I can’t seem to get this to go above 30C? Whether I have it hooked up directly to 5V on my UNO or off a digital pin on my Nano V3.0 Board it just doesn’t seem to heat up as it does in the chart (~60C in 2 min) or the video.
Anyone else had better luck with it? I am trying to make a dough proofing box, so doesn’t need to be that hot, but if limited to 30C it will take forever and may not keep up with any thermal losses.
Hi Dion,
Depending on your source it might not be able to pull enough current to reach 60 degrees. Can you share a picture of your setup?
Hi Jack,
It’s a Nano V3.0 with a 100K NTC thermistor on A0 and an LED with 220ohm resistor to signal when element is on.
I also have a 10k pot to set target temp between 20-30C.
Hi Dion
If you have got it sitting out on the table like shown you are never going to get the thermostat to 30º. It is only 3W which is not much. If you enclose it in something or wrap it up you will have a better chance.
I think this bit of the text says it all
" ideal for things like hand-warmers and other heated garments."
Cheers Bob
Edit
Just had a bit of a study of your pix. Very hard to tell what is going on. Even the parallax error with the camera does not make it easy to tell which Arduino pins are connected to what. A schematic would be helpful.
It looks like you are trying to drive the heater directly from an IO pin. If so can’t do that. The heater needs 600mA and the IO pin will only supply 20mA so well short.
post a schematic showing EXACTLY what you have and we can go from there.
Cheers Bob
3 Likes
Thanks Bob, I think you’ve nailed it. It’s the low current available.
I’ll have to add an additional power source and a switch of some sort.
1 Like
Hi Dion
If you want to switch it with the Arduino a Mosfet configured as a low side switch would be the easiest way. You will need a power supply for the heater, I would suggest one that is capable of at least 1A.
Cheers Bob
2 Likes
Operating Current: ~600mA. Maybe it’s not getting sufficient current.
Hi Ila
If it is being driven from a GPIO you can bet on it
Cheers Bob
So I set up a parallel circuit with a 5V power supply going into the element and VIN on the Nano. I use a IRF520 MOSFET to control power supply to the element. When I connect the power supply direct to the element and bypass the Nano, it is drawing ~450mA.
When I use the MOSFET it barely draws 20mA. It’s as if it is being powered by the IO pins?.
In the picture, the original terminal is connected to a USB cord that goes either into the wall via an Apple phone charger or my laptop. I also have a 100k NTC probe and 10k pot to control the set temp but for simplicity have left them out.
I have tested the MOSFET using a led as a proxy and it works. For some reason it won’t draw the full amperage.
2 Likes
Hi Dion
From what I can make out in the top pix it should work. It sounds as if the Mosfet is not turning fully on.
Try disconnecting the gate from the Nano and touching the gate to 5V momentarily. It should switch on and stay on until the gate cap is discharged to ground.
The gate is a short to ground at switch on courtesy of the gate capacitor (360pF) and as the output of the Nano is very limited in current this cap might be killing it. You need about 1kΩ in series in the gate lead. If that does not work and the connection to 5V as above does work you might (read will) need a driver circuit to provide enough gate current or a dedicated gate driver IC.
Cheers Bob
1 Like
Thanks Bob.
Should I just use a small 5V relay instead?
1 Like
Hi Dion
It would have to be an Opto coupled set up as the Nano could not drive the relay direct.
Try what I said above. Firstly by momentarily connecting the Gate to 5V to make sure turning it fully on achieves the desired result and there is not something else wrong.
Then with the 1kΩ series resistor to the Gate. This will limit the Gate inrush from the Nano to a safe value and stop the cap from killing the 5V nano output.
Cheers Bob
2 Likes
So momentarily connecting Gate to 5V does turn the MOSFET on but current fluctuations wildly up to 400mA. It is usually around half of that or less.
Once I reconnect the gate to the Nano pin it drops back to 30mA.
I added the 1000ohm resistor in series between my the Nano and the gate and it also just sits on 30mA.
Hi Dion
OK so there is something fundamentally wrong here. I would strongly suspect the breadboard connections. The cheaper breadboards I don’t think are all they are supposed to be. I don’t actually know what the rated maximum current is for these things and I don’t recollect seeing anything published. Does someone here know? I have only used these in the mA current levels.
You might have to break out the old soldering iron and get rid of the breadboard at least to isolate the problem.
OK so the Nano is not driving the Mosfet fully on. Get rid of the other problem first and find out what is wrong before you tackle this one. There are a few questions and things to measure once the first problem is eliminated. It is no good going further until then. If you try you will be going in circles for weeks with no result.
