Come up with a brilliant solution .......

"quietguy" wrote in message news: snipped-for-privacy@REMOVE-TO-REPLYconfidential-counselling.com...

Here's what you wrote:

Except now you don't know when the LED lamp array has failed, so the diagnostic is useless.

" Geocacher" wrote in news:dgvroi$j5c$ snipped-for-privacy@ctb-nnrp2.saix.net:

I think it would be best in the long run to solve this by the cheapest way, and that is to lean on the computer suppliers for the manufacturer to modify their code or hardware. Your business seems to be in the middle of a new trailer lighting paradigm and it is obvious that the future lies in solid state lighting. So what about more effort in debating with the computer makers? They must be aware of the impending change, they may even be planning engineering solutions now. If they are stone walling, maybe point out the marketing advantage of an LED friendly system would give them an advantage over competitors. After all, YOU are the customer of these suppliers and they need to listen to YOU.

Aside from the fact that most manufacturers are based in the US/Europe and couldn't give a flying f*ck about the Aus market. From what I can gather is that for some trucks you cant even get a program that matches some certain australian engine-transmission combos. Why would they care about a LED lamp on trailer?

OP is apparently in South Africa so there's even less chance....

Ken

A simple comparitor cct using an op amp. Everyone wants to turn this into a major project.

the

lamp

back

including

Stupid)

Let

Per light? Sheesh, who's making it a major project?

Ken

I have it down to one comparitor and 1 transistor plus a few passives. I reckon i can get it down to at least 3 transistors, perhaps 2 and no comparitors. Mind you. thats in spice, not in the real world, and thats with a few glasses of red wine too....Dang its been a while since I used spice...

It starts getting complicated because ; - LEDs can fail short-circuit OR open-circuit. - To get adequate luminance, the LED-lamp will probably consist of several parallel 'series-strings', where each 'series-string' would have several LEDs and a current-limiting resistor or driver connected in series fashion. - The environment is not electronic-friendly, what with noise, spikes and wide operating temperature-range.

Spice your wine...? Cool! :-)

All that electronics so you can use LED's in a trailer light - I think this project is in serious need of management oversight. :-)

Cheers.

Ken

After thinking about the situation I am beginning to think the problem may be more complex than it seems at first glance.

I presume the OP is talking about the sort of trailers that are attached to 'semi-trailer prime movers, and that the computer thingo is located in the prime mover. If this is so, then somehow that computer must have some way of telling how many tailights/sidelights/number plate lights/brake lights?/clearance lights etc are located on the trailer. Do these vary from trailer to trailer? And remembering that one practice is for (say on the Sydney to Melb run) there to be a change over of trailers at Tarcutta, then the computer must be able to know how the lighting setup of any trailer that might be connected to the prime mover.

Is any of this relevent to the OPs situation?

David

two bob wrote:

"quietguy" wrote in message news: snipped-for-privacy@REMOVE-TO-REPLYconfidential-counselling.com...

Having just fired off a reply to your other post, this one deserves some credit. That's a very valid point because either the controller in the rig gets told how many lights on the trailer or it is assumed by the manufacturer that there is a general response curve by a lamp, and that any number (within some limit, no doubt) will behave in a certain fashion. The LED's of course react differently. An intersting point is what happens if only *some* of the lamps are replaced with LED's. At what point does the system fail to detect the lights?

Ken

You could get even trickier, use 1 4bit uP and a transistor. Even better if the uP has an ADC on board..

to

way

measures

DC

that

that

thread.

looking

just a

has a

the

this

There are some neat networked chips that fit into an RJ45 socket - now if they just had wireless......

Ken

seems to me all it would take is a capacitor in parallel with the lamp and possibly a resistor to limit the current a bit...

my guess is try 100-200 ufd and if it works add in a bit of resistance until it quits working (use a 1k-10kpot) then back up till it works again and measure the resistance...replace with a similar value fixed resistor and your done.

"quietguy" wrote in message news: snipped-for-privacy@REMOVE-TO-REPLYconfidential-counselling.com...