I have a 120V DC Motor that I need to control the speed of. I am doing this through a High Voltage PNP transistor (MJW21193) rated at 250V and 16A with a peak of 30A as I understand it from the data sheet. The Diagram below is the circuit in question. I am using an optical coupler to supply the required 5V signal from the 3.3V logic. The circuit works. There is some heat but not enough that you can not touch the back of the transistor... on my test bench. The unit only runs for 6 seconds at a time.

However, when I place the unit in the field the electronics randomly fail. Yesterday I lost D5 and possibly the transistor. Thought it test correctly. The only real difference between the bench and the field is the length of cable run. Test bench length is about 20ft. Field length is 250FT.

The stranger part--at least to me--is that there are identical circuits on my circuit board. Both with independent rectifiers feeding the IN and with separate outputs connected to the out. At first my thought was that the motor itself had a problem. However when I change the motors around (they have twist lock plugs on my control box) both resonators will run off the good circuit.

I ran the system for about 3 weeks on the bench testing both circuits. There were no issues. Intact I ran them about about 400% duty (timed runs) to attempt to verify the longevity of the circuit and associated field unit.

The motors have an steady state current draw of 1A at 120VDC. With a inrush current of 3A at the same power. At least that is the power that I see with my meter.

One NOTE: Due to the 5V logic needing to be connected to the 120VDC logic (for the gate to fire) I do see the negative side of the power (always on) on the output side of my optical coupler. About -50V. The diodes are there to prevent this from back flowing.

I am an Aerospace Engineer by education. So this is starting to get above my head. When all else fails ask for help right? Can any one help me out on this one?

MOSFET Circuit

  • 1
    \$\begingroup\$ The MJW21193 is a bipolar PNP transistor, not a MOSFET. \$\endgroup\$ – Turbo J Apr 24 '15 at 15:40
  • \$\begingroup\$ PNP transistors operate on negative voltage and current. If you really are using an MJW21193 then your circuit should not work at all. Also you have "IN 120VDC+" on both IN-1 and IN-3. \$\endgroup\$ – Bruce Abbott Apr 24 '15 at 16:59
  • \$\begingroup\$ The symbol in your schematic is for an IGBT, not a PNP transistor. (Technically, an IGBT is a hybrid PNP and NMOS merged together into a single device.) \$\endgroup\$ – Zulu Apr 24 '15 at 17:27

Your power device appears to be a PNP transistor, not a MOSFET. As such, it requires significant base current to become fully saturated.

The datasheet that you linked to says that the minimum Hfe is 20. You should probably target a forced Beta of 15 to ensure full saturation.

So now you need to tell us what your motor current is. Then we can help you calculate the required base current. I'm also guessing that your opto-coupler is way undersized for the base current that you need.

Finally, your schematic diagram is extremely confusing in regards to the polarity markings on the 120 Vdc terminals. Because this is a PNP device, the Emitter should be the most Positive terminal.

BTW - The transistor symbol in your schematic diagram is the wrong symbol for the device that you are using. You should be using the symbol that is in the datasheet that you linked to.

  • \$\begingroup\$ Dwayne... thanks for the help--I corrected the original entry--however to be honest I was confused when I ordered the transistor. DigiKey (my usual source) lumps them all together. I mistakenly did not look at the type of transistor and selected the wrong one. My understanding now is that a MOSFET is a better candidate because it does not require full saturation in order to activate the gate. \$\endgroup\$ – mbtskysurfer Apr 24 '15 at 16:39
  • \$\begingroup\$ I did mislabel the diagram above. I cut it from a larger sheet and added the polarity markings. The emitter is connected to the - output of the rectifier (DC In) and the motor - is connected to the collector. That would mean that the emitter is the most positive, right? \$\endgroup\$ – mbtskysurfer Apr 24 '15 at 16:44
  • \$\begingroup\$ If this is a PNP transistor, the Emitter needs to connect to the rectifier (+) output, Collector goes to motor (+), rectifier (-) goes to motor (-). \$\endgroup\$ – Dwayne Reid Apr 24 '15 at 17:41
  • \$\begingroup\$ It's a NPN transistor. \$\endgroup\$ – mbtskysurfer Apr 24 '15 at 18:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.