1. Home
  2. Electronics Design
  3. Brilliant idea needed!!!
  4. Page 2

Brilliant idea needed!!!

" Geocacher" schreef in bericht news:dgu6ht$g24$ snipped-for-privacy@ctb-nnrp2.saix.net...

This is the place for some brillant questions. How often is that pulse send? How long does is last? What current does it expect? How much do the LEDs take? Is the voltage of that pulse the same as the normal operating voltage? Is that 12V or 24V or other? Is that pulse also send when the lights are on?

Why can't you keep that lights on? Or do they continue to simulate that disco? What's the flash rate?

Why? I've done that kind of things some more times. It's obvious that the computer is to old to recognise the LED assemblies but apparently a modification of it - let alone a new one - is more expensive then the bulbs or resistors. I feel with you that it does not satisfy the technician but economics rules you know.

Well, telling the computer requires a current high enough to make it think it sees incandescent bulbs. Bulbs or resistors come first into mind as you already know. You can circumvent it only if you have acces to the inside of the measuring system. It's almost sure that somewhere inside is a small resistor in the pulse circuit, used to measure the current by measuring the voltage accross it. If you can replace that resistor by a larger one, the voltage accross it will increase so the system will "see" more current.

Another approach is to provide the bleeding resistor only when the computer sends the measuring pulse. That's why I asked so many questions above. When the pulse comes, the resistor is switched on by a transistor, a timer keeps it on as long as the pulse lasts and switches it of afterwards. It's some simple classical electronics once you know how to detect the start of that measuring pulse.

Can't call that innovative. It's stretching the lifetime of an otherwise obsolete computer system. Can nevertheless be attractive.

petrus bitbyter

Reply to
petrus bitbyter
Loading thread data ...

Trigger a one-shot on the turn-on of the tail lights. The one-shot gates a power mosfet on, which connects a properly sized ( 10 ohms ?) resistor across the LEDs for the on period of the one shot. Make the on period of the one-shot twice the duration of the diagnostic pulsing period.

Ed

Reply to
ehsjr

Perhaps a Zener diode + resistor (in series) would do the trick. That combination should be wired in parallel with the LED's.

The Zener voltage should be less than the voltage pulse, so that it conducts only when the pulse is sent. The resistor is chosen to provide whatever current the computer is looking for.

Mark

Reply to
redbelly 

--
Well, somebody there must know how these things are wired up, so why
don\'t you post a schematic to alt.binaries.schematics.electronic or
a wep page somewhere so that we can see what going on and come up
with a solution for you?  How\'s that for a brilliant idea?
Reply to
John Fields

The diagnostic system checks the lights individually so it boils down to the dual 3-way switch controlling a dummy load that exceeds the threshold. Both lamps ON or OFF means no dummy loading, and by sizing the current relays, CR, for the LED load, you end up with a valid diagnostic for the truck computer with no additional power consumption during normal mode of operation. If the lamps are tested as a single entity, then the circuit inverts the logic by presenting a dummy load only when one but not both LED clusters have failed, and there is no getting around dummy loading during normal operation because the input state during both modes, test and run, is the same. But I will not waste my time on the obsequious and clueless non-technical types like this particular OP.

View in a fixed-width font such as Courier.

. . . L R . | | . +--|>|-----------+-----------|

Reply to
Fred Bloggs

You didn't mention the pulse duration, but if it's short enough you might get away with a capacitor in parallel with the LEDs. At the pulse edge it looks pretty much like a short so you may need to have a low value resistor in series with it. Recall that a cold filament also has a low resistance value. With the lights on the capacitor charges up and doesn't draw any additional current.

- YD.

--
Remove HAT if replying by mail.
Reply to
YD

A little suppression would be nice: View in a fixed-width font such as Courier.

