That’s interesting, so you can flip the relays all you like without trouble as long as the 24DC supply isn’t connected? If that’s true then your problem presumably isn’t the typical inductive kick from the relay coil. It looks like your relay board has stuff on it which is presumably drivers and snubbers so let’s assume all of that is adequate to the job.
So, if it’s inductive kick from the valve solenoid it’s being coupled all the way from there, back through the 24DC supply to the outlet, then forward through the USB supply to your shift register, which is impressive! But not implausible.
Anyway, three places I’d add some stuff:
The main thing you need is a snubber network across the valve solenoid coil itself, ideally physically close to the valve (you want to minimize the area of the loop formed by the valve coil - wiring - snubber). Something as simple as a freewheel/clamping diode would probably help a lot. This will also improve the lifespan of your relay contacts which are probably arcing a little.
Small decoupling cap on your breadboard, say 0.1µF on the power supply rails, to keep your logic happy.
Larger decoupling cap on your 24VDC rails (the bus on the left), just to eat any transients the snubber doesn’t deal with. Maybe 1-10µF or so?
Thanks for your thoughtful response… and sorry it took so long to get back to you. I tried different combinations of capacitors, a diode on the load lines, etc… nothing worked. And then I put a 0.1uF capacitor directly between the power leads on the valve itself… and everything started working fine. Admittedly, I’m not 100% sure why… but I won’t complain :)
That makes sense, it forms a simple snubber network. A capacitor in series with a low-value resistor might work even better. Did you try a freewheeling diode directly across the valve leads?
Yes you can absolutely breadboard it. Forcing a current is as easy as following ohm’s law. Make sure there is a certain voltage across a resistor and ensure that only a negligible amount of that current is leaked elsewhere. A difference amplifier is a good way to ensure this, as long as you pay attention to the amplifier input currents.
If current regulation isn’t super important, a highish voltage (say 24+ V) and a large resistor will also work because the variation of threshold voltage will be so small that the voltage across the resistor will be relatively stable.
I think there is some confusion about the word diode here. The transistor is effectively an inverting amplifier, that is that the drain voltage is reduced if the gate voltage is increased. By tying them together, they reach a stable configuration where the gate is just high enough to make the drain low enough for them to be equal. In this configuration, there are two terminals, hence the di in diode. Like a traditional diode, it has a very nonlinear voltage-current relationship. If you apply 10V to it, theoretically the current would be thousands of amperes. Practically that won’t happen but you will blow up the transistor.
I don’t know enough about radiation and semiconductor physics to answer your other questions but if I were you I would just build it and test it. MOSFETs and resistors are cheap and if you do have a radiating source on hand it might be easier to try and fail than to hope someone here can tell you what your part will do when exposed to conditions outside of the manufacturer recommendations.
The two through hole pads by C19/20 arent soldered and the one above it looks to be a bad joint
jumpers on the right arent connected in the same spot
extra traces by the IC
components can die, especially ICs and capacitors from ESD and drops respectively
PCBs themselves can sometimes just be faulty from the factory and have damaged traces.
Another thing is that small ICs like that tend to be fairly difficult to stick down all the pads. Reworking that might be a good option if checking the above doesnt work
Thank you, I will reflow C4 and add C16, but C10 and C14 look fine from the sides.
The messy/missing joints near C17, C19, and C20 are optional components for the other side. (Unused surface mount USB port with through-hole mounting pins and an unused jumper.) Right-hand jumpers are all optional and unused from an earlier prototyping stage.
What do you mean by extra traces by the IC? And you’re absolutely right about the small IC. It’s the MAX17640, and at 2mm x 2mm it’s a piece of work to land correctly.
I was referring to it more as a difference which might be the source of the issue. For example, one via is fairly close to the pad for C16 which could be shorted (probably isn’t, but still a good idea to check). It could be a wiring issue but could also be a board manufacture issue.
You can try checking whether the 54v supply has any voltage spikes if you haven’t already. 60-to-5 converter is most likely a switching converter and they draw high current instantaneously and can cause spikes if not filtered properly which may affect the functionality. This may not show up in DC voltage measurement, and daisy chained boards may still function if they are filtered well. One of the daisy-chained boards may be causing the disturbance too, and maybe top board is filtered well but bottom isn’t for some reason, maybe a soldering error(e.g. filter capscitor ground connection, poor soldering may be conductive but with high inductance, hindering filter functionality). So it is good to check supply quality. Also what do you mean by not powering up? Is 3.3v not working? What about 5v?
