UK Monster Owners Club Forum » .: Technical :. » Electrics » Warning light is dim after fitting led indicators.

Reply
 
Thread Tools Display Modes
Old 01-08-2020, 03:07 PM   #1
utopia
No turn left unstoned
 
utopia's Avatar
 
Join Date: Jun 2010
Location: leicester
Bike: M750
Posts: 4,543
Warning light is dim after fitting led indicators.

I've been trying to neatly fit a set of panigale led indicators to my yr 2000 M750 for ages.
After much pondering I finally managed to attach the rear ones to the teatray without them looking like they'd been slapped on willy-nilly.
Front ones were relatively easy.
I know .. there should be pics at this point but I'm lacking in that dept, I'm afraid.
Anyway, to get to the point ....
I fitted an led-specific relay to correct the flash speed.
This was also quite neat as it was exactly the same dims as the std relay so it fitted straight into the rubber strap thingy where the old one was.
It was one of those with a speed adjustment knob so it was dead simple to reset the flash speed.
So far so good and I thought that was the job done and dusted.
It had taken some perseverance and I was close to abandoning the project on a few occasions but I persisted as a sort of memo of Buzzbomb because we had been discussing the issue over many months.
Anyway, and here's the rub, the indicator warning light is now much dimmer than it used to be.
In fact it is hardly visible in daylight on a bright day.
Now there are those in the club who already know that I can sometimes be a tad forgetful about cancelling my indicators these days
In fact this can be quite dangerous and it is something that I'm actively working on.
The last thing I need is a dim warning light.
So the question is, do any of you electric-savvy members have an explanation for this new dimness (the bike, not me) and better still any suggestions for a cure ....?
So far, my thoughts are that one solution might be if I replaced the original flasher relay and corrected the flash speed with resistors instead, but that's not quite so neat as using the led-specific relay.
I also considered replacing the bulb in the warning light with an led, but I've got no further with that than a vague pondering.
Thanks in advance.
utopia is offline   Reply With Quote
Old 01-08-2020, 04:03 PM   #2
gemantsdad
Registered User
 
gemantsdad's Avatar
 
Join Date: Jun 2013
Location: Wolverhampton
Bike: S4Rs
Posts: 53
Hi, a couple of guesses, could there be a voltage dropper in the flasher to work with leds 4.8v, or the adjuster is reducing the 12v input voltage to set the flash speed, but the incandescent bulb in the dash on the output circuit isn't receiving 12volts
gemantsdad is offline   Reply With Quote
Old 01-08-2020, 04:46 PM   #3
Mr Gazza
Lord of the Rings
 
Mr Gazza's Avatar
 
Join Date: May 2013
Location: Norwich
Bike: M900sie
Posts: 5,766
That would be my thinking too. I imagine the adjustment is stepping down the current and so depriving the warning light.

Easy test for this would be to wind the relay back up again, thus increasing the winker speed, but observing the brightness of the warning light at the same time.. Obviously not a cure, but an indicator as to watts going on. (couldn't help it!)

Messing with the wattage of the warning repeater might not be any good?
As far as I know the repeater is connected to the indicators both sides in series, and does not light the opposite side indicators because the resistance in the repeater prevents enough current from reaching the other indicators.. Or something like that?

If you were to replace the repeater with an LED it might not work at all since LED's are very voltage sensitive and also polarity critical (ie, a diode) and I think the current needs to be able to flow either way depending on which side is switched on?

I've probably exposed my slender grasp of what's happening here, but someone could probably explain it better or blow it out of the water?

As an aside, would you consider an audible repeater? Perhaps it would be inaudible or not have sufficient current to work? Or just be too irritating?
__________________

Last edited by Mr Gazza; 01-08-2020 at 04:48 PM..
Mr Gazza is offline   Reply With Quote
Old 01-08-2020, 05:57 PM   #4
utopia
No turn left unstoned
 
utopia's Avatar
 
Join Date: Jun 2010
Location: leicester
Bike: M750
Posts: 4,543
Thanks for your responses so far.
My thinking as to the cause of the problem was also along similar lines to your suggestions.

I tried winding the flash speed back up but it seemed to make no difference to the brightness of the warning light.
In fact I wound it between max and min flash speeds and could detect no change in warning light brightness.

I have yet to fully digest the nature of the circuitry and also the likely effect of messing with the wattage of the indicator bulb.

