Problems with homemade X-Ray power supply and grounding

[Luca c.]

Hello, I'm making this post because I would like to hear your opinion about a problem I encountered today when I tested my homemade x-ray power supply:
I start by describing the schematic: it uses a switched mode power supply to transform the 220v to 24v, then regulator 1 makes it possible to regulate the voltage between 0 and 23v, for this test, I used 7v (this gives me around 30kV at the output of the multiplier), then the output goes to a ZVS driver which feeds a flyback that gets multiplied with 3 stages CW voltage multiplier, the output then goes to the x-ray tube, which itself is powered by a second regulator that emits few volts and around 2-3 A for the filament of the x-ray tube.
Schema2_230507_122610.pdf
while testing (see video below) the high voltage is turned on by pressing the push button with a stick as shown in the video, you can hear the high voltage hissing. No x-ray is seen on the fluorescent paper in front of the tube (this might be due to low high voltage or low current in the filament, but is not the problem) then after some time you can hear an explosion, that is some sort of component dying.

/>I checked and the main power supply doesn't work anymore, the lower left transistor (D13009K) haves some black residue near the base and by testing it with a multimeter it seems to be the problem.

I have checked the rectifier and it is ok, the model of the transformer is S-360-36.
The two regulators are reg 2: "ZK-4KX" , and reg 1: "DPS5015".
I didn't ground the negative side of the power supply and the other part of the circuit because I thought that there may be some problem with connecting the earth ground (that is connected to the chassis of the power supply) to the negative polarity of the different voltage regulators, mainly because by using a multimeter I have seen that the negative (input/output) side of the regulator doesn't have continuity so by connecting them I thought I might cause some problem (note that the earth ground of the chassis is not connected internally to the negative side of the switch mode power supply).
So this is my question: the reason for the transistor exploding is that I didn't ground the power supply? I can only think of this as an explanation, but it seems unreasonable, mainly because the power supply has the chassis earthed so it should act as a Faraday cage and protect the internal component from any static electricity.
Do you think that I should ground only the power supply, the supply and the 2 regulators and the CW multiplier or only part of the component?
if you have any other comments or suggestions I would love to hear that.

[davekni]

So this is my question: the reason for the transistor exploding is that I didn't ground the power supply?

Probably yes.  Corona discharge or even ion wind from the HV (30kV) output will at least in part go to surroundings (ground).  That current must return somehow.  Will force negative side of HV supply below ground until isolation insulation within supply fails.  That failing insulation path from supply secondary to primary may have confused supply control electronics.

If I understand failure timing, it occurred while button was pushed in, not as button was just being pushed in.  If just as button is pressed in, then it could be the common ZVS startup issue overloading supply.  However, supply presumably has over-current protection that would keep it from frying even if load current becomes too high.

Do you think that I should ground only the power supply, the supply and the 2 regulators and the CW multiplier or only part of the component?

Supply output as a minimum.  If other following voltage regulators are not isolated, then they shouldn't be bothered by HV return current through them to grounded supply output.  If other regulators are isolated, then they also need grounding.

[曹靖]

I will provide you with a schematic diagram for modifying the voltage regulating switch power supply. Perhaps it will be helpful for you. A few months ago, I referred to this schematic diagram and made myself a 1KW adjustable switch power supply with an output of 0-150V. I hope it can help you now

[Luca c.]

UPDATE: thank you very much Dave, today I had the time to make the suggested modifications, grounding everything (even the CW multiplier) and it worked flawlessly, I had the time to make a brief test and take only 1 picture but there it is:

I used "DENTAFILM ECO 30" a dental x-ray film with the developer inside.
the exposure lasted around 50seconds with a current that i am not really sure about, the analog A-meter was off-scale, the maximum it could read is 5mA, but I have some doubt it could be this high, because the DC regulator that feeds the zvs was saying that only 15W was being consumed, so I have to do further testing to determine who's lying. The voltage, should be around 40kV.

this is the subject I x-rayed, it is a high voltage resistor, you can see the darker spots of the resistive material that is visible on the x-ray photo, the distance was about 3cm from the window.
I will make more tests and experiments, in the future I will post my X-ray photo on the correct part of the forum.

I will provide you with a schematic diagram for modifying the voltage regulating switch power supply. Perhaps it will be helpful for you. A few months ago, I referred to this schematic diagram and made myself a 1KW adjustable switch power supply with an output of 0-150V. I hope it can help you now

Thank you for the schematic, it may be very useful, maybe in the future I will try to replicate.