Shop OBEX P1 Docs P2 Docs Learn Events
Uhf mosfet — Parallax Forums

Uhf mosfet

datacpsdatacps Posts: 139
edited 2008-01-30 02:52 in General Discussion
My next project involves a Propeller and I need to know if there is a MOSFET or any High Voltage ,, Like 1000 V device that can go upto 30Mhz.· I may be asking too much from a MOSFET .. I also did a search but most of the JFET went to 1Mhz.. Thanks

Comments

  • LeonLeon Posts: 7,620
    edited 2008-01-21 21:29
    What on earth are you making?

    Leon

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Amateur radio callsign: G1HSM
    Suzuki SV1000S motorcycle
  • datacpsdatacps Posts: 139
    edited 2008-01-21 22:56
    I am modulating a high voltage NEON tube. I don't want to use a transformer to do it. It's for an experiment I am conducting. If·I have to use a·a flyback·transformer than I will. I just don't want to if I can do it with out it. I need the High Voltage to light the Argon tube and be able to change frequencies.·
  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2008-01-22 04:29
    datacps,

    Unless you have something up your sleeve, you will need to regulate more than 1000V. A typical Neon transformer puts out 15,000 Volts and can require almost twice that as a starting voltage. Also, if you are planning on running DC, then you need to take extra precautions against the potential of thermal runaway destroying your tube. This is also seen with florescent tubes as well running on DC, but not as severe, since the starting voltage is around 400V. Thermal runaway can occur when the gas in the tube is in steady plasma. Electrically this plasma state looks like a short circuit, so the contacts can overheat and burn up, not to mention the tube itself under severe circumstances. An AC or pulsed DC stream provides a mechanical means to "break" the short circuit and prevent a thermal runaway condition.

    What you should see by altering the frequency are standing waves producing illuminated "bands" within the tube, dependant on tube length, and the chemistry of the gas within the tube.

    When I was in high school, I had the opportunity to help with the construction of a few exhibits at the science museum. One of which was very similar to what you are trying to do, although I don't know that the frequency went that high. Instead of powering a Neon transformer directly off of the mains we had a variable speed Dynamotor driving the Neon transformer. A Dynamotor is basically a motor/generator in one device. The motor was powered by the mains and could be easily varied in speed. Because of this, the output frequency could also be varied... I think 10 kHz was the max.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • Dennis FerronDennis Ferron Posts: 480
    edited 2008-01-22 08:10
    The requirements you described scream "vacuum tube"! I'm not saying that just to be nostalgic - heck they were before my time anyhow - but honestly, you really and truly need a vacuum tube for that! They still make them, especially in Russia. You can still buy new ones, or old ones on E-bay. As for how to use them in a circuit, I'm told they work much like MOSFETs.
  • datacpsdatacps Posts: 139
    edited 2008-01-22 13:03
    Thanks for your replies gentlemen. I have the tube working already with the way I set it up. It is a pulsed DC and the tube is argon with mercury so it lights up good when I stated at 1K hz. I can limit the current with a 5 watt resistor. I am using a voltage quadupler circuit with capacitors. The experiment involves frequencies changes up to 30Mhz. It works great but I am limited by the mosfet switching speed.

    "What you should see by altering the frequency are standing waves producing illuminated "bands" within the tube, dependant on tube length, and the chemistry of the gas within the tube."

    Beau, this experiments is in that range of study. As you stated it has to do with standing waves. The tube is only 12 inches long. I just have to take the frequencies up to 30Mhz to meet my Max target range. I can get the bulb to light I just need to see if there is a driver capable of that freq range. Thanks again guys for your help
  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2008-01-22 17:04
    datacps,

    At first I looked for 1000V Mosfets, and found this one... http://www.datasheetcatalog.com/datasheets_pdf/I/R/F/P/IRFPG30.shtml·...The rise and fall times look ok, but the turn Off delay would limit you to 11MHz

    Then I searched for 500V Mosfets.... Several of these had worse delays for rise time and fall time ... I found one that looks like it might work...
    http://www.alldatasheet.co.kr/datasheet-pdf/pdf_kor/68303/IRF/IRF820.html
    ...The rise/fall and on/off times look good on these...

    Note:·· You can place mosfets in parallel to decrease RdsON, and to increase the current capability.
                     RdsON = 3 Ohms
                     Id    = 2.5Amps
    Gate Threshold Voltage = 2 to 4 Volts
     
                 Rise Time = 8.6nS
                 Fall Time = 16nS
             Turn On Delay = 8.0nS
            Turn Off Delay = 33nS
    
    

    ...The bottle neck is the Turn Off Delay of 33nS = which places you right at 30.3MHz

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2008-01-22 19:52
    The MicroSemi DRF1200 is a hybrid device incorporating a (claimed) 1000V 30MHz MOSFET with a driver chip. I have no idea how much they cost or where to get them. Perhaps you could get one from the mfr. as a sample.

