Electronics > Projects, Designs, and Technical Stuff
Solenoid driver design
<< < (6/6)
langwadt:

--- Quote from: mbless on December 27, 2019, 07:15:23 pm ---
--- Quote from: MagicSmoker on December 27, 2019, 12:25:10 pm ---Sure, I'm sure: your explanation on circuits 1 and 2 is correct, but that doesn't make what I wrote incorrect. As you noted, D1 does nothing in the first circuit, but also note that D2 is redundant, so it doesn't do anything, either, as long as the TVS breakdown voltage is higher than the supply voltage (which it should be), and it is, indeed, a bidirectional type as shown.

--- End quote ---

Yes, the TVS is bi-directional and rated for 27V, and the supply voltage is 24V. I tested your suggestion see here.
(Attachment Link)
Removing D1 and D2 decreased the current turn-off by ~5%. Incidentally, I did a test without any of the diodes installed, and the turn-off time was reduced from 1.75ms to 0.3ms! I'm surprised the solenoid current can decay so fast without any diodes. Any thoughts on why that would happen, and is then better to leave it without diodes or keep them for protection? (I do have a 470uF electrolytic on the supply rail.)

--- End quote ---

without the diodes the diodes the "turnoff voltage" will be clamped to the Vbr of the mosfet, look at it's avalanche
(and power) rating to see if it is happy about that
T3sl4co1l:
Even if it's rated for avalanche (they usually are these days), I don't recommend using it on a repetitive basis.  Solenoids are probably alright anyway, but you can just as well use a TVS with a voltage rating somewhat lower (say by 20-30%) than Vds, and get nearly the same slew rate.

Again, the nice thing about two-switch with rectifier diodes, is efficiency; you won't get the highest possible slew rate, of course.

At some point, it may be worth checking how much slew you actually need.  Take measurements of the mechanicals, figure out how much it actually moves in the hundreds of microseconds, or milliseconds, the solenoid discharges in.  Tiny stuff like injectors and print heads (do they still use those anymore? ;D ), probable.  Big stuff like solenoid actuators/valves, probably not so much.  If the actuator itself is slow, there isn't any point in a maximum-speed electrical solution, and then you can consider other goals like efficiency.  If electrical speed is still required, a less efficient or more complicated solution is then an option.

Tim
langwadt:

--- Quote from: T3sl4co1l on December 28, 2019, 12:00:22 am ---Even if it's rated for avalanche (they usually are these days), I don't recommend using it on a repetitive basis.  Solenoids are probably alright anyway, but you can just as well use a TVS with a voltage rating somewhat lower (say by 20-30%) than Vds, and get nearly the same slew rate.

Again, the nice thing about two-switch with rectifier diodes, is efficiency; you won't get the highest possible slew rate, of course.

At some point, it may be worth checking how much slew you actually need.  Take measurements of the mechanicals, figure out how much it actually moves in the hundreds of microseconds, or milliseconds, the solenoid discharges in.  Tiny stuff like injectors and print heads (do they still use those anymore? ;D ), probable.  Big stuff like solenoid actuators/valves, probably not so much.  If the actuator itself is slow, there isn't any point in a maximum-speed electrical solution, and then you can consider other goals like efficiency.  If electrical speed is still required, a less efficient or more complicated solution is then an option.

Tim

--- End quote ---

why not use avalanche? (or some of the protected fet with active clamping)

afaikt the two switch is the best of both worlds, slow decay without dumping energy into FET/TVS for the pwm current control and fast decay for turn off, dumping the energy in FET/TVS
mbless:

--- Quote from: langwadt on December 28, 2019, 12:49:57 am ---afaikt the two switch is the best of both worlds, slow decay without dumping energy into FET/TVS for the pwm current control and fast decay for turn off, dumping the energy in FET/TVS

--- End quote ---

I will attempt to modify the 3rd version to make it like a true two-switch driver, though, I will probably have to spin a new board. Is the TVS usually omitted in such configuration, so it's a fast recovery from low-side drain to supply and another fast recovery from ground to high-side drain?


--- Quote from: T3sl4co1l on December 28, 2019, 12:00:22 am ---At some point, it may be worth checking how much slew you actually need.  Take measurements of the mechanicals, figure out how much it actually moves in the hundreds of microseconds, or milliseconds, the solenoid discharges in.  Tiny stuff like injectors and print heads (do they still use those anymore? ;D ), probable.  Big stuff like solenoid actuators/valves, probably not so much.  If the actuator itself is slow, there isn't any point in a maximum-speed electrical solution, and then you can consider other goals like efficiency.  If electrical speed is still required, a less efficient or more complicated solution is then an option.

--- End quote ---

This is for an injector, so I was going to check the end of injection with my high-speed camera at work (won't be back to work for a few weeks). I have already matched the current profile of the OEM equipment, so it will be interesting to see if the faster decay actually results in a shorter end-of-injection delay. In reality I don't need to be faster than OEM (and shouldn't for repeatability purposes), so if the two-switch is the best way to go, then I can proceed with that.
T3sl4co1l:
Correct, you don't need the TVS if there are diodes doing the clamping.  (Note the supply itself should have a bypass cap somewhere nearby.  How near, depends on how fast the transistors are switching, and how much current.  You can probably afford quite some distance, and quite some series gate resistance to slow it down and also save on EMI output from the switching edges being slower.)  Indeed it's better that way, as you aren't risking pulling the full supply across the TVS if/when voltage swells -- not common events, but typical features of automotive tests: battery jumping, reversal, load dump.

For example, on a 24V system, 100-200V diodes and MOSFETs will pass probably any test used in the industry.

Aftermarket stuff may not need to handle this (or be expected to) -- but YMMV, caveat emptor and all that...

High speed camera, nice. 8)

Tim
Navigation
Message Index
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod