1
\$\begingroup\$

I have purchased the following H-bridge and I'm controlling it with a microcontroller:

https://www.amazon.com/H-Bridge-Brakes-Forward-Reverse-Full-Duty/dp/B07GFFVZKS

My problem is that the MOSFETs will only conduct when the bridge supply voltage Vhv is in the 10-40V range.
I need finer voltage control for my application and would like Vhv to be able to be as low as 2V - giving a 2-40V range.
Is there a way I can bias in order to reduce the minimum bridge operating voltage?
My power supply voltage is 48V. I was thinking to use a separate power supply and bias the gates on the MOSFETS. Any advice appreciated

Thanks

___________________________________________

Here is a question restatement. Use as desired.

I am controlling a Peltier device.
I use a separate power supply to supply a voltage (Vhv) of 2 to 40V smoothed DC to the Peltier device. (This is derived by PWMing a 48V DC supply and then filtering the PWM to provide smooth DC).

I wish to use a full-bridge / H-bridge to change the polarity of the Peltier device so that I can change it from heating to cooling.
The fully-assembled Full bridge PCBA that I am attempting to use will only operate over a 10V to 40V Vhv range.
I wish to determine why the device is limited to a minimum of 10V on Vhv and whether I can modify it to operate over a 2-40V range, or even 2-48V. Any suggestions would be welcome.

*Arduino.

\$\endgroup\$
  • \$\begingroup\$ Comments are not for extended discussion; this conversation has been moved to chat. Any conclusions reached should be edited back into the question and/or any answer(s). \$\endgroup\$ – Dave Tweed Oct 5 '19 at 0:45
1
\$\begingroup\$

This solution is 'highly likely' to work.

Summary:

  • Provide driver IC with an external 12V supply - see data sheet.

  • Possible bonus: The existing maximum 40V Vhv limitation is probably due to either the maximum allowed voltage of an onboard regulator IC or Vdsmax of the MOSFETs used - or both.
    When Vdd is supplied from an xternal 12V supply the regulator can be removed and the board will then be able to be operated over the range fron 2V (probably) to whatever the MOSFETs allow.

______________________

The webpage that you cite states "3. Working voltage 10V-40V".
Obviously, your 2-40V violates that requirement.

The HIP4081A data sheet specifies a Vdd of 12V typical and 9.5V to 15V operating. The supplied board needs to get this voltage from somewhere.

The board contains a horizontal TO220 package device. It is likely that this is a regulator which derives Vdd from the high voltage H-Bridge input.
The 10V Vhv minimum allows 0.5V for regulator dropout.

A likely solution to your problem is to remove or disable the regulator (if that's what it is) and to operate the HIP4081A from an external power supply - typically 12V.

________________________________________________

Added:

Can I simply desolder [the regulator] from the board and use the naked pads to apply the 12V to IC?

All care no responsibility advice :-) :
Yes, remove the regulator and feed 9.5 to < 15V to the prior regulator output.

Do look at the H-Bridge datasheet and convince yourself that it does not "care" if Vhv is say 2V, but I think it is 'uncaring' about this.

If possible check that the bootstrap gate drive voltage is safe for the FETS - but you'd hope that this was designed. You'd hope.

Note that the regulator has Vabsmax of 35V (page 5) - I guess that that accounts for the (hopeful) 40V max Vhv rating :-).

As someone noted, now you have the low VHv sorted (probably)
you can if desired PWM the HBridge to vary the voltage across the whole voltage range with either polarity.
Switch one side high and PWM the other side. Swap sides to swap polarity.

\$\endgroup\$
  • \$\begingroup\$ Is the HIP4081 going to be able to put a negative voltage in the PFET gates when the load supply is so low as to require that? The real issue here is that the bridge is being misapplied. \$\endgroup\$ – Chris Stratton Oct 3 '19 at 13:44
  • \$\begingroup\$ @ChrisStratton If the board works as I think it does then 1. There are no external PFETS and 2. It will probably work and 3. It's easy to try. || The driver IC has a minimum Vdd spec (which I missed initially) so my suggestion of "works at 10V" is correct. The IC datasheet does not limit the lower Vhv - upper is 95V. || The assumption is that 4 x NFETS are used with the high side gates being provided by charge pump. The pumping here won't add much but you'll get Vdd-a diode drop on BHB to start. || I think it's highly likely to work. I'll edit my answer slightly. \$\endgroup\$ – Russell McMahon Oct 3 '19 at 22:33
  • \$\begingroup\$ @ Russell McM you are right. I checked the Datasheet for the component in question and it is indeed a regulator: st.com/resource/en/datasheet/l78m.pdf can I simply desolder this from the board and use the naked pads to apply the 12V to IC? \$\endgroup\$ – Jordan Gallacher Oct 4 '19 at 1:26
  • \$\begingroup\$ @JordanGallacher All care no responsibility advice :-) : Yes, remove the regulator and feed 9.5 to <15 to the prior output. Do look at the H-Bridge datasheet and convince yourself that it does not "care" if Vhv is say 2V, but I think it is 'uncaring' about this. If possible check that the bootstrap gate drive voltage is safe for the FETS - but you'd hope that this was designed. You'd hope. || Note that the regulator has Vabsmax of 35V (page 5) - I guess that that that accounts for the (hopeful) 40V max Vhv rating :-). || As someone noted, now you have the low VHv sorted (probably) \$\endgroup\$ – Russell McMahon Oct 4 '19 at 11:07
  • \$\begingroup\$ you can if desired PWM the HBridge to vary the voltage across the whole voltage range with either polarity. Switch one side high and PWM the other side. Swap sides to swap polarity. \$\endgroup\$ – Russell McMahon Oct 4 '19 at 11:09
0
\$\begingroup\$

You achieve fine control by providing fine timing of the PWM signal that you feed to the h-bridge.

For example if you power the H-bridge off high voltage for it to work and you want 2V output you need a duty cycle of 2/12 = 16.6%. To get 3V you would need a duty cycle of 25%.

You state you have an inductor and capacitor on the output to smooth the ripple, that is good. However you need to operate the PWM at a suitable rate for those components. To do that you will need to use the hardware PWM generation capability of the Arduino processor. The maximum rate you can achieve with 256 level PWM on the hardware is about 60kHz. Your inductor needs to be suitable for that.

The voltage actually coming out of the H-Bridge will be approximately the same as the supply (eg +/- 12V) although there will be probably be a volt or two dropped in the H-bridge. The voltage after the L-C filter will be the low voltage you need.

\$\endgroup\$
  • \$\begingroup\$ @ Russell McMahon you are right. The Datasheet for the device in question is indeed for a voltage regulator: st.com/resource/en/datasheet/l78m.pdf can I simply desolder this component and apply the external 12V using the naked pads? Thanks for your inputs! \$\endgroup\$ – Jordan Gallacher Oct 4 '19 at 1:36

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.