. . L -- R . | | | | . | | --- | . | | /// | . | | | ** . +------------------------------------------PO . | | | | | . | | +------------- | | . | | | | | | . | | (-)| FWB | | | . | | --------------- | | | . | | | | | | | | . | +--------|-+-||-+-|-----|--+ | . | | AC| | | |AC | | | . | | | +-|>|-+-|

Reply to
Fred Bloggs

This one will only dim the untested lamp if the test is performed during running operation- and it does open the possibility of bucking that CR off to activate the dummy load by any number of means- the CR coil drop is assumed to be 1/2 or less that of a single diode under load. This is the last post on the subject- not going to get suckered into another waste of time: View in a fixed-width font such as Courier.

. . L -- R . | | | | . | | --- | . | | /// | . | | | ** . -------------------------------------------PO . | | | | | . | | +------------- | | . | | | | | | . | | (-)| FWB | | | . | | --------------- | | | . | | | | | | | | . | +--------|-+-||-+-|-----|--+ | . | | AC| | | |AC | | | . | | | +-|>|-+-|

Reply to
Fred Bloggs

Well- does it do that by momentarily interrupting voltage to an individual lamp?

The LED lamps are stock standard in voltage and form factor so not a lot you can do there.

Reply to
Fred Bloggs

"Fred Bloggs" schreef in bericht news: snipped-for-privacy@nospam.com...

Don't these diagnostic systems keep checking while the lights are on? I would expect that..

Yes, well, that's why I suggested something that draws the same amount of current, but dumps the energy not needed by the LEDS into the battery again, as efficient as possible. After all, the OP wanted something brilliant and not something obvious ;)

--
Thanks, Frank.
(remove \'q\' and \'.invalid\' when replying by email)
Reply to
Frank Bemelman

"Fred Bloggs" schreef in bericht news: snipped-for-privacy@nospam.com...

If it can send a pulse to an individual lamp and measure the current, I assume it can also measure the current when it is on continously. Unless the dahsboard switches are still hardwired and short ciruit the diagnostic circuit. I don't know what is common practice.

--
Thanks, Frank.
(remove \'q\' and \'.invalid\' when replying by email)
Reply to
Frank Bemelman

Frank Bemelman wrote: > "Fred Bloggs" schreef in bericht > news: snipped-for-privacy@nospam.com... >

That's a good point and as far as I can find out the current is measured continuously- the pulsing the OP talked about must have been an artifact of the controller flashing a dashboard lamp when it read the whole circuit dead. In the US it is a seven "circuit" SAE standard from the tractor to the trailer with six light circuits and a common chassis ground- no VBATT is available- and these are big circuits 20, 30, and

100 amps. There is nothing the OP can do except load it down.
Reply to
Fred Bloggs

Everybody seems to forget that the lamp test must correctly detect an open circuit LED lamp. A dummy load which allows the computer to record a successful test must actually include the LED lamp in a fail safe manner. In other words, assume the dummy load is in place and the LED lamp was not connected in parallel. The computer would still carry out a successful test I would imagine. Now how meaningful would this result be?

Reply to
Ross Herbert

...

It wouldn't, but it would get the truck to start. Let's face it - LEDs are so reliable that they'll probably outlive the computer anyway. The point of this exercise was to spoof the obsolete computer in the first place, wasn't it?

Thanks, Rich

Reply to
Rich, Under the Affluence

The OP is designing trailer lights, not ones for the vehicle itself. There is no modification of the vehicle. LEDs make a *lot* of sense here (and are often used in new vehicles).

I've replaced *many* bulbs. The last safety inspection caught a blown "cyclops" on my 2001 pickup. I'd replace the thing with an LED unit, were there one in the same configuration. I don't like driving around with lights out.

The OP is trying to match a trailer to existing tractors. Newer tractors likely do accept LEDs (they're not at all unusual). Given the reliability and redundancy inherent in LEDs, replacing single filament lamps is a

*good* idea. 
--
  Keith
Reply to
keith

If, in successfully fooling the computer, the relevant vehicle design standards were being ignored or breached, then I do not agree that the proposed solution is effective or wise.

One can stay within the design standards and satisfy the vehicle computer by simply reverting to incandescant bulbs, which, to my recollection, have proved remarkably reliable for many decades. In my own vehicle (a '93 Mazda 626 V6) I have not yet had cause to replace a single incandescant bulb of any type. Of course, I realise that rough roads and severe vibration can lead to the early demise of bulbs so perhaps there would be some argument for replacing them with LEDs in some circumstances, but I wouldn't do it if I were in the OP's position.