However a cursory glance at the wiring diagram would suggest that current needs to flow in each direction through the warning light according to which side (left/right) is flashing .. so yes (or do I mean no) it would seem that an led warning light will not work due to it being uni-directional.

I have briefly considered an audible warning device but yes it would be annoying, no it might not be loud enough and also it doesn't fit with my minimalist design philosophy.
My memory isn't so bad that a decent, bright warning light and a bit of focus shouldn't be enough.

Looks like the led warning light idea is a non-starter though .. so you've saved me unnecessary time and expense there.
I'm still wondering if a higher wattage conventional bulb might be worth trying. That might be just because my grasp of the situation is vague though.


ps. the original bulb is a type T5 (also designated 286) and is marked 12V 3n OS ..... correction, the marking is actually 12 / 3n OS
I have no idea what the 3n OS bit means.

Last edited by utopia; 01-08-2020 at 06:41 PM..
utopia is offline   Reply With Quote
Old 01-08-2020, 06:22 PM   #5
spuggy
Registered User
 
spuggy's Avatar
 
Join Date: Apr 2016
Location: Farnborough
Bike: M900sie
Posts: 171
This might explain the problem:

Quote:
The (relatively ) short answer to your question is that you can add some load resistors and it'll almost certainly fix the problem you're describing. You don't need to do all four corners; I'm using load resistors on my rear turn signals only.

...

The long answer is to understand why this happens. Have you ever noticed that the turn signal indicator blinks out of phase with the turn signals themselves? (i.e., if you've turned on a turn signal or hazard lights, the indicator is on when the turn signal(s) are off and vice versa)

The main thing to notice (if you can read schematics) is that when the turn signal lever is switched on (either direction), the indicator light will be off when the relay closes (and the turn signal lights turn on), but it should turn on when the relay opens because the indicator light still has 12 V connected to one side, but the other side is then grounded through the turn signal lights, which (in their original incandescent form) are much lower resistance than the indicator bulb. So... raise the resistance (and the LEDs are effectively much higher resistance than the original incandescents), and... indicator bulb no longer has a good low-resistance path to ground. QED

There are probably other ways to solve this problem (e.g., reworking the way the indicator light works), but it probably is most expedient to just solve it with load resistors.

I'd suggest experimenting with the highest value resistors you can that get the job done. I can't tell you what values I used (I'm at work and my car is at home), but 25 Ω load resistors and 6 Ω load resistors are quite common. Remember: the lower the resistance, the more current the circuit will draw and waste in the form of dissipated heat.
spuggy is offline   Reply With Quote
Old 01-08-2020, 06:48 PM   #6
spuggy
Registered User
 
spuggy's Avatar
 
Join Date: Apr 2016
Location: Farnborough
Bike: M900sie
Posts: 171
And for an example of what they mean by "reworking how the indicator light works", this is more complicated, but actually might be the "correct" fix; different symptoms, similar cause: https://www.customled.com/blogs/cust...d-turn-signals

Quote:
WHAT CAUSES THE 4-WAY BLINK PROBLEM?
This problem actually has nothing to do with the flasher relay on the motorcycle. A flasher relay can not cause, or fix this particular issue.

The problem stems from a lack of restive load (incandescent bulbs) on a blinker system that was designed for incandescent bulbs - however the best solution is actually NOT to add resistors. It is actually best to solve the problem at the source, which is the turn signal indicator lamp in the gauge cluster. The turn signal indicator lamp is the little light that blinks next to the speedometer when you are turning.

It is at the turn signal indicator lamp where both the left and right turn signal circuits are brought together. Left turn circuit on one side of the indicator lamp "filament" and right turn circuit on the other side. With the incandescent turn signal bulbs installed, the system behaves as designed where voltage on either the left or right turn signal circuits illuminates the turn signal indicator lamp. This is achieved simply by allowing the opposite side turn signal circuit from the one activated, to function as a ground path for the indicator lamp. The small amount of current that passes through the indicator lamp in the gauge cluster is not enough to illuminate the power-hungry incandescent bulbs, and they simply function as a ground path without lighting up at all. When removing these incandescent bulbs and installing LEDs, you have removed this ground path - which usually leaves the turn signal indicator lamp inoperative - AND you have introduced voltage to the opposite side turn signal circuit which is enough to illuminate the super efficient LEDs.

OK, SO HOW DO I FIX THE 4-WAY BLINK PROBLEM?
The pure solution to the 4-Way Blink Problem (not the simple one, but the correct solution) is to diode isolate the two turn signal circuits leading to the gauge cluster, and provide a dedicated ground for the indicator lamp.

