(Image: Header Graphic)

Tuesday, September 22, 2020

Doug's Domain

Doug Vetter, ATP/CFI

Like what you see?

Donations to dvatp.com are now processed via Stripe. Like this site? It's easier than ever to show your appreciation.

Wednesday, May 20, 2015

OBC Backlight Repair

(Image: Closeup of OBC main display backlight failure)Several years ago I observed the main display backlighting of my 18 button OBC turning off intermittently and no amount of fiddling with the unit would fix the problem. As randomly as it would turn off, it would turn back on again. Since it worked more often than not and I had bigger fish to fry I tolerated the occasional outage and got on with my life. Recently, as winter changed to spring, the display failures become more frequent (and hence annoying) so I decided to address the problem.

From a prior discussion with the developer of the OpenOBC project I knew that the backlighting was controlled by a couple FETs -- one for the main display and another for the clock display. I naturally assumed at least one FET was at fault so a few weeks ago I decided to pull the OBC out to see if I could get a part number off the devices and either source those units or an equivalent.

As it turned out the board uses two different FETs in varying capacities but the ones connected to the two lighting circuits were identical. Both were made by Harris Semiconductor and, as expected, long since discontinued. I did, however, find a couple suppliers with stock and purchased the devices despite minimum order requirements, thinking that if the repair was successful I could provide the devices in smaller quantity to others looking to complete the repair themselves or even provide the repair as a service.

Although the parts came in several weeks ago I wanted to wait until the backlight failed again before repairing the unit so I could more quickly determine if the device replacement actually fixed the problem. As I hopped in the car today I found the backlight inoperative so when I got home I popped the unit out of the dash and brought it up to my bench. Looking at the unit more closely I discovered three wires on a ribbon cable leading from the backlights in the display back to the FETs...two for power and a common return. In an effort to localize the fault I decided to replace only the device associated with the main display and accomplished that repair today in less than 20 minutes despite one of the FET leads being a bit stubborn.

The good news is that when I returned the unit to the dash I found the backlighting working again but the display still remains a bit dimmer than the clock display (or the clock display is brighter than the main display, depending on how you want to look at it). I seem to recall this as a consequence of being cheap and replacing only the clock bulb the last time it failed, so it may be something easily solved by replacing all the bulbs. Or it may be evidence that the other device needs to be replaced as well. If another device on the board is flaky it might be stealing gate drive from the main display FET and thus preventing the FET from fully turning on (or reaching saturation). I suppose time will tell. If the display fails again I'll have my answer, but until then, I'll enjoy watching the temperatures on the external temperature display climb as we move solidly into the summer months.

Mileage 259570