Flasher units

Looks like it wants a 1W load to flash normally otherwise it detects a lamp failure, so (If I'm not being thick) you need to put a 80 milliohm resistor in series with each of the LEDs, or failing that measure the resistance of the LEDs (not much) and alter the 1.5 kilohm resistor down a long way - possibly down to tens of ohms - get a varistor and try playing with it.

P.

On or around Thu, 26 Feb 2004 19:26:09 +0000, "Paul S. Brown" enlightened us thusly:

further study of the unit I'm looking at suggests that the shunt resistor which comes out at approx 30milliohm isn't present, unless it's done by careful calculation of the track on the PCB.

I think I'll try cutting the bulb-failure detect track, which should stop it doing that. I doubt it's possible to make it work meaningfully on LED units, which ain't gonna draw much current.

Twas Fri, 27 Feb 2004 09:41:51 +0000 when Austin Shackles put finger to keyboard producing:

you could sell flasher units modified in this way surely, buy them, cut a track, sell them to go with LED bulbs.. £££££££

remember it was my idea when you get rich from this.

-- Regards. Mark.(AKA, Mr.Nice.) ___________________________________________________________ "To know the character of a man, give him anonymity" - Mr.Nice.

On or around Fri, 27 Feb 2004 11:33:22 +0000, Mr.Nice. enlightened us thusly:

It had occurred to me... be nice if the bulb-failure could be made to work, but I suspect not. may try it, in due course - 's too cold and snowy to play with it ATM.

On or around Thu, 26 Feb 2004 16:42:07 +0000, Austin Shackles enlightened us thusly:

well...

I cut the track to the bulb-failure leg of the chip, and I now have a flasher unit that ain't bulb-sensitive.

I then took the lid of the other flasher, cut the same track on that'n, and it doesn't work any more.

Then I looked at it more closely... and it's not a U643B, it's a U 269B. Bugger.

so I went a-googling, but the only places which look like they might have datasheets for the U269B are places that want me to subscribe and pay mucho $, which I'm not that keen to do...

if anyone in the "industry" is already registered with one of these places, and can track down a data sheet for it, I'd be suitably grateful.

made by TFK, this one, same as the other - TFK appears to be now known as TEMIC, but their website is a waste of space on the subject of datasheets.

I may be able to solve it by observation of what goes where, since it's obviously a similar circuit, but it's evidently a different pinout on the IC.

On or around Fri, 27 Feb 2004 23:25:04 +0000, Steve enlightened us thusly:

unfortunately, Atmel don't list data for the one I'm currently trying to get data for. They do list it for the one I modified, and that's how I knew what to cut... but the other chip isn't the same. I'd be prepared to bet that there's a similar connection to it, but it's not the same pin.

For data, try ATMEL, who bought out a lot of Temic's lines - I am using an Atmel processor at the moment with a (T)emic part code.

The devices are looking for around 80mV across the current sensing resistor, so you could easily replace it with a larger value resistor to get what you want.

Steve

If I remember I'll hassle my distributors on Monday. Steve

On or around Sat, 28 Feb 2004 00:19:22 +0000, Steve enlightened us thusly:

I'd like to see if I can get this one going, since it does the trailer light thing. It should be possible to disable the bulb outage and leave that intact, I hope. I may yet see if I can make the bulb outage work with the LEDs, but the difference in current between 1 led and 2 led clusters isn;t going to be large.

hoverer, solved the immediate thing and no longer need load resistors. I suspect this makes the leds slightly brighter - the resistor must create a slight voltage drop, after all.