DIY esd gun.

[newbadboy]

Just for fun and for learning intensions.
It will handle 500V-7kV with 500V steps and will only have positive output polarity. It will have single shot and 10times @ 1Hz modes.

[newbadboy]

more pics

[newbadboy]

more pics

[newbadboy]

more pics

[mcinque]

this is interesting. do you want to share any schematics?

[niconiconi]

Very cool. How did you make the discharge tip for the gun? Did you make your own drawings and ask a machine shop to have them custom made, or was it something else? I always want to make an industry-standard ESD gun but I don't have any mechanical knowledge.

[T3sl4co1l]

Huh, what's the dielectric strength of FDM like, anyway?  Depends on fill and tuning?  Is porosity ever low enough to use reliably (insulating large areas)?

Tim

[Zero999]

I've used a piezoelectric igniter before. The ones from disposable cigarette lighters work quite well. Of course it's just a rough test, not calibrated in any way.

[joeqsmith]

Nice looking project.  It will be interesting to see the waveforms once you have it working.   

[niconiconi]

Nice looking project.  It will be interesting to see the waveforms once you have it working.

Testing or calibrating the air-discharge waveform would probably be the most tricky one... IEC 61000-4-2 requires the use of special lab room setup with ground planes, resistors, and shielded cable going into a large shielded box for all the instrumentation... Fortunately for a quick qualitative pre-compliance test instead of traceable quantitative measurements, you probably don't need it...

Transmission Line Pulse is an alternative component-level ESD test method precisely due to the unpredictability of gun-type tests...

[SiliconWizard]

I've used a piezoelectric igniter before. The ones from disposable cigarette lighters work quite well. Of course it's just a rough test, not calibrated in any way.

Same here.

[Karel]

I use a Milty anti-static gun for (rough) testing. I extended the electrode a bit so that it's more exposed and I
attached a grounding wire to the trigger. It's pretty effective.

[Swainster]

My experience with HV (at a few 10s of microamps) and FDM in prototype construction is that you can treat the plastic as a perforated for HV purposes I.e. it goes straight through it. You need to rely on the air gap (taking into account creepage) for any insulation. For that matter, even bog standard injection moulded plastic has a significant chance of failure over time, due to imperfections/air bubbles picked up in the shot. Silicone tubing can be used to supplement things though. Surprisingly (at least to me) vinyl tubing is absolutely rubbish at HV isolation.

[fcb]

Good luck - be interesting to see the results.  Have you got a fast scope to check rise-times?

Have you used a mercury wetted relay? Looks like a Pickering 67 series - dry relay? What is your target max discharge voltage?  Discharge tips are also typically a ball not a point (more difficult to turn!).

[newbadboy]

Huh, what's the dielectric strength of FDM like, anyway?  Depends on fill and tuning?  Is porosity ever low enough to use reliably (insulating large areas)?

Tim

I have no idea. just hoping (and believeing) that having alot of space and infill will work.

[newbadboy]

Very cool. How did you make the discharge tip for the gun? Did you make your own drawings and ask a machine shop to have them custom made, or was it something else? I always want to make an industry-standard ESD gun but I don't have any mechanical knowledge.

I have a friend that has a mechanical workshop that made it for me

[newbadboy]

I've used a piezoelectric igniter before. The ones from disposable cigarette lighters work quite well. Of course it's just a rough test, not calibrated in any way.

Same here.

As mentioned before, it will only handle contact

[newbadboy]

Good luck - be interesting to see the results.  Have you got a fast scope to check rise-times?

Have you used a mercury wetted relay? Looks like a Pickering 67 series - dry relay? What is your target max discharge voltage?  Discharge tips are also typically a ball not a point (more difficult to turn!).

I have not measured it yet, waiting for my HV diff probe, my scope is fast enough but the diffprobe is 100MHz so it will probably limit the measurement capabillity. Yes i use the pickering relay due to cost factor. It will be driven out of specc but should work. Max 7kV is the goal.

The ball discharge tip are for air disharge, i will only use it for contact which then is the point tip.

[newbadboy]

I did a small startup of the HV part.

First attempts kind of failed. When injecting 6V of 10kHz the secondary output was 170Vdc. Also the LDO powerin the primary went down all the time like if it went into protect mode.

I guees i will need to tweak it a bit more

[newbadboy]

this is interesting. do you want to share any schematics?

Sure, will upload soon

[EggertEnjoyer123]

I'd suggest using a similar circuit (but with a capacitor connected in parallel with the inductor coil). You'll probably need to experimentally determine its value.

The way these transformers usually work is that you put current through the primary, then you quickly shut it off and the collapsing magnetic field makes a massive voltage on the secondary. The circuit provided (which I stole from the chinese seller's website) does that and it should work (with the capacitor added, and maybe a few protection diodes to make sure you don't blow up your transistor). These transformers will not put out anywhere close to the rated voltage if you drive them normally (with a function generator). Voltage feedback can be accomplished using a resistor divider and a comparator, and it can be used to change the duty cycle (maybe consider using a dedicated switcher IC instead, like the UC3845). Just make sure everything is stable, which is easier said than done.

[newbadboy]

I don't have access to my laptop so i cannot share the full schematic at the moment.

The drawing you attached is basically what i'm doing. At least what i believe. I have a transistor down to ground with 10kHz square wave. The upper part is then supplied with a variable LDO to ba able to set out outputvoltage by altering the input voltage. I don't really have any controll of how energized the coil is in reality but that might be the issue here

of course i have protection diodes also

What am i missing

[fcb]

You need feedback from the HT generator to detect both sparking and to regulate/set the HT.

I'd probably have gotten higher than 170V before going to the bother of designing a case.

Designing HT supplies is non-trivial, you'll probably want to pot things in clear silicone/RTVwhen you get higher. Also, when a discharge happens you can easily nail the FET/transistor, so look at adding protection in the form of over-current and over-voltage.

If you are going single polarity then LOPT's might be better to start with, they are designed to work around 16KHz and have the output rectifier diode(s) built in and are designed to drive a capacitive load (eg CRT).

Also, if it's contact only, then I'd be tempted to just use the relay to discharge/safe the probe-tip (via a few hundred K) when the trigger is released.

[newbadboy]

 My intent is to measure what the LDO needs to output to set certain secondary voltages. Feedback should not be needed?

I hope to be able to step up the out output with 500V steps from 500 till 7kV. This by regulating the primary voltage. In reality the ldo is controlled by a mcu and a digital pot.

There is a guy making a esd gun on youtube with the exact same transformer as i use and he has no problems getting 10kV out of it. He is using the 555 as pulse generator and the LM317 as primary voltage source to set the output level.

Yes adding more protection to FET is probably a good idea.

I'm not sure what LOPT stand for, sorry

Discharging the pulse thru that high resistance will not really stress the DUT enough.

[fcb]

Feedback allows you to:
1. accuratley set the output voltage
2. know where you are in the charging cycle
3. use this to detect breakovers in your DUT

LOPT - "line output transformer" (CRT monitor/TV part)

Pulse discharge network for contact discharge testing is specified in the various standards as is tip design etc..  You want to try and get close to this as it will mean that any testing you do will put you ahead of the game when it comes to testing in a proper (as in $$$) EMC lab.

As for FET protection, you will obviously need to allow the transformer primary to fly fairly high, so you could protect this with anything from a transzorb, zener, RC network or even a some sort of pumped-clamp rail.