HP97 The journey begins

[teenix]

Hi all,

I was talking to my plumber son the other day and he gave me a bit of work as a gopher, go for this go for that type of thing, so I was able to make a few $ that I otherwise wouldn't have had. This should enable a bit more experimenting.

I did do some formal training years ago, but I am not an electronics engineer by any standard especially with analog. I was having some thoughts on how I could salvage what I have so far without reinventing the wheel and have to deal with even more problems.

The DC brush motor driver that I chose is a DRV8832 IC which among other things has printers listed in its battery powered applications and the drive capability and supply voltages are good. Great I thought. One of its main tasks is to monitor battery voltage and adjust the PWM motor drive signal and so keep the motor spinning at a constant speed.

It also monitors the voltage across the motor terminals and integrates that to get an error offset, so I thought that would compensate for torque as well. Unfortunately, as mentioned, it does not seem to do this.

Ideally, the printer head mechanism should have relatively constant resistance as it moves from right to left while printing. HP obviously discovered otherwise and hence had to incorporate speed control based on the voltage produced by a free wheeling dc motor and some wizardry inside the PIK chip. This setup is there to account for battery voltage and torque changes.

As I understand it, a change in load on the motor changes the speed. This will increase or decrease the motor current, leading to pretty much a short circuit if the armature stops.

I was thinking of putting a low value resistor, somewhere around 0.2 ohm as a start, in series with the ICs supply pin. The chip is designed to cope with with varying battery voltages anyway, so the idea is that as the motor current varies under load, there will be slight changes in the sensed battery voltage. This will cause the chip to vary the PWM duty cycle to try and compensate. Also if the battery voltage changes, then that is catered for as well.

Obviously there will be some fiddling so that the chip doesn't under/over compensate, but it might work as a good enough solution, and if so only some time and a few cents. A good quality current sense resistor should suffice.

I'm open to ideas :-)

cheers

Tony