Not entirely sure, but maybe these help you somehow:
The relay has a coil which requires 0.35W. The chip seems to have a maximum output current of 35mA.
The ‚switch on current‘ of an inductive load is usually 3 to 5 times higher than the ‚hold‘ current.
The valve may not have a free-wheeling-diode. This could create an issue by creating strange voltage spikes on all your supply voltages (connected by GND).
Yeah, decoupling cap here might as well just help OP fry the chip more effectively by ensuring it can sink all that current.
Solution here is probably a transistor/MOSFET that the chip turns on, which in turn turns on the relay. Relay coils are inductors, so that probably also needs a diode to protect the transistor from inrush current and also the kickback when it turns off: inductors resist changes, so it’ll try to keep sinking the current and result in temporary spike of very high voltage: spinningnumbers.org/a/inductor-kickback.html
Using MOSFETs as TID sensors is common enough. A term that you can use for more research is RADFET. The best way to measure threshold voltage is to sweep the gate voltage. In my experience however, if you intend to measure this in a non-lab environment (say, in a satellite), I would recommend that you instead connect the gate to the drain, force a small constant current (maybe 10uA) from the drain to source, and measure Vds (which is equal to Vgs in this configuration). This won't give you the threshold voltage per se, but this will produce a voltage that changes as dose accumulates, is a far easier metric to measure, and is as equally valid as measuring the threshold voltage to determine TID. You can't really predict the shift in threshold voltage according to TID unless your MOSFETs are all from the same batch (manufacturing defects and tolerances), and you need to calibrate them in order to obtain a calibration curve (this is done by simply going to irradiate several (the more the better, I suggest at least 10 for statistical significance). Alternatively, you can buy pre-calibrated ones from companies who make MOSFETs for this intended purpose like Varadis. If you really want to measure the threshold voltage, you could read MIL-STD-750, which outlines how to measure the threshold voltage.
Hey I messaged them a bit, this is an undergraduate project with no budget (so a MOSFET tester is out of budget). I also suggested a sort of sweep method using an MCU and some op-amp glue, but I don't think they have sufficient background to get this kind of thing working yet (in fact I barely do, so probably it won't 'just work' with whatever I came up with off the top of my head).
What I was thinking is perhaps they can set Vds and Vgs to fixed values such that a particular MOSFET conducts a fixed current, e.g. 100mA, somewhere near-ish the start of the linear region. Then record the Vgs required to achieve this current for each of a set of MOSFETs, say a few dozen (because of part variation).
Then after exposing them to varying amounts of radiation (a few for each exposure level), put them back in the same test conditions and measure how the output current has changed, what Vgs will restore the same current, draw some graphs, discuss the advantages and disadvantages relative to the Vth method with regards to radiation dosimetry, conclude, and call it a day.
Think it would work? No need for an MCU or signals processing this way, so the science can get done with the tools they have.
Also I never had free access to strong radiations sources in undergrad, so am a little jealous. I barely got to use tritium, and that sparingly.
Hello, do i still need to apply a supply voltage of Vgs, or will only the current at the drain be the main source. Also, is this MOSFET sensitive enough for a cs-137 (661 keV).
Ultimately, yes you will need to bias the gate. If you put the MOSFET in the diode connection mode, the gate will automatically be biased when you force a constant Ids current. While I have never worked with this MOSFET nor with Cs137, I don’t see why it wouldn’t be sensitive. A few notes:
This MOSFET is in a TO-247 package, so make sure that you have the front of the MOSFET pointed towards the Cs137 source during irradiation, otherwise the leadframe will likely act as an attenuator, reducing the sensitivity.
This MOSFET is a HEXFET, which normally aren’t designed for continuous high DC power dissipation (they are meant for switching). So keeping the dissipated power in the FET would be best.
I’m not sure if there is a difference in general sensitivity between HEXFETs and other MOSFET types like VDMOS or traditional monolithic planar MOSFETs.
