Spanner / wrench on startup?

[21gg]

My mistake on the resistance bit, having a bad day at work ! (shouldn't try and surf and work at the same time!)

Ok so the leds are artificially loaded high then, fair enough, that makes sense.  If you put led indicator relay in ebay loads pop up, would just need to identify the model that matches the original. 

Gav

[Steve V8]

13 hours ago, DrRick said:

@21gg, I could be convinced that the resistance needs a tweak, however I think you've got the purpose muddled. The idea of the resistor is to simulate a 'good' halogen bulb, rather than a blown one and spoof the system into not displaying the features of a blown bulb, i.e. the hyperflashing. To do that the resistance of the LED bulb needs to mimic that of the halogen bulb. If you put it in line, surely it would still show up as a blown bulb?

Or do I have it totally wrong?

I've double checked the image Greg sent me too, and this is how I've wired my resistors in...

@Steve V8, I had to do this to my Defender when I replaced the halogen rear lights (all of them) with brighter LED units: new relays installed.

I ain't an electrician, but that doesn't look right to me,

Imagine those wires to be pipes, the current as water flowing through, and the load resistor as a half closed valve to restrict the flow.... As it is in the diagram, the valve isn't doing anything to restrict the flow, as water (current) is still able to flow down the live pipe and out of the earth pipe unrestricted. 

My next issue would be the quick wire clips, (scotch clips), the easiest and worst way to join wires, in use they cut through the insulation but will also cut several strands of the core, causing more resistance than the resistor. Soldered joints with heat shrink are the way to go.

 

 

[DrRick]

OK. You point is put well. I am however only going by what is recommended by the vendor / fabricator.

And as for the Scotch clips, I'm not going to disagree, however right now I just want the systems to work. I can do a proper-job joining when I'm in that situation.

Still nothing back from Greg either, which is disappointing

[21gg]

Just identify the indicator relay and change it, job done.

 

[DrRick]

@21gg, do you reckon with or without the resistors?

The GRP Elise units do only the relay according to the instructions

[21gg]

In theory it shouldn't make any difference,  but I personally would take them back off.

Gav 

[Brendonian]

8 hours ago, Steve V8 said:

I ain't an electrician, but that doesn't look right to me,

Imagine those wires to be pipes, the current as water flowing through, and the load resistor as a half closed valve to restrict the flow.... As it is in the diagram, the valve isn't doing anything to restrict the flow, as water (current) is still able to flow down the live pipe and out of the earth pipe unrestricted. 

My next issue would be the quick wire clips, (scotch clips), the easiest and worst way to join wires, in use they cut through the insulation but will also cut several strands of the core, causing more resistance than the resistor. Soldered joints with heat shrink are the way to go.

 

 

Your plumbing analogy is quite good, but you've misundersood the problem. The reason for the hyper flashing is that the car thinks the bulb has blown. A blown bulb in plumbing terms would be a solid blockage in the pipe - no water flow.

The issue with the LEDs is that they have a very high resistance, so they act like a restriction in the pipe, passing so little current that it looks *almost* like a blown bulb.

Adding a resistor in parallel acts as a bypass, allowing more current to flow, tricking the car into thinking the bulb is still working.

Obviously, it has to be a resistor, not just a wire, as that would allow unrestricted flow, which would melt wiring, blow fuses, and potentially damage the battery.

[Steve V8]

14 hours ago, Brendonian said:

Adding a resistor in parallel acts as a bypass, allowing more current to flow, tricking the car into thinking the bulb is still working.

I did say I wasn't an electrician didn't I?... @DrRick Your resistor wiring diagram appears to be spot on, and Brendonian should take up a teaching job if he's not doing it already. His ability to explain in a plumbing way was excellent.

Using resistors appears to be the best way forward, as changing just the relay won't protect the bulbs from overloading and will affect other filament bulbs in the circuit unless they're all LED.

Could the problem be the value of the resistors used? Fooling the lighting control module may require far more precise resistorisation than fooling the relay, if precise values are required, it may be necessary to take some voltage and resistance readings to get the correct resistors in place and delete the spanner warning light. And definitely bin those Scotch clips, they are rubbish!

TPMS....I have no idea how this system works, but if it's using the chassis or radio waves to transmit it's signal, is it possible that the 8 or so volts being dumped (blead off) to earth by the resistors could have an effect on the readings?

Sorry for finding this topic so interesting, it's like an episode of the Krypton factor.....Steve 

[Clive59]

ohms law gives the following:  60W @ 12V = 5A, so effective resistance of bulb is 12/5 or 2.4 ohms. cold resistance with a multimeter will be less of course. 

I am guessing that each bulb circuit has a small resistor in series, in the milliohm range, which will have a small measurable voltage across it when the bulb is on, and if is all is well the ECU will see this. As was suggested before, the resistance value might be critical. The 6ohm resistor in the diagram above might be too high, but really depends on the particular car and how it was designed.

