24V air/vacuum pump questions pertaining to inrush current

Munce31

Hello

I have a couple of questions with regards to aspects of directly wiring a 24 volt vacuum pump to the duet wifi.

To assist my queries I have the following:

Duet wifi version 1.03 -- I understand that this already has a flyback diode installed.
The following vacuum pump vacuum pump in 24v:
1N4007 1000v 1A diode for flyback

Firstly, I have measured the resistance of the motor of the vacuum pump, turning the shaft at each step and the lowest resistance measured was 13.8 ohm (I removed the vacuum pump assembly so that I could properly turn the shaft of the motor with my finger). With 24 volt, and to be conservative with the resistance let us round down to 13 ohm, then theoretical maximum inrush current would be 1.9 amps.

Is it necessary to run in series with the motor an NTC thermistor? From my understanding 1.9 amps is well within the maximum amperate capability of the fan mosfet, which from what I've read is 3 amps.

Would I be right in saying that the 33ohm thermistor is only required with 12v motors where the resistance can be less than 8 ohm, therefore exceeding the 3 amp maximum for the fan mosfet?

Secondly, I currently have at my disposal a 1N4007 diode which is rated to 1 amp, for use as a flyback diode at the motor.

Would it be more suitable if I used a diode such as a 1N5408 which is rated to 3 amp? I understand that the Duet 1.03 has the flyback diode already installed, but from what I've read, it wouldn't hurt to install another to protect against PWM.

Perhaps I would be better if I ran an external mosfet, however I'm unsure if they could take PWM frequencies up till 25khz?

Hope I'm clear with what I'm trying to convey.

Thanks in advance

EasyTarget

@Munce31 said in 24V air/vacuum pump questions pertaining to inrush current:

Hope I'm clear with what I'm trying to convey.

Much clearer than most

I think the 1N4007 flyback would be fine; it has a 30A forward surge capacity according to this datasheet, 1A is it's continuous rating. Similarly the Mosfet probably has a surge rating too.

For the inrush surge, a correctly rated NTC might help, even if the current is within the overall capacity of the mosfet/system you might get a vin drop which could cause issues. Also I'm not as expert as others here but I think there is an inductive resistance/load to consider too which may either increase or reduce the surge. I just don't know DC motors well enough to say.
But I do know from experience that the inrush surge on my CNC's spindle could brownout it's mega256 based controller on occasion, until I beefed up the supply.

Munce31

@EasyTarget said in 24V air/vacuum pump questions pertaining to inrush current:

I think the 1N4007 flyback would be fine; it has a 30A forward surge capacity according to this datasheet, 1A is it's continuous rating. Similarly the Mosfet probably has a surge rating too.

For the inrush surge, a correctly rated NTC might help, even if the current is within the overall capacity of the mosfet/system you might get a vin drop which could cause issues. Also I'm not as expert as others here but I think there is an inductive resistance/load to consider too which may either increase or reduce the surge. I just don't know DC motors well enough to say.
But I do know from experience that the inrush surge on my CNC's spindle could brownout it's mega256 based controller on occasion, until I beefed up the supply.

Ahh thanks for clearing that. My concern with the diode was that its was smaller than the inrush current and may overload it. Good to know that it's appropriately sized when it comes to burst current.

I wonder, could i put two diodes in parallel to share the current load and flyback?

Dougal1957

the diode is there to protect against back EMF and not surge. AFAIR the fame mosfets are actually recommended not to exceed 1.5 amps.

Munce31

@Dougal1957 Sorry i probably didn't word my previous question clearly. I mean to put two diodes in parallel to prevent back EMF. May be overkill?

So if the Fan mosfets are actually recommended to not exceed 1.5 amps, does that mean the minimum resistance for the thermistor should be:

24v / 1.5 = 16 Ohm?

Munce31

Perhaps this thermistor would be suitable?

https://au.mouser.com/ProductDetail/Semitec/15D2-08LD?qs=sGAEpiMZZMv1TUPJeFpwbn2nzGmZq6KjDmRwitVxse4%3D

Munce31

Actually I have another thought for the thermistors.

Could I use 5 of 100 ohm thermistors in parallel?

This should theoretically bring the resistance down to 20 ohm. 20ohm should reduce the maximum current down to 1.2 amp?

Does the math check out?

Why those thermistors, you may ask. Well because they're readily available at the local electronics show in Australia, whereas the shipping price from Mouser or Element14 far outweighs the cost of an appropriately sized thermistor.

EasyTarget

@Munce31 NTC thermistors work by getting very hot (eg Heating up to the point where their resistance falls off dramatically, High resistance initially.)
If you put them in parallel you need to ensure the current through each one is high enough they can still heat up.

To find that current you need to measure the steady state current of your pump, then size the NTC's to that. It's a long time since I did this but the graphs in the datasheets show you how their resistance will change. Their rating is not the 'inrush' rating, it's the nominal maximum steady-state current they can handle.

Two diodes sounds like overkill, but it will not break anything. And do not 'test' your NTC by touching it, I have a small burn on one of my fingers to remind me of this.

Munce31

@EasyTarget According to the data sheet of the air pump, the unloaded current is equal to or less than 450mAh, so i'd expect more when loaded.

Unfortunately the details on the thermistor are thin, however it is mentioned in the specifications that the DC current rating is 200mAh. I'm assuming that is maximum.

So if I use three in parallel that would bring total resistance down to 33ohm and total current capacity to 600mAh, which might be close to loaded current draw of the pump.

I guess there's no harm in trying it out I guess.

https://www.jaycar.com.au/100-epoxy-dip-ntc-thermistor/p/RN3432

EasyTarget

@Munce31 said in 24V air/vacuum pump questions pertaining to inrush current:

According to the data sheet of the air pump, the unloaded current is equal to or less than 450mAh, so i'd expect more when loaded.

Something that can be tested I guess.

So if I use three in parallel that would bring total resistance down to 33ohm and total current capacity to 600mAh, which might be close to loaded current draw of the pump.

I think so too, so long as there is enough power in the thermistors to heat them up, but not exceed the 500mW max. Most NTC discussions I have ever seen focuss on AC supply, not DC.

I guess there's no harm in trying it out I guess.

That's the spirit! I had a small point, I guess the pump runs continuously? if it is stop/starting make sure there is enough time between cycles for the NTC's to cool down. They are not suitable for bang-bang type controllers.

Munce31

@EasyTarget said in 24V air/vacuum pump questions pertaining to inrush current:

That's the spirit! I had a small point, I guess the pump runs continuously? if it is stop/starting make sure there is enough time between cycles for the NTC's to cool down. They are not suitable for bang-bang type controllers.

Yes the pump will be run as part of a berd-air-like system, so mostly will be on and not switched on and off.

Munce31

Weighing up my options I think the most effective option is to remap a heater to a fan port. I'll investigate this further. Thanks for your help!

dc42

@Munce31 said in 24V air/vacuum pump questions pertaining to inrush current:

Weighing up my options I think the most effective option is to remap a heater to a fan port. I'll investigate this further. Thanks for your help!

Yes, when using a pump having a brushed DC motor, it is better to drive it from a spare heater port if you have one. Don't forget to add the flyback diode.

Munce31

@dc42 Thanks for confirming that using a spare heater port is a better option for an air pump.

Yes, I will be installing a flyback diode.

However, would it be beneficial if I use two? One screwed in at the heater port terminals to be closer to the mosfet, and another at the motor to reduce any interference created by the PWM?

Also, will the heater port pwm up to 25khz, as recommended in other threads?