Testing my module this past WE, I think I fried my H-Bridge module.
It is this kind :
To test it, I put 3 LEDs in series (I'm feeding 6,5V to the Bridge) and tried to control it via the Arduino.
At first it was working well and then all of a sudden, the LEDs were still responding (I put the Fade sketch example to test it) but went very dim. I thought I had fried the LEDs but when supplied with the 6,5V, they light up just fine.
Then I realised I might have been connecting the PWM entry to a constant +5V at times to try and see if it responded (as well the PWM being 255 at some point in the sketch as well). I read that I shouldn't go above 98% duty cycle.
Could that be the reason why it fried?
If so, can I troubleshoot what went off or is it most likely the transistors (and I'm not equipped for CMS soldering...)
Looking at that board there appears to be an octal buffer but i can see no sign of an h bridge driver ic.
Do you have a circuit diagram and show how you are driving it.
Its very easy to destroy an h bridge with shoot through when driven incorrectly.
Boardburner2:
Looking at that board there appears to be an octal buffer but i can see no sign of an h bridge driver ic.
Do you have a circuit diagram and show how you are driving it.
Its very easy to destroy an h bridge with shoot through when driven incorrectly.
The BTS7960's are complete half-bridges, have their own driver built in I believe.
You should never put 6.5V into a 5V only circuit, you risk blowing up the
Arduino and on this board the 74HC244 (or whatever the octal buffer is).
The 74HC244 is rated for a maximum supply of 6V, but more importantly
the BTS7960's are rated for 5.3V absolute maximum on the logic inputs (though
that looks a bit spurious to me)
I don't understand the point of the buffer chip at all, without it the BTS7960's can
be driven direct from 3.3V or 5V logic systems, with it there is no guarantee it will work
from 3.3V logic... Doh!
So, use 5V only, if its not working is a desoldering station and replace that 74HC244,
and hope that's the only casualty.
Im pretty sure that board has a seperate supply for the logic. I have assumed that the 6.3v is only applied to the bridge, 3x2.1 v for the leds.
If 6.3 for the logic supply then yes its fried.
Drivers on chip yes , takes care of the high side drive but does not adress the shoot through problem which will occour if the drive signals are is mistimed.
Scrub the shoot through comments, i finally found the datasheet and it appears the driver does take care of that.
Docs for this board are awful though.
The fact they are on but dim makes me think the bridgi is working but its not being driven correctly, ie he may pe pwm dimming them due to a software mistake?
I have a DC-DC buck converter providing me 6,5V from a 12V battery.
Thing is this : the fade sketch did fade in and out the LEDs perfectly until at one point they just appeared dim and ever since, with no change in wiring whatsoever, I only have very dim LEDs or when I try to control the DC motor it will control, it won't even move...
I tried to have it go by connecting the PWM input directly to a +5V but to no avail...
Iv just ordered one of those myself for a similar project which iv posted here.
Problem im having prompts me to suggest using a 6v battery if you have one and see what happens.
I know that you said leds light on the supply , but you also said you had tried a motor.
Im just wondering if the inductance has done a mischief to your psu.
Be aware that these type of h-bridges probably have a charge pump/bootstrap setup to control the high side NPN MOSFETs. This is why the pwm frequency needs to be something like 98% or less. A constant high level on the h-bridge control pin would cause the charge pump to fail, possibly causing a damaging high resistance condition in the MOSFET.
Well I will try the other side (by reversing the polarities) of the H-bridge.
@Boardburner2 :
Thing is, it seemed to fail before I even connected it to the DC motor but yes I wondered as well if the inductance could have fried my DC-DC. It doesn't seem so so far. The DC-DC is rated for 300W (although chinese) and the small motor will probably pump out 20W, I guess there is some margin
Zoomkat.
Thats very interesting and rather important i think.
Thing is i cannot find reference to it in the device datasheet.
My experience with h bridges is dated and of the discrete driver type.