The resistor being in parallel doesn't bleed off volts, but it does bleed current(and therefore power, which creates heat in the resistor). So each light will have its own built in heater of maybe 20-50W, which is quite significant in a small area. Is this of concern? And it sort of negates the power efficiency aspect of LED bulbs a bit too.

Also, the LED bulb is hopefully more than a simple series/parallel set of LEDs, but should have some driver electronics built in the globe to regulate the input voltage to a constant current source for the LEDs. This presents a complex load and will not make any sense at all with a normal multimeter.

As to why putting the LEDs causes the other problems, one thought that comes to mind is electromagnetic noise(EMF) from the driver circuits in the globes. If not designed correctly, the types of circuits can be great noise generators, because of the high switching speeds/currents in the electronics. This noise could be affecting the relevant sensors you are getting odd behavior from. Do the warning lights come on if the LEDs are off? If they do, then that is probably not the issue.

If the sensors are being corrupted by the noise, be very wary that they are working correctly.

[Brendonian]

10 hours ago, Steve V8 said:

Brendonian should take up a teaching job if he's not doing it already. His ability to explain in a plumbing way was excellent.

Thanks Steve. No, not a teacher, but I guess that as a computer programmer, my job is to break down complex ideas into elements simple enough for a computer to understand. That's teaching of a kind

 

@Clive59 I suspect you're onto something with the electrical noise. Maybe the cables should have some kind of noise suppression filters on them?

[DrRick]

On 1/11/2018 at 08:49, 21gg said:

Just identify the indicator relay and change it, job done.

 

OK, can't do that - the Evora doesn't have a separate relay for the indicators, there's a control box of some sort that does all that (at £200 a pop). This would be why the Elise kit has a relay and the Evora kit has a resistor set.

[DrRick]

9 hours ago, Clive59 said:

ohms law gives the following:  60W @ 12V = 5A, so effective resistance of bulb is 12/5 or 2.4 ohms. cold resistance with a multimeter will be less of course. 

I am guessing that each bulb circuit has a small resistor in series, in the milliohm range, which will have a small measurable voltage across it when the bulb is on, and if is all is well the ECU will see this. As was suggested before, the resistance value might be critical. The 6ohm resistor in the diagram above might be too high, but really depends on the particular car and how it was designed.

The resistor being in parallel doesn't bleed off volts, but it does bleed current(and therefore power, which creates heat in the resistor). So each light will have its own built in heater of maybe 20-50W, which is quite significant in a small area. Is this of concern? And it sort of negates the power efficiency aspect of LED bulbs a bit too.

Also, the LED bulb is hopefully more than a simple series/parallel set of LEDs, but should have some driver electronics built in the globe to regulate the input voltage to a constant current source for the LEDs. This presents a complex load and will not make any sense at all with a normal multimeter.

As to why putting the LEDs causes the other problems, one thought that comes to mind is electromagnetic noise(EMF) from the driver circuits in the globes. If not designed correctly, the types of circuits can be great noise generators, because of the high switching speeds/currents in the electronics. This noise could be affecting the relevant sensors you are getting odd behavior from. Do the warning lights come on if the LEDs are off? If they do, then that is probably not the issue.

If the sensors are being corrupted by the noise, be very wary that they are working correctly.

I've got some resistance readings now from Greg, so I'm going to take equivalent ones from my OEM lamps and GRP lamps and see what's what.

Warning lights come on irrespective of lamp status.

Temperatures - the resistors are currently NoNails taped to the back of the lamps, not in the boot itself. I could change that of course.

Greg seems to have taken this up as a challenge too, he's not seen these reactions anywhere else. All this means however is that I've probably done something stupid down the line.

The noise thing could be an interesting one, but it's the TPMS sensor that's furthest from the lamps that is most affected. I'm thinking that somewhere deep down in the wiring / controls that the indicators / TMPS / Stability Control share a module. It's only logical.

It's also only logical that the difference between the lamps has caused the issues, so if I can identify that, then I can get a fix in place.

[Bravo73]

1 minute ago, DrRick said:

Greg seems to have taken this up as a challenge too, he's not seen these reactions anywhere else.

Yes, he has. PM me, if you’d like. 

[21gg]

No relay, thats a bugger! Thats the thing with canbus cars, they get very tetchy when you start fiddling !

[Clive59]

Noise can be radiated or conducted, so distance from the TPMS is not necessarily going to help.

As to why you get this effect even when lamps are off is a real mystery.

What happens if you just have no lamp at all? Do you get a lamp warning and no other side effects with TPMS etc?

It does sound like the lamp sensor circuitry is interfering with the TPMS sensor input, and sounds more like a software bug in the controller

[DrRick]

16 minutes ago, Clive59 said:

Noise can be radiated or conducted, so distance from the TPMS is not necessarily going to help.

As to why you get this effect even when lamps are off is a real mystery.

What happens if you just have no lamp at all? Do you get a lamp warning and no other side effects with TPMS etc?

It does sound like the lamp sensor circuitry is interfering with the TPMS sensor input, and sounds more like a software bug in the controller

Haven't tried the no-lamp situation. I'll give that a go next time I get a chance. I'm getting quite adept at swapping the lamps out.