    -Phil
  • datacpsdatacps Posts: 139
    edited 2008-01-22 19:53
    Thanks Beau that one may work. If you run into another one with a higer voltage let me know. I will look around some more too.
  • Bruce BatesBruce Bates Posts: 3,045
    edited 2008-01-22 20:48
    datacps -

    You can request samples from the link below:
    http://www.microsemi.com/catalog/part.asp?PARTNO=DRF1200

    Regards,

    Bruce Bates

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Veni, Vidi, Velcro! - I came, I saw, I stuck around!
  • datacpsdatacps Posts: 139
    edited 2008-01-23 02:56
    WOW Thanks Phil and Bruce .. That is exactly what I am looking for. Now I can finish my project with no problem.. Thank you guys very much...
  • James LongJames Long Posts: 1,181
    edited 2008-01-23 03:25
    Phil Pilgrim (PhiPi) said...
    The MicroSemi DRF1200 is a hybrid device incorporating a (claimed) 1000V 30MHz MOSFET with a driver chip. I have no idea how much they cost or where to get them. Perhaps you could get one from the mfr. as a sample.

    -Phil
    I wonder if you could use that to build a small induction forge .

    Haven't looked at the datasheet....but would be usefull if it could do the job of the switching.

    James L

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    James L

    Partner/Designer
    Lil Brother LLC (SMT Assembly Services)
  • steve_bsteve_b Posts: 1,563
    edited 2008-01-23 13:24
    Of interest,

    Radars that use the same antenna for transmit and receive (monostatic) have something called a TR Cell or LImiter (TR being Transmit/Receive).
    Because you are using the same path to transmit and receive, there's always the risk of transmitting directly in to your receiver...200kW in to something that's looking for mW won't survive!
    So there's something called a Transmit/Receive Cell or Limiter (TR Cell)....this is an enclosed piece of waveguide that has a radioactive (below regulatory levels) gas inside that will ionize upon large power levels. (Ever see those gas charged lightning arrestors...same idea)
    So, when transmitting this little guy ionizes and essentially blocks (not perfectly, but good enough) the transmit pulse from going down the receiver path.

    The caveat to this device, is that it's not instant on/instant off! There's a rise/fall time to it's operation.
    If you look at your local NexRad image, you might see a little donut in the center of the screen (essentially where the radar is)....this is, in part, caused by the TRcell being active and blocking any energy before the receiver.
    If you see the donut getting bigger, it's a sign that the cell is starting to have problems....if the cell fails, well, there won't be any data on the image as you've already blown your receiver!

    Anyhow, to the point....the roll off time for these things are differently from device to device (assuming different gases inside).
    I'd suspect a similar thing using a neon tube!
    What's your duty cycle and pulse width? I suspect a longer tube would have a longer 'turn off time'....
    Are you trying to put a 30Mhz carrier on top of a 60Hz drive signal?

    Cheers

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    <FONT>Steve



    What's the best thing to do in a lightning storm? "take a one iron out the bag and hold it straight up above your head, even God cant hit a one iron!"
    Lee Travino after the second time being hit by lightning!
  • datacpsdatacps Posts: 139
    edited 2008-01-24 04:11
    Thanks Steve
    I am running a 50% duty cycle and no carrier wave it just a straight DC·pulse or freqout signal through the tube. I had my 10 inch bulb working upto 5Mhz but I need to go upto 30Mhz.· I have to test the output when I get it working and see what effects it will have on the lenght of the wires and the bulbs.
    Thanks again
  • steve_bsteve_b Posts: 1,563
    edited 2008-01-24 21:08
    Have you tried a longer tube at 5Mhz? That might correlate a length with frequency (like wavelengths and antenna lengths).

    How big does your tube need to be? IF the length changes things, you might have to shorten your tube a bit....will be neat to see!

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    <FONT>Steve



    What's the best thing to do in a lightning storm? "take a one iron out the bag and hold it straight up above your head, even God cant hit a one iron!"
    Lee Travino after the second time being hit by lightning!
  • datacpsdatacps Posts: 139
    edited 2008-01-25 02:45
    Steve when I get it all working I will post the result. I am going to start the prop program and get that out of the way.. I am going to try the 500V mosfet too. Thanks for the info and your help.
  • datacpsdatacps Posts: 139
    edited 2008-01-30 02:52
    I have one more question on the MOSFET. In the DDS Chip they use a AD8008 RF amp . I was wondering if that AMP will be a good choice to drive the MOSFET or do you know of a better choice. Thanks
Sign In or Register to comment.