Hi All,

One of the circuit boards where I work is experiencing occasional failures of two motor drive ICs. One is a Toshiba TA8428K, the other is a Toshiba TA8429H. We sometimes have boards come back with the components burned out, looks like they've overheated. In the design the TA8428K does not have a heatsink, the TA8429H has a small heatsink. I have not yet measured the actual current drain of the motors they drive.

Looking at the datasheets for the parts they appear to include a thermal shutdown block. Shouldn't this be preventing the parts from overheating?

Thanks,
FW

 

Hello,

What current is drawn by the load?
The TA8428 is capable of 1.5 A cont and 3 A peak.
The TA8429 is capable of 3 A cont and 4.5 A peak.

I have attached the datasheets so the others can also have a look at them.

Bertus

 

check ALL the power connections to the motor and be sure nothing is corroded or loose. Bad connections can be hell on a micro as in; when a large motor draws surge currents through a resistive connection(corrosion), or a connection becomes resistive while under a large load(loose). Those situations can create voltage and signal conditions which will KILL silicon devices not properly protected, even then sometimes they die anyway.

 

Probably a good place to start looking for the problem is page 13 in both data sheets. It gives information that HAS to be followed when using these IC's.

 

Do you have a schematic of the circuit they are in?

Failing that, a good, clear photo of both sides of the board?

For the motors, what are their ratings/type? Voltage and current draw?

 

I have not made current measurements yet but hope to do so today. I do have a schematic I can post perhaps tomorrow.

Kermit2 : I don't fully understand how the loose connections could damage the parts. Is the theory that a loose connection would cause surges in current drain, and the surges would kill the part?

Thanks all,
FW

 

You'll want to review your current limiting. Your designed load should fall within the packages ability to dissipate the heat.

Also check for suppression on the output terminals. Feedback due to regen or noise should be shunted.

 

Hello,

Also take care about enough time between forward and reverse operation.

Bertus

 

Thanks guys.

I was able to measure the current drain. The motor connected to the first chip draws .37A normally but 2.4A stalled, so the stall load is clearly a problem. Mechanical jams in the system could cause stalls.

The motor connected to the second higher power driver draws .15A and .83A when stalled. I've seen evidence on some of the PCBs of overheating of this driver chip (discolourations on the circuit board) and it's only fitted with a small heatsink.

I'll post the circuit diagram here if I can upload it as an attachment.

GetDeviceInfo : what sort of components would you use to add suppression at the output pins? Capacitors? I don't fully understand why that's needed on the output.

 

Sorry, I was having two conversations at once. I wanted to attach the regen to the outputs, and noise to the supply side. Not sure of your opperating environment, but if AC parallels your outputs, you could consider noise on the outputs as well.

 

Thanks guys.

I was able to measure the current drain. The motor connected to the first chip draws .37A normally but 2.4A stalled, so the stall load is clearly a problem. Mechanical jams in the system could cause stalls.

The motor connected to the second higher power driver draws .15A and .83A when stalled. I've seen evidence on some of the PCBs of overheating of this driver chip (discolourations on the circuit board) and it's only fitted with a small heatsink.

Are the motors the same size/type? One drawing over twice as much current isn't good if they are supposed to be identical.