I’m not sure if you already planned this, put generally it is recommended for the MOSFET to have all pins grounded during irradiation. Biasing of the MOSFET can affect its sensitivity (depends on every MOSFET), so having all pins grounded keeps them in a constant state during irradiation (and lets all accumulated charge get shunted to ground, preventing ESD damage).
I’m not entirely sure, but I suspect that the die is mounted to the leadframe on the flat side of the package. In this case you should point the flat side towards the source.
I’m using lm334z as a current source, i’m still thinking and deciding on where to connect it to the MOSFET. Can you propose the wiring for the current source input once the MOSFET is in diode connection. Like for example, in a breadboard i used a wire to connect the drain and gate. then, i’ve applied a constant current at drain terminal with source connected to the negative. So to check the change in voltage, should i measure the voltage in Gate to source?
Hello, I have already tried to apply a microampere current (around 40 μA) using only Ohm’s law. I used a 2V input and a resistor in series. However, the LM334z is not working, and I am unsure of the reason for its failure. Will using an input voltage and a resistor in series work just fine?
I have also measured the Vgs, which reads at around 0.117 mV. What do you think about the measured Vgs? Technically, it should be around these values, right? Considering that there is a very small current.
You will probably need to increase your voltage. I haven’t ever used the LM334, but it will need a minimum voltage across it. I don’t know if you are still using the IRFP250N, but if so it has a threshold somewhere between 2V and 4V for 250uA, so the threshold won’t be as high but it should be close. I would try using 5V if it’s fine with your setup.
For the suitability of the resistor method, you should do the math on how a change in Vds will affect the current, and then calculate how much this variability in current will affect your readings. If this error/innacuracy is acceptable, then why not.
Do you mean that to properly operate the LM334, we need to apply or reached the threshold voltage of the MOSFET which is around (2 and 4V)? But, i’ve tried to used it (LM334z) as a standalone and still it doesn’t work. maybe the LM334z is the problem hahaha *Regarding the resistor method, What do you think about the measured Vds, Vds=117mV (With an input current of 40uA)? i’m confused with the “how a change in Vds will affect the current” *Isn’t it Vds the parameter required for determining the dose? So if there is a constant current and resistance, how does the radiation affects the MOSFET? Will resistance or current increase upon irradiation leading to an increase in the measured Vds?
You will need to provide a voltage of at least the threshold voltage PLUS the minimum voltage of the LM334Z. If the LM334Z circuit by itself doesn’t work, that will be the first problem to figure out. Make sure you completely read through the datasheet www.ti.com/lit/ds/symlink/lm334.pdf?ts=1689139734…, there are example circuits in it for reference. For the resistor method, keep in mind that the current isn’t constant; the voltage is. Your current is dictated by your resistor and the voltage across it, which is the supply voltage minus the threshold voltage. If your threshold voltage changes as the dose increases (which is the typical behaviour), the voltage across the resistor will change, therefore your current will change, which will generate error in your reading. The only way to minimize this error is to apply a very high voltage (100’s to 1000’s of volts) to the resistor, such that a change in the threshold would become a rounding error.
I work with a lot of particle detectors. Instead of this, I recommend a photo diode and laser diode. Light dispersal particle counting is relatively easy to pull off by comparison to scraping off coating and all that. You could even use a very small segment of solar panel and a light source or an IR detector and IR LED similar to a smoke detector.
Doing some quick math, the transistor will have a base current of 5 milliamps, which a Pi should be able to supply. At a fairly typical beta of 100, the transistor could drive the fan at up to .5 amps, which is plenty for a small fan. A MOSFET transistor is generally better suited for switching high current loads, but for this a BJT (as drawn) should be fine.
I recommend EEVBlog’s OpAmp tutorial. His explanation is pretty simple to understand. Basically there are two rules (note these rules are ideal, but the exceptions can usually be ignored):
No current flows into the inverting and non-inverting inputs.
For negative and positive feedback circuits, the OpAmp wants to keep the inputs the same by changing its output, and will sink power to its positive or negative power rails to achieve this.
The point of the diode is to prevent reverse current that gets induced when a (brushed) motor is turned off. It essentially turns into a small generator while spinning down, and the diode essentially short circuits that. It prevents damage to the rest of the circuit. If that motor is brushless (with an integrated control board), you likely won’t need it but it doesn’t do any harm either.
askelectronics
Active
This magazine is from a federated server and may be incomplete. Browse more on the original instance.