Serial device activated ni-chrome trigger problems

Discussion in 'The Projects Forum' started by techx, Mar 31, 2011.

  1. techx

    Thread Starter New Member

    Sep 21, 2009
    14
    0
    Currently having problems with a triggering device I am working on for an engineering project. I'm attempting to use an IRD510 which I have read a couple of things on this forum to state it might not work correctly, I'm looking for detailed feedback. Two formats, first is testing, second is the actual device.
    Diagram 1: [​IMG]

    This shows a 3v(2x1.5v AAA's) power source on the left used as the gate voltage. 6v(lithium-ion) source on right goes into drain and out source of the IRD510 to the 1.9-2.4v/20mA LED pushing a 220ohm resistor(only one I had at the time and is cutting voltage to ~1.6v). Out of LED configuration and back to ground.

    This seems to not light LED with the 3v source but works fine with 6v source, but sometimes the LED lights without power being applied, very odd timing. The LED system is going to be replaced with a Nichrome Wire wrapped around a 550lbft parachute cord. See second diagram.

    Diagram 2: [​IMG]

    The trigger for the gate on the final product is a bluetooth serial device. It will lay idle until we are ready to activate the "burn cycle" and then it will send a stream of serial data (pushing up to 3v but minuscule amperage) to trigger the gate. Then the mosfet or transistor will need to conduct 6v and up to 3-5amps of current to heat the 6-12inches or so of 28ga ni-chrome wire(approx 6.5ohms/ft) to 2000degreesF.

    If the IRD510 is not appropriate is there a possibility one of the following transistors would be an appropriate replacement or can someone suggest a way this will work with parts available from a local Radioshack. The project is in the final week and half until it needs completed and I do not have time to order from Mouser or Digi-Key or the like. I'm a Mechanical Engineering not Electrical so I am not as well versed in the intricate electronic parts. Advanced thanks for any help.

    Available transistors in my shop: TIP42(PNP), TIP120(Darlington), TIP3055(NPN), and a LM317T-Positive Adjustable Voltage Regulator. If the serial data voltage could be amplified to trigger the IRD510 then I would need advice on method of amplification.

    TechX
    Morgantown, WV
     
    Last edited: Mar 31, 2011
  2. JDT

    Well-Known Member

    Feb 12, 2009
    658
    85
    Diagram 1:

    This is a "source follower". The threshold voltage of the IRF510 is between 2V and 4V. So you need about ~3V between the gate and the source to switch the FET on. This is in addition to the voltage on the source. This is why a 3V signal does not switch on the LED.

    The reason why the LED sometimes lights without a signal being supplied is because the gate terminal is insulated from everything else (it has a very high input impedance). If it is open, it can pick up a static charge sufficient to switch on the transistor. It's not a good idea to let the gate be un-connected because it can pick up a static charge large enough to destroy the device.

    So, your input will not in fact be a steady voltage but a stream of data. I would have a diode and capacitor to store the peak voltage and a resistor to slowly discharge the voltage when the signal stops. Don't forget, the gate-source voltage will be 2 - 3 V so the source resistor is set to control the current required. The attached is a very simple circuit and if you were controlling a load of any power you would need to do something a bit better.

    EDIT: Doh! LED wrong way round!!
     
  3. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    JDT,
    That's not going to work. You have the current limiting resistor on the source terminal, which will limit Vgs. Also, I know you're trying to keep the peak voltage of the Bluetooth output on the cap, but Radio Shack doesn't carry any Schottky diodes at all, just a few standard silicon diodes like 1N4148, 1N4004, etc. and they have too high of a Vf to be useful.

    I don't have a good solution offhand.

    What is the duty cycle of the bluetooth output, and what is the actual current and voltage output? Say, at 1mA load, what's the output voltage? Need some numbers here.
     
  4. JDT

    Well-Known Member

    Feb 12, 2009
    658
    85
    OK, it was only a very hasty sketch! It suffers from another problem in that the FET will only be turned partially on with a small signal.

    Really it needs a peak rectifying circuit - I only used the schottky to maximise the gate drive - followed by a comparator with a bit of hysteresis.

    If I get time I'll draw up a better circuit!

    Is techx still there?
     
  5. techx

    Thread Starter New Member

    Sep 21, 2009
    14
    0
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    You didn't need the Google Docs part to get to the datasheet:
    http://www.rovingnetworks.com/documents/RN-41.pdf
    That just gives the specs of the bluetooth device. It's going to send/receive communication signals, and you're going to need to interpret them with a microcontroller or something.

    You're not just going to send a bunch of bits and expect that to drive a transistor, are you?
     
  7. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,170
    394
    It would be helpfull if you had a small sig. NPN, like 2N3904 2N 22222, or a comparator, but try this. If a LED is needed, connect in parallel with load with 200Ω current limiter.
     
  8. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,170
    394
    Hit submitt preme.
     
  9. techx

    Thread Starter New Member

    Sep 21, 2009
    14
    0
    Actually I opened the rovingnetworks.com pdf but because I was under chrome it opened as a google doc, thus when I copy and pasted the link it saw it as my google doc, instead of the host location. The bluetooth is only meant to connect from the Fez Embedded board to the 15 ft away remote location of this "burn" trigger. The idea was to connect the bluetooth devices and then send a serial stream to provide enough voltage to trigger some form of "switch". As I'm not an electrical engineering or technician, I am unfamiliar with a large amount of electronic elements. This is why I'm here in this forum. It seems that whenever the bluetooth device is powered up it automatically has 3v on the tx terminal though, which will not work for me. I need it to be able to be triggered. I understand that a transistor usually requires a certain amperage to work, but because the serial device provides a useless amount of amperage that was a no go. I asked a friend what he would use and he stated a MOSFET because of the simple gate voltage and minor amperage needed. So now I'm unsure how to approach the problem.