The real solution is to pressure the vehicle computer manufacturer to design a system to cater for modern, high reliability LED lamps. Eventually, these will be so popular they will have to do something anyway.

Reply to
Ross Herbert

I fully realise that he is designing a trailer lighting system. Where I come from that exercise also comes under vehicle design standards. The fact that the existing tractor electronics is unable to cater for the additional LED trailer lighting without some sort of additional method of fooling the vehicle computer means that the modified trailer wiring would also possibly contravene vehicle design standards where the OP is located.

Yes, I dare say you have replaced many bulbs. It may depend on the roads you regularly drive on, to some extent, and the quality of the bulbs used. In the case of tractors there may be some added vibration which might lead to early incandescant lamp failure but unless the tractor is used on public roads (many aren't) the requirement for high reliability LED lamps would not be so great. It is fairly easy to find out when an incandescant bulb is not working by simply walking around the vehicle - brake lights do need an extra person however. An automated lamp test does remove that requirement and saves some time, but surely people are not so time poor or lazy that they can't do simple safety checks when going on public roads. Truckies driving road trains have to do it regularly, and they do have big rigs.

formatting link

I do understand the OP's problem. His trailers can probably be used on tractors of all ages, with or without computers. Older tractor computers may only cater for incandescant's and newer ones may handle LED's, but he can't determine who is going to use his trailers. However, if he is forced to fudge a fix to enable older tractors to work with the LED's, he may be in breach of vehicle design standards. It would be far simpler, imo, to offer both incandescant and LED options to cater for both types depending on the customer requirements. In this case, one size doesn't fit all.

Reply to
Ross Herbert

Would this make you happy, or would it still violate some design standard?

----+--------------[Rload]-------------+ | | [R] | | / +-----[comparator]---[oneshot]---| | | \\ [LED] Vref | | Gnd Gnd

Comparator triggers one-shot when it sees Vfled.

I'm clueless as to how the testing is done by the computer - I don't know if it is a one time test at turn on, or if it is continuous. If it is continuous, remove the oneshot. Ed

Reply to
ehsjr

Use a current sense R in the LED array. Amplify the voltage drop, use this to drive a current shunt across the supply to the array.

If the measurement is indeed a pulse measurement, use AC coupling.

Reply to
The Cheese Machine

Ed,

Your arrangement would appear to trick the vehicle computer into providing a fail safe test result, however the cost of including this custom modification into each LED lamp assembly would imo make it uneconomical compared to simply using incandescent lamps.

I would imagine that the additional components would also need to be integral to the LED lamp itself so that it formed a complete unit in order to meet the relevant vehicle design standards (for Sth. Africa in this case). I would also ssurmise that the design standards for the trailer would specify the type of lighting fixtures, wiring, connectors etc, but additional circuitry to make LED lighting agreeable to an existing vehicle computer would not be covered. Since the modified lamp assembly as proposed by your suggested circuit is not a standard product then it would need to be approved under the design standards before being allowed to be installed.

On the subject of LED automotive lights I did some further research;

When attempting to use LED lighting in high reliability, long life and efficient operation, it appears that some type of regulated driver circuitry is required which preferably incorporates diagnostics. Including LED lamps to work with existing vehicle computers is bound to create some difficulties particularly where the computer only caters for incandescent lamps which are standard on the vehicle itself. Trying to mix incandescants and LED's on the same vehicle appears to be a no-go situation.

A search for LED drivers for automotive applications revealed that using LED's is not as simple as just wiring a series of LEDs into an array and connecting them to the vehicle power supply. Some sort of control module may be needed.

Melexis make an IC specific to this application. However, to include a diagnostic function seems to be relatively complex.

formatting link

ST Microelectronics have an IC which does includes a diagnostic function

formatting link

Analog Devices have the AD8240

formatting link

This recent EDN article sheds a little light (pardon the pun) on the subject of automotive LED lighting and some considerations

formatting link

Reply to
Ross Herbert

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.