivers. These can source/sink enough current on their own and I could get rid of the P-channel JFET fast shutoff. And they still have low quiescent currents so I can still use RC filtering to suppress noise from the isolated switcher.

[T3sl4co1l]

You're still locking yourself into one limited solution...  What's wrong with:
- Series limiting resistor e.g. 10R into bidirectional TVS; TVS capacitance can be used as part of filter capacitance to set bandwidth.
- Current limiting depletion MOS; has low resistance at signal levels, zeners or clamp diodes protect amp.
- Plain old polarity protection circuit like you'd use on a battery, chain two complementary ones and bias them to +V/-V so they pass from (-V + Vgs(th)) to (+V - Vgs(th)) and shut off outside there.
- Classic, biased diode-bridge current limiter; bias can be from HV BJTs between supply rails.

Tim

[Siwastaja]

Yeah I was planning to put a TVS diode downstream.

You realize a TVS has a lot of capacitance which continues supplying the load?

For the idea of turning a load off in nanoseconds, it's absolutely vital to know what the load is. You may need to modify the load, for example short any capacitors in it. Large short circuit currents in nanoseconds then mean high dI/dt, which means problems with inductance.

Unless it's top secret, tell us what you are actually doing.

[inse]

What about a dedicated optocoupler to turn off the power MOSes fast?

[shapirus]

I'm a little worried about noise being injected back into the input of the switcher and making its way into more sensitive circuitry that way.

Check https://www.analog.com/media/en/training-seminars/tutorials/MT-031.pdf. You may find some applicable ideas on the isolation of noisy components from sensitive ones there.

[matthuszagh]

You realize a TVS has a lot of capacitance which continues supplying the load?

No, it doesn't. I'm planning to use a SP00R6, which has 0.2p capacitance. Just because some TVS diodes have a lot of capacitance does not mean that all do.

[T3sl4co1l]

But that's rated for 2A (8/20us) and 12kV (ESD, contact). Can you limit currents accordingly?

With 50 ohms in series, you'd pretty much be done.  Not quite, as 10uF * 50R = 500us not 20us, but a bigger diode could be used trivially without affecting bandwidth.

I don't get the series resistance at the gate, when it can do much more upstream of the transient protection.  This isn't rocket science, it's basically solved passively this way.

Tim