This is done by finding these two turn signal circuit wires leading into the gauge cluster (typically the same wire colors as found at the blinker lamps themselves). Once you properly identify them in the harness leading to the gauge cluster, you cut them and install a Diode (1N4001) in each circuit (with the gray band towards the gauge cluster) on the two wires feeding power from the turn signal circuit. You will then combine the outputs of the diodes into one of the two wires leading into the gauge cluster, and ground the other wire going into the gauge cluster.
The diodes will keep electricity from being able to cross over from one turn signal circuit to the other, but it will allow the two circuits to be combined to power the indicator lamp properly.
We all like diagrams, right?



Even though you don't have this exact issue, I think this'd also fix the "dim indicator" issue - by providing a good path to ground for the bulb (or LED).
spuggy is offline   Reply With Quote
Old 01-08-2020, 06:52 PM   #7
spuggy
Registered User
 
spuggy's Avatar
 
Join Date: Apr 2016
Location: Farnborough
Bike: M900sie
Posts: 171
Quote:
Originally Posted by spuggy View Post
And for an example of what they mean by "reworking how the indicator light works", this is more complicated, but actually might be the "correct" fix; different symptoms, similar cause: https://www.customled.com/blogs/cust...d-turn-signals



We all like diagrams, right?



Even though you don't have this exact issue, I think this'd also fix the "dim indicator" issue - by providing a good path to ground for the bulb (or LED).
Oh, and it would allow you to replace the indicator bulb with an LED - as current would only flow through it one way...
spuggy is offline   Reply With Quote
Old 01-08-2020, 07:42 PM   #8
spuggy
Registered User
 
spuggy's Avatar
 
Join Date: Apr 2016
Location: Farnborough
Bike: M900sie
Posts: 171
Incandescent bulbs are just a coil of wire. A little heater coil that glows white-hot. They don't care which direction the current is flowing, they work either way round. They have a fairly high resistance - which is converted into heat as well as light.

Many old indicator circuits are designed around using bulbs; as these aren't polarity-sensitive at all, as soon as you provide current with a path to ground, it's "job done"...

An LED is just a diode that happens to emit light (they actually generate small amount of electricity if you expose them to light as well, but usually not enough to be useful)

The voltage the LED itself wants is determined by the doping of the substrate - and thus the colour they produce. White LEDs only work with 3V, red ones want 1.9V, UV ones 4.1V. They either have a current-limiting (aka "ballast") resistor built-in to the package to allow it to be directly connected to a 12V circuit, or sometimes multiple - eg 4 - LEDs are wired internally in series in the package.

LEDs only allow current to flow in one direction. Even with the ballast resistor, they have a very low resistance to current flow in this direction - which is why they draw so little power.

(A 2W bulb, at 12V, draws 160ma, and a 3W bulb 250ma. A typical white miniature LED uses 10ma - and probably brighter than a 10W bulb. It'll generate a little warmth in the ballast resistor - but still a lot less power overall than a bulb.)

However, current flow in the other, "wrong", direction, with an LED has a very, very, high resistance - think of it as infinite. Much, much more resistance than a bulb. That's the "diode" part of "light emitting diode".

With non-LED flasher relays and no load resistors, the relay hyper-flashes because the the forward resistance of the LED - even with the ballast resistor - is nothing compared to a bulb. So it behaves as though a bulb is out.

If you resort to load resistors, you pretty much throw away the "low current draw" advantage of LEDs (the load resistor just turns the extra current into heat), but you do get to keep the "I don't blow bulbs", "they're brighter" and "they respond almost instantly" benefits.
spuggy is offline   Reply With Quote
Old 01-08-2020, 07:55 PM   #9
utopia
No turn left unstoned
 
utopia's Avatar
 
Join Date: Jun 2010
Location: leicester
Bike: M750
Posts: 4,543
Ok .. for the sake of being able to respond promptly I'm going to say .. thanks very much for that and it seems to explain the situation.
I say "seems" to explain simply because I need to study it all a bit longer before I can say that I've fully grasped the whole picture.
However, I'm now pretty much up to speed with the thing about the other side of the indicator circuit serving as an earth path for the warning light.
Also the diode installation looks at first glance to make sense.
Really, I just wanted to respond and say thanks without necessarily giving the impression that I fully understand the full picture just yet.

Perhaps of