0
\$\begingroup\$

enter image description hereA H-bridge (12V) driven by a 555 astable multivibrator (about 2Hz) and using 2x p-channel (RFU9024) and 2x n-channel (1N60) mosfets starts up fine, on half period 1 the bridge conducts one way, then the next half period it conducts the other way, but then it stops and the side that does not switch over starts to overheat. The 555 is decoupled with a 100nF and a 47uF across the 12V supply rail; pin 5 is connected with 100nF to ground. When breadboarding this circuit it works fine, a custom built pcb does not work. All traces have been checked and re-checked for shorts, open sections, wrong connections, wrong parts, defective parts,... no errors there.

The load is resistive.

Question: what makes a h-bridge stop its oscillation after just 1 period (and in fact short circuits on one of its legs)enter image description here?

\$\endgroup\$
  • 1
    \$\begingroup\$ Very strange gate drive. Connect your oscilloscope and measure Vgs for each transistor. \$\endgroup\$ – winny Apr 5 '18 at 20:25
3
\$\begingroup\$

As the others have said the P and N channels go can on at once, so this is a bad circuit.

As this is only running on 5V, you can probably connect M3-Gate to IN1. (I assume cmos 555 with 0-5V signals) Then it is effectively two cmos invertors chained as long as Rload is not too high


It is however very interesting that it worked on your breadboard, and using a scope on your breadboard setup might be very interesting.

[Don't know why I thought M1,M2 meant motor - oh well] An actual DC motor for Rload has discontinuous currents as the commutator goes around, and flyback and back emf voltages. Protoboards may turn the current pulses into voltages (e.g. at the sources) that effect positive feedback. Or stray capacitance my be interacting with voltage spikes. Try putting a 100n capacitor across the motor (on your protoboard circuit) and see if it still works. Try a 1000uF cap across the 5V. Does it still work? (I see your pcb has 10000uF. Did the protoboard?)

Another thing that happens with fets, is they can oscillate in the MHz when you get them anywhere near linear operation, especially with any sort of inductive load. That is why you see them with a 47ohm series resistor right at the gate. Breadboards exacerbate that sort of thing, however it might be making you circuit work by adding positive feedback.

If you have discovered the self-level shifting bridge drive, your name may yet join other hallowed minimalist giants: the inventor of the Regen radio, the teledyne opamp, and the discoverers of cold fusion.

\$\endgroup\$
0
\$\begingroup\$

There exists an intermediate state where 3 FETs are (potentially) on. If V2 is zero, M1 is off. If the points M1 and M2 are both at 2.5 volts there is no current through the motor, yet M2, M3 and M4 are full on. The input V2 has no effect on this voltage, and its exact value will depend on details of the FET characteristics.

You need a proper set of drivers for your gates.

\$\endgroup\$
  • \$\begingroup\$ If it works time and again on a breadboard using the same components what would get this to work on this custom pcb? \$\endgroup\$ – Erik84750 Apr 5 '18 at 20:46
  • \$\begingroup\$ using the same component values surely there must be a specific reason for it not to work on this custom pcb? If the one pair first conducts, then switches over to the other pair yet one of the initial mosfets does not switch off, what would make it to switch off properly? \$\endgroup\$ – Erik84750 Apr 5 '18 at 20:52
  • \$\begingroup\$ You might be getting lucky on the breadboard because of series impedance limiting the shoot-through current and/or parasitic gate impedances changing the gate timing. \$\endgroup\$ – John D Apr 5 '18 at 20:52
  • \$\begingroup\$ What would be a proper way to drive these 4 mosfets when using a 555 oscillator? \$\endgroup\$ – Erik84750 Apr 5 '18 at 21:00
  • \$\begingroup\$ @Erik84750 - Go to Digikey and search on "gate driver". \$\endgroup\$ – WhatRoughBeast Apr 5 '18 at 23:43
0
\$\begingroup\$

Obvious Shoot-thru fault on M4:M2 with Vt=0.7V on 5V supply.

with Vgs= 2.5 and -2.5 for Nch and Pch respectively they will both conduct >3 amps directly across Vcc to 0V return.

(poof)

You can do this with Vt=2.5V but not 0.7V

\$\endgroup\$
  • 1
    \$\begingroup\$ Who ninja'd who? \$\endgroup\$ – WhatRoughBeast Apr 5 '18 at 23:41
  • \$\begingroup\$ I didn't see your answer when I posted mine. Is it posted in reverse order? Great minds think alike....but M3 doesn't matter. Only M4:M2 will conduct 3A more or less. or share 15 W with no heatsink ... bad form. Methinks he saw the PEH harvest design and tried to use it as a inverter. meh. Now a transformer primary between drains with enough ZL(f) as long as it oscillates. \$\endgroup\$ – Sunnyskyguy EE75 Apr 6 '18 at 0:03
  • \$\begingroup\$ But of course M3:M1 shoot thru is controllable with V2 \$\endgroup\$ – Sunnyskyguy EE75 Apr 6 '18 at 0:24
0
\$\begingroup\$

My design is junk. Impossible to get the inverted signal to work properly from the drain of M1. Only solution is to invert the signal V2 with an external inverter (in my solution using a p-channel with gate to signal and drain connected to gate of M2 and 10k load to ground) and apply that to the gate of M2

\$\endgroup\$
  • \$\begingroup\$ If the M3 gate connect to IN1, then it should be working properly at M1M3 drains. BUT this is only if you have proper 0-V1 (5V) square signal at IN1. If it does not get up to 5V, M3 will stay on. If you use bipolar 555 maybe this is problem. Pull In1 up with 1k resistor. \$\endgroup\$ – Henry Crun Apr 6 '18 at 9:02
  • \$\begingroup\$ Interesting. But this will remove the interlocking between M3 and M4. \$\endgroup\$ – Erik84750 Apr 7 '18 at 9:25
  • \$\begingroup\$ What did get this to work is add an inverter to drive the opposite leg. \$\endgroup\$ – Erik84750 Aug 27 '18 at 21:25

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.