So...
I made a some research, sorry if I ask some obvious questions but I am completely new to this...
Would the IRF520 work for this?
I've even found it already pre-soldered to some board which are sold specifically for Arduino so my polite guess it posivite about it...just wanted a confirmation...
Thanks a lot guys!
Ale
No. Don't you currently have an IRF520? If you look at the datasheet, you see this
RDS(on): VGS = 10 V 0.27
This means they spec the Rds value when the gate voltage is 10V - not a logic level transistor, You need one that specs this at VGS = 5V.
If you look for IRLxxx series, the L stands for logic (vs. the IRFxxx)
Another good Logic Level MOSFET is a 30N06L. I just used a few in a motor controller of sorts.
Those items on Amazon seem to work with arduino. The reviews says that it has been tested and working with Arduinos.
If you decide to go with the FETs by themselves, @wolframore showed me the way to drive the FETs. See attached picture.
Good luck.
OMG, sorry, you are right, the IRF520 is exactly what I have, I was convinced, I don't know how, that I had the 540...
Maybe a stupid question...
But why, even in the Arduino starter pack, they provide the IRF520 if they cannot be driven properly by Arduino?
Also, how can they sell those module if Arduino cannot open the gate 100%?
Thank you for the hint about the 30N06L, but apparently it doesn't exist on Italian Amazon...any other suggestion?
Thank you for the scheme! The fans MUST be in series or they could also be in parallel?
Thank you very much all!
Ale
Otherwise, as an alternative, may I use a normal BJT transistor?
In this case which one should I go for?
Thanks a lot guys, I am really not an expert...
Yes, you can use a BJT. Just make sure it can handle your voltage (24V) and the current demand (plus margin) Since you have 180mA x 3 = 480mA, a 1A BJT will work just fine. Look for "Darlington Pair" to make sure you can handle the current (common types are TIPxxx)
Thanks a lot for your answer!
In the meantime I found this article: assuming it is right, it says that at 5V the IRF520 should be able to output 3A.
As I need a total of 480mA, just as you said, why it doesn't work?
Thanks a lot for all the explanations!
Ale
A TIP102 transistor will work fine.
In series, each fan will not get a full 24v. The voltage to each fan will be divided evenly among the fans plus any circuit resistances.
I would advocate for parallel connection. One of the reasons is that if a fan fails or begins to fail, the other two fans will be affected too. Vice in parallel, the fans will operate independently.
Another reason is that they should run faster, which may not be necessary in your case. See attached pic and note the higher current to each fan in parallel config, plus a full 24v to each fan.
Are these fans rated for 24v?
Hello Darrellread,
thank you for your explanation and scheme!
I actually put the fans in parallel (without knowing exactly the reason why but I thought it was better: I knew about the fact that if one failed the other ones would too, in series, but I didn't know about the dropping volts...).
Thanks for the transistor name!
Unluckily even this one is not available on amazon.it... very strange I cannot find anything at logical level!
The mistery remains for me about why Arduino provides the IRF520 in the starter pack if they cannot be controlled properly...
So, to be clear the IRF520 will still work with an Arduino.
If you look at the IRF250 datasheet it shows that the IRF520 will conduct about 1amp at 4v Vgs and about 2amps at about 5v Vgs. See attached picture.
Compared to the 30N06L, the datasheet shows that the FET will conduct well over 10amps at 5v.
I'm pretty sure that your fans will not draw more than an amp, so you should be fine with the IRF520.
If you wanted to go with a transistor like the ones in the picture provided, they will not conduct enough current for your purposes.
So why don't we take a look at how you hooked up your IRF520 FETs and see if we can optimize your circuit.
Please be careful how you read the datasheet:
according the datasheet Vgs threshold it could be anywhere from 2V to 4V (guaranteed threshold) which means somewhere between this value you should be able to conduct 250 uA
Due to this manufacturing tolerance... it may work or it may not work at 4V ( I don't count 250 uA as working)
Note how they are performing the test: 20 uS pulse width - sounds like a single pulse. It's 20/1,000,000 th of a second.
Even if you were to pass 2A with 5V at gate. you have no idea what the Rds is so while it may pass 2A but for how long and how hot will the MOSFET get? They have done all the Rds rating at 10V Vgs because I'm sure it's not great at 5V and it was really designed for higher voltage switching. Notice also that the gate voltage goes up exponentially.
I estimate a reasonable Rds for 5V is about 2-3 Ω (it's 0.25 Ω or 250 mΩ at 10V)
I believe if you switched 24V 2A we are talking about dropping about 4-6 volts at the mosfet... that's going to get hot dissipating 8-12 watts. Wasting power and getting hot!
It will also reduce the voltage going to your fan which it probably needs.
There are a number of other issues including switching speed and other inefficiencies.
the specs for the 30N06L is better for logic levels.
Why keep insisting that it will work... it's like using the wrong tool for the job. There are better mosfets out there for our logic level switching which will work better, waste less power and I highly recommend saving the IRF520 and 540 for when you need something that has higher gate voltage.
I am in no place to argue with wolframore.
However, there seem to be a lot of arduino projects using the IRF520 FETs. They must be able to perform somewhat reasonably with the arduino.
While I appreciate that there is a right tool for every job, Ale_V has had a very hard time sourcing an alternative. Might need to make it work.
I say this with the utmost respect, wolframore.
darrell hope the information helps, was hoping to just clarify the issue. Yes it will switch, and if the current is low enough you might be ok. But be careful with the datasheets... there's some marketing involved with those specs.
these are in the TO220 packages everyone seems to be familiar with - easy to breadboard
IRLB8721
IRL520
IRL520
IRL540
IRL640
IRL3803
the list goes on and on...
Quick search in Digi-Key gives you over 54,000 choices for MOSFETs... no one knows them all.
Go for the IRF520 and get started.
The other option is to use a general purpose transistor to switch the gate of the MOSFETS he has.
it would invert the logic but it's a better idea.
36 posts and still questions. Can another 10 new helpers step in and add 10 more new alternatives to bring this proj to its goal?
do we have a time limit?
He was using the IRF520 - wasn't working... and now we keep pushing him back to IRF520. You could drive it with a transistor but it's not easy.
If you use one transistor it would stay on at power off because it needs a pull up. We have the wrong MOSFET yet again... it has to be done with a P-CHANNEL.
Another option is to use two transistors. but it really starts to get messy. The OP needs to get the right device to do this correctly.
To me the IRL520 looks like an N-channel logic MOSFET and it ought to work. Either the item is damaged, not connected the proper way or something else is wrong. Lets spot what is wrong.
Helpers could use their time in a more productive way than just moving around the spoon in the kettel. Adding new guesses is waisting brain energy as I feel. It's like keeping a communication running in idle without adding anything solving the problem.
Yes, it's free to leave the topic but what about OP wanting the problem to get solved?
