.

.

Page 1 of 3 123 LastLast
Results 1 to 15 of 34

Thread: FETS instead of relays for track connections

  1. #1
    Join Date
    Jan 2008
    Location
    mechanicsville Virginia
    Posts
    361

    FETS instead of relays for track connections

    Is anyone using FETS instead of relays? Europe? Big relays are expensive.
    Anything worth doing is worth overdoing

  2. #2
    Join Date
    Jun 2002
    Location
    St Charles, Illinois, USA
    Posts
    2,292
    I do not think you understand what a 'FET' is. FET is an abbreviation for Field Effect Transistor. They can act as a switch, but cannot handle massive amounts of amps. They CAN be used to help switch another relay or heavier duty transistor. It is not a magic solution to replace starter relays.
    The SRT circuit board uses transistors to turn the track relays on/off in their timing system.
    l.d. kelley, M.A. Ramcatlarry@aol.com

    50 year pin 1959-2009
    Racing slot cars in America
    USRA 2009 member 0028, 2017 #404
    IRRA, ISRA/USA, Hardbodies 1/24 & 1/32
    retired raceway owner(for now)
    Omni/Cidex service center

  3. #3
    Join Date
    Jan 2003
    Location
    Denmark
    Posts
    148
    I am using FET's in my Controllers.
    In CS-2's and CS-4's the FET's in the power-section are able to handle more than 250A at a case temperature of 100degrees C, and more than 350A when cold. In Peaks the current capability is a lot higher, thousand of Amperes.

    A big Advantage of using FET's is that they dont arc, as relays can do. Further relays are getting worn out when they are conducting DC-currents, somewhat less so when it's AC.

    Now: go find a 200A relay. Bring Money!

    Steen
    Last edited by smichslot; 03-08-2018 at 01:04 AM. Reason: Spelling

  4. #4
    Join Date
    Jan 2003
    Location
    Denmark
    Posts
    148
    When it Comes to switching power on and off on a track I have been thinking about using FET's for several years.
    I think that it's a good idea. FET's will last a lot longer than relays, and modern FET's will be able to carry a lot more current than even good relays.

    As and example: IRFB7530 is able to carry almost 200A (in fact: 195A) continuosly, has a internal resistance of less than 0,002 ohm, which is somewhat less than any relay.
    2 of These in parallel will cost the same as a single 70A relay.

    The only challenge using FET's for swithing track-power is that in order to switch the FET's on you need a voltage of appx 30V (but virtually no current). It's not rocket science.

    Steen

  5. #5
    Join Date
    Jan 2008
    Location
    mechanicsville Virginia
    Posts
    361
    Good info Steen. I know how they work Larry. With a hint or two from Steen, I built a few fet output controllers.

    I have a track owner who wants to add some power and switching. That got me thinking why not use fets. Searching at Mouser I see they sell fets packaged as big screw down units just for high power DC to DC switching. Trouble is, they are expensive. Dang...I just want to use the TO220 style and fab the mounting myself.

    I'll keep you posted.

    Have you seen my controller? https://youtu.be/exIVAuhThdw
    Last edited by dlatch; 03-08-2018 at 06:27 AM.
    Anything worth doing is worth overdoing

  6. #6
    Join Date
    Jun 2002
    Location
    St Charles, Illinois, USA
    Posts
    2,292
    Each physical size case has amperage limits. Light duty controllers use the to220 size and most more robust commercial controllers use the to3 size as do the track rental systems I have seen from Difalco and Cidex, so I imagine IF you wanted to offer more than 50 amps to a track a larger solid state transistor could replace a point based relay or controller as used in higher voltage industrial starter systems.
    I hate solid state surface mount circuit boards that have no repair options.
    l.d. kelley, M.A. Ramcatlarry@aol.com

    50 year pin 1959-2009
    Racing slot cars in America
    USRA 2009 member 0028, 2017 #404
    IRRA, ISRA/USA, Hardbodies 1/24 & 1/32
    retired raceway owner(for now)
    Omni/Cidex service center

  7. #7
    Join Date
    Mar 2005
    Location
    Canton
    Posts
    190
    Some typical TRIAC specifications[6][7]
    Variable name Parameter Typical value Unit
    Gate threshold voltage 0.7-1.5 V
    Gate threshold current 5–50 mA
    Repetitive peak off-state forward voltage 600–800 V
    Repetitive peak off-state reverse voltage 600–800 V
    RMS on-state current 4–40 A
    On-state current, non-repetitive peak 100–270 A
    On-state forward voltage 1.5 V
    "Ever onward."