    This is what I need to happen, the Fez microcontroller will run its software to determine the timing to trigger the "burn", when it receives the signal or the countdown timer reaches the appropriate time it will connect via bluetooth to the waiting receiving unit(the one I'm trying to get to work) and once connected it should trigger the gate of the MOSFET. Then allowing the collector and emitter to draw in the 6v circuit and output this to the ni-chrome. Anything come to mind that can work with this plan? Possibly something with rts or cts?
     
  10. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Radio Shack does not stock any MOSFETs that will turn ON with a Vgs of 3v.
    They only stock the IRF510, if they even have them in stock, which is an old N-ch MOSFET that is standard level; meaning that a Vgs of 10v is required to fully turn it on; and even at that rate, it's only rated for around 5A drain current - and at that current, it will need very good heat sinking.

    Have you connected up this Bluetooth device to see what the signal levels are at the output? I am not familiar with them.
     
  11. techx

    Thread Starter New Member

    Sep 21, 2009
    14
    0
    Right the MOSFET I picked up was the IRF510 but when I read it I assumed 2-4volts gate was what I wanted and the 1-60v collector and emitter with 3-5amp current levels would be enough. I saw a solid 3v output on the tx pin of the bluetooth when powered up and while connected or unconnected. I'm wondering if I start sending serial data if the voltage drops.
     
  12. techx

    Thread Starter New Member

    Sep 21, 2009
    14
    0
    Using my computer I was able to connect to the bluetooth, open spp protocol to the device, opening putty and connecting to the bluetooth I still see a 3v signal at the tx pin, but when I send any data it begins to drop down under 1v. Any way to sense the 3 to 1 voltage drop and be able to use it as a trigger?
     
  13. techx

    Thread Starter New Member

    Sep 21, 2009
    14
    0
    maybe it was a loose wire, but now I'm seeing a solid 3v signal from the tx. this will not work at all. I'm scratching my head..
     
  14. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Like I said before, I think you're going to need some kind of microcontroller to communicate with the bluetooth device. The bluetooth device will simply output bits of data, and the uC will need to interpret the commands/text that it receives. I don't think you're going to have time to deal with that.
     
  15. techx

    Thread Starter New Member

    Sep 21, 2009
    14
    0
    Yea seems like quite a dilemma, although I might have a solution... I happen to have an Atmel 168 sitting in another project box that I'm going to steal for this. This becomes a nuisance as I will have to program both the Fez on the transmit side with one program, and the Atmel on the receiving side with a different program in a different language. Although with the Atmel I have analog io and digital io at 5.5v and would be enough to trigger a standard relay or transistor.
     
  16. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    Be very careful with the MOSFET symbol.

    Your symbol shown a P-Ch MOSFET, with the arrow pointing outward.

    A n-ch MOSFET symbol should have its arrow pointing towards the Gate.
     
  17. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    If your bluetooth serial output is indeed sitting in Mark state (level high = +3V) and stays in that state until it's receiving a signal then I don't think you need to read the serial data to turn on a BJT switch. The attached circuit was designed based on statements you made, so in order for this to work the following assumptions were made..

    (1) Prior to accessing Bluetooth the serial output sits in BREAK mode at +3V.
    (2) When receiving data or accessed, the Bluetooth serial output switches to (MARK/SPACE) serial data output that alternately swings from +3V to 0V.

    This circuit has no brains at all, so if another Bluetooth or other wireless system can pull your Bluetooth out of Break mode they will trip this circuit. The other issue with this circuit is that S1 MUST remain OFF until the Serial In is connected to the Bluetooth. This is because the Serial In must see a steady state +3V in order to keep Q3 OFF!

    Personally, (IMO) a uC that actually reads serial data would be preferable to this circuit and a heck of a lot safer. A Picaxe 08M (8 Pin Dip) would do the job easily.

    BTW, this parachute cord... Are we to assume that there is no human
    being attached to this parachute?

    Note: I didn't actually simulate this circuit with a serial pulse train. As you can see I used a 50% duty cycle square wave. I don't think this fact is an issue though.
     
  18. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,170
    394
    Re post 17, might it be better if the charging time constant ,R4-C1, were increased from 220μs to around 200ms as to require several data pulses to turn on Q? Also might add another 2N2222 to make Q2 into a darlington pr to raise its input impedance.
     
  19. techx

    Thread Starter New Member

    Sep 21, 2009
    14
    0
    to clarify this is a weather balloon with payloads, the receiver is located on the parachute cord connecting the balloon to the parachute. the problem faced is that during change of wind or after the balloon bursts and falls over top the parachute we may need to signal the cord to detach between balloon and parachute to allow the balloon to fall free from the entire system and allow the parachute to properly deploy and carry the system back to earth.

    i'm still unsure of a solution, although as I did mention if the serial transmission can not trigger a circuit then I can use a separate processor the atmel168 that I have just laying around to receive the serial data in UART and once the cutdown signal is seen on serial then the atmel168 can trigger a relay or mosfet to burn the parachute cord. does this seem the simplest and most reasonable solution to my problem?
     
  20. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Doing so may stress Q3 because of the very slow rise time, and nichrome's resistance is lower when cold. Do we know how much current he's going to draw for his nichrome?
     
Loading...