We did exactly as the schematic/picture showed, but when we power it up all lights come on and buzzer sounds with no variation, they are just steady on. I have been talking to a friend via email who is better at electronics and he suggested measuring voltage at coil connections when coil was connected, and it is a steady 2V. No variation. Which he said seems like the switching via the MOSFET was not occurring.
Our only deviation from original poster's design is that we did not have small copper wire long enough to do 25 turns at 20 cm so we did 25 turns with 26 gauge aluminum. AND we did not have specific value of resistors for the 2M and 300R so we had to use a couple lower values in a series to get the right resistance which I understand is common.
My question...and I almost hesitate to ask it. Will 26 gauge alum wire work? Should there be more or less turns. I asked the OP about what the specifications should be and what would need to change if we had a different coil...but no word yet.
Any clues are greatly appreciated, or if you have another good quality metal detection project you could turn us onto...
The aluminum is fine, except that sometimes there is a problem of making a good contact between aluminum and other metals (which is why the experiment of using aluminum for house wiring was abandoned in North America).
What you have is a straightforward hardware troubleshooting task. The fastest way for us to help is to post clear detailed images of your actual work.
Okay. This is our 1st attempt. I will say before you look at pics. Try and remember my 8 year old did the wiring, I traced it and to me it seems to check out...even if it is a little messy.
I hope it is okay that I shared through google photos so you all can see the higher quality pics instead of doing an upload through here and losing quality.
Do you have a white wire jammed into the same hole on the breadboard, as a resistor (follow pin A0)? Please upload those photos here. They won't lose any resolution. What op amp did you use, a MAX951? It got obliterated in the photos. Have you done a (repeat) point by point check of all the wiring? Confirmed resistor values with a DMM?
Don't you think it would be a good idea to supply power to the op amp via pin 8?
Also I don't think that is aluminum wire. To me, it looks like tinned copper hook up wire.
yes, apparently there was a white wire in the same hole...but it is meant to be on the same rail. The op amp is a LM741CN. This is still our first build out of this particular project. It took him most of an afternoon to wire it up, and I double checked his work after every few additions. Then I checked again before we applied voltage. Did I check each resistor? No, but they were brand-new out of a "kit" of them...probably should, huh?! Perhaps I should supply the op amp power, but I actually don't know much about them...and was going on the man's schematic. Should I put power to pin 8?
I will work on getting those pics uploaded here. Upload out here in the booger woods takes a hot minute or 3000.
jumper_d_1981:
Perhaps I should supply the op amp power, but I actually don't know much about them...and was going on the man's schematic. Should I put power to pin 8?
I give up on uploading to this site. I have terrible upload speed...and it seems to tell me to talk to the forum admin after it spends 10 minutes deciding whether or not to fail.
Sorry, google works from my phone which is just bizarre.
IRF840 is actually a bad choice in that project, as it is not a logic input gate MOSFET, it needs at least 2 to 4 volts to start conducting and the specs basically suggest a gate voltage of 10V to saturate drain to source.
The FQU13N06LTU MOSFET suggested in the SmartPhone version should work as it is a logic level MOSFET.
I always thought LM741 is synonym for "it doesn't really matter, we just have to put a number on it, but you can use just about any OpAmp you happen to have floating around in your parts bin as any halfway modern one will outperform the 741 anyway".
The MOSFET as suggested in the article will work well with a 12V supply the article writers probably used (as is evidenced by the 62k/47k voltage divider on the output of the OpAmp).
300Ω is rather low a series resistor for those LEDs.
A solderless breadboard is a terrible way of implementing a high frequency LC oscillator circuit - which I think this is (based on the description, it's very hard to recognise from that wiring image how it's supposed to work - a proper schematic would help wonders).
I apologize for the upload issues. Thank you TomGeorge for hooking up an upload on my behalf.
I am using two Lithium 18650's in a little 2 battery pack giving me 8.5 volts for power supply. Perhaps that is not enough?
Tom suggests FQU13N06LTU MOSFET. It or any logic level Mosfet will work, correct?
Should I be worrying about the OP AMP as well?
If I go up to a 12v p/s the 300 ohm res. will need to be a little higher I would assume.
I fully intend on placing the components over to an actual board but I wanted to verify that the thing worked first before spending hours soldering only to be let down.
Apologies if I sound half baked. It is morning and I have not have my "go juice"/coffee yet.
wvmarle:
The MOSFET as suggested in the article will work well with a 12V supply the article writers probably used (as is evidenced by the 62k/47k voltage divider on the output of the OpAmp).
The MOSFET is being controlled by the controller pin not the "opamp" output through the divider.
One circuit is Blue the other is Purple.
Tom....
I have a buddy who has a scope and he thinks he can track down the problem. We are working on getting together on a pretty day so we can be out in the sun, for social distancing reasons...better safe than sorry.
I am hoping he can help because I have looked at 9 or 10 other DIY metal detection projects on here, hackster, and instructables and this is the most sensitive one I can find, that I have all the components for more or less...there was one that was way over my head...well this one currently is too, but I have faith.
BTW. I tried 10 and 12v, neither one made much difference but the Mosfet wants to get pretty warm(uncomfortable to touch)...I understand they can handle up to around boiling temps...but not sure it should be getting that hot on this...
jumper_d_1981:
I have a buddy who has a scope and he thinks he can track down the problem. We are working on getting together on a pretty day so we can be out in the sun, for social distancing reasons...better safe than sorry.
I am hoping he can help because I have looked at 9 or 10 other DIY metal detection projects on here, hackster, and instructables and this is the most sensitive one I can find, that I have all the components for more or less...there was one that was way over my head...well this one currently is too, but I have faith.
BTW. I tried 10 and 12v, neither one made much difference but the Mosfet wants to get pretty warm(uncomfortable to touch)...I understand they can handle up to around boiling temps...but not sure it should be getting that hot on this...
Good, make sure you check that the MOSFET is turning ON when ever the gate signal is high.
Tom...
TomGeorge:
Good, make sure you check that the MOSFET is turning ON when ever the gate signal is high.
Tom...
Well, my friend Alex came down with his scope and poked around for about 2 hours on this. He checked my wiring against the picture on the original poster's page and said everything checked out, he replaced my 741 op amp with another because it did not have any voltage where it was supposed to, and that fixed the coil not having a pulse. So now there is a pulse but the pulse is tiny like itty bitty...he said he would dig more into it and perhaps even set up his own version of the breadboard to look into it more, but it seems like that voltage pulse is not enough to be read from the arduino.
Big FAT pin put into this because the wall continues.
Meanwhile if someone had a metal detector project that was halfway usable I would love to hear about it. We did two other projects for metal detection from the hub and they weren't very sensitive having a hard time sensing anything smaller than a baseball...does not a metal detector make.
jumper_d_1981:
So now there is a pulse but the pulse is tiny like itty bitty...he said he would dig more into it and perhaps even set up his own version of the breadboard to look into it more, but it seems like that voltage pulse is not enough to be read from the arduino.
You need to look at the signal on the MOSFET Gate, it should be 5V, then check the signal on the MOSFET Drain, it should be from gnd to the powersupply voltage that the coil is running off.
I suspect that the 5V gate signal is there but it is not high enough to turn the MOSFET on, as mentioned earlier, the MOSFET is designed for a much higher voltage to turn it on.
Tom...