Cheers Bob
Add on
You might have to find another Mosfet. All the figures in the data sheet are at a gate voltage of 10V. The gate threshold is quoted at up to 4V so even at the full 5V this could be marginal. That is for a quality product. If your device happens to be a cheap knock off this could be anything. It could be a case of someone found some that can be driven with the Nano 5V and straight away assume that they are all the same. That happens a lot lately (5V relays @ 3.3V???).
You will have a better chance with one that is listed specifically a “Logic Level” device. They all like to be turned on fully.
1 Like
I am wondering if this is because without connecting back to the Nano the lead is “hanging in the air” behaving like an aerial transmitting random voltage signals to the gate?
The mosfet was part of the Arduino Uno starter kit. Don’t know the quality of it but it worked for the “Motorised pinwheel” project (Project 9). In that project it supplies power to a motor with an auxiliary 9V battery. In that project the motor was drawing ~200mA.
Perhaps I should replace the Nano with the UNO to see if the microcontroller is the problem? It has its own breadboard as well.
Hi Dion
Not very likely. In fact when you touch the gate momentarily to 5V it will turn on and stay on. This is because the gate capacitor is charged up and the gate resistance of the Mosfet is so very high (could be up in the TΩ region) it will take some time for the charge to leak away and could stay on for weeks or months. In fact if you remove power for a while and then restore power IT WILL STILL BE ON. And will remain that way until the gate cap is discharged back to source to turn it off.
As I said, get this bit working first. DO NOT persevere with 2 parallel faults or you (or us helpers) will get nowhere fast.
What the motor draws has nothing to do with the Mosfet switching.
That might not be a bad idea at all. The UNO GPIO pins have more grunt than the Nano. And I still think the breadboard could have something to do with BOTH problems. Especially if you have been using that breadboard with header strips.
A bit of trivia. I did look at a couple of breadboard product descriptions on the Core site. They seem to have 2 versions of breadboards, one version very cheap and the other somewhat more expensive. The more expensive one DOES state a contact current rating as being 2A. This might be a bit optimistic but you should have been within that. The cheaper ones do not carry this specification, does not rate a mention. I don’t know which you have.
Cheers Bob
1 Like
Ok I’ve had some good progress. I switched to my UNO, which initially wasn’t successful until I decoupled it from the common power source it was sharing with the element and plugged it into its own power source via its USB. It worked pretty much as it should be doing, pulling about 450mA. This means I have to have two power sources which isn’t practical, so I have to work out why the current through the mosfet drops to 35mA when I attach 5V to VIN.
When I repeated this approach with the Nano it would only pull 200mA? I measured voltage across the digital pin going to the gate on the UNO and it was 5V. On the Nano it was 4.7. Is wouldn’t think that difference would cause current to drop by more than half?
More answers but more questions…
Hi Dion
Possible. 4V is very marginal. If you look at the graphs in the Data sheet for that Mosfet you will see why. That device likes about 10V on the gate to urn fully ON
Have you done as I suggested and got rid of the breadboard to see if there is a problem there. If not why not. I said before you have to get rid of one problem at a time or you (and us) will be going in circles which is something I am not prepared to do.
There is one potential problem on that breadboard I can see. That orange terminal block is designed to mount in a PC board, NOT a breadboard. The leads will be quite short for breadboard use and maybe the contact is not reliable enough. I have some of those so if I get time I will have a look. I am busy arranging a funeral at the moment so will see.
I can’t think of any reason you would need 2 power supplies. I am not sure of using USB and 5V on VIN at the same time though. Maybe you can’t do that.
There is something else basically wrong and the pin length on that terminal block could be the culprit as it is common to most everything.
Cheers Bob
Add on: and you might do better with a dedicated “Logic Level” Mosfet.
Yes, by changing to the UNO I used its dedicated breadboard. With new configuration of replaced breadboard and seperate power supplies to element and UNO the element is working as it should, therefore I am satisfied that the mosfet is working as it should receiving 5V from the UNOs digital pin. The mosfet came with the Arduino kit so not sure why it needs 10V to open?
Consolidating to a single power supply is my next issue.
Agree it is not ideal but for prototype purposes it works. I will solder it into a PCB when I have this all sorted for longer term solution.
For simplicity I only want one power supply. When I connected the VIN to active off orange terminal, I disconnected the USB as I am aware it can fry it. It seems logical to power the element and microcontroller in parallel off the orange 5V terminal but for some reason it is creating an issue.
Hi Dion
Follow on from the last reply
You would be better off using a IRLB8721 Power Mosfet. Core SKU ADA355.
Note that “I” could be a “1”.
The graphs say this should be fully ON at a Gate voltage <5V so would be a lot better.
Cheers Bob
2 Likes