    Nelson Swanberg

  8. #8
    Join Date
    Jan 2003
    Location
    Denmark
    Posts
    148
    Hi Larry,

    I am sorry, but some of your information is seriously out of date. The current capabilties of modern FET's, even relatively small TO220-types, far surpasses the current capabilities of bipolar transistors. Their power handling might not be as good, but that can be compensated for by adding more FET's since they are small and relativeley cheap.
    Nelson's post is clearly demonstrates the difference: a forward voltage of 1,5V for a 40A, 800V Triac. For our purpuse its far away from a modern FET.

    Steen

  9. #9
    Join Date
    Jan 2008
    Location
    mechanicsville Virginia
    Posts
    361

    something like this maybe

    I'll set this up in my shop for a test. (the Hot and Ground buss' represent the track supply). I like the diode in the 220 package as well for ease of mounting. rather than solder I like the idea of clamping the terminals with screw heads and star washers so they can be changed with hex driver.

    All comments welcome.
    Attached Images Attached Images
    Anything worth doing is worth overdoing

  10. #10
    Join Date
    Mar 2005
    Location
    Canton
    Posts
    190
    Red post on the ground buss?
    "Ever onward."

    Nelson Swanberg

  11. #11
    Join Date
    Jan 2003
    Location
    Denmark
    Posts
    148
    Why are you suggesting P-FET's? N-FET's has less resistance and can handle more current.

    What is the purpose of the diode? Virtually all power FET's today has a body-diode that is able to carry as much current as the FET itself.

    Steen

  12. #12
    Join Date
    Jan 2008
    Location
    mechanicsville Virginia
    Posts
    361
    My diagram shows an NFET on circuit high side. Positive to Drain, Source to load. Regarding the diode, I was under the impression that the body diode was not sufficient in all cases. on my controller I did have nfets blow with pwm brakes until I added an external diode. Good to know it would not be needed here. Still learning.

    Yes Nelson...the brake post is circuit ground (meaning ps negative...if not actually earth ground)
    Anything worth doing is worth overdoing

  13. #13
    Join Date
    Jan 2003
    Location
    Denmark
    Posts
    148
    Sorry, I misread the drawing (maybe because of the resistor, which really ought to go from gate to source in order t6o be sure that the FET switches of).

    Steen

  14. #14
    Join Date
    Jan 2008
    Location
    mechanicsville Virginia
    Posts
    361

    something like this maybe

    Here is the test (gate pull down to source, thank you catching that, Steen) no diode and 2 IRF630 in parallel. I guess the reason to use more than one is because the small connector legs are the weak link?

    It works. No drop I can measure Drain to Source with 13.7 supplied and a 24 volt wall wart for the gate control. The on / off looks fast and clean on the oscilloscope...LOL...don't ask me to measure it...pretty sure it's fast enough.

    I don't know how to calculate the gate inrush for 16 switching on at once. Seat of the pants will do me unless you, Steen, or some other engineer lurking can tell me. The IRF630's were laying around. The 195 amp fet mentioned above costs about 2.5 dollars so I'm open to suggestions as to the best fet to use.

    Are you sure we don't need the zing back diode?
    Attached Images Attached Images
    Anything worth doing is worth overdoing

  15. #15
    Join Date
    Jun 2002
    Location
    St Charles, Illinois, USA
    Posts
    2,292
    Still looks like you want to try to choke the power to the track through less than a 20 ga wire - smaller than the motor maybe. Since many tracks are getting 2 ga welding cable for power sources.
    l.d. kelley, M.A. Ramcatlarry@aol.com

    50 year pin 1959-2009
    Racing slot cars in America
    USRA 2009 member 0028, 2017 #404
    IRRA, ISRA/USA, Hardbodies 1/24 & 1/32
    retired raceway owner(for now)
    Omni/Cidex service center

Page 1 of 3 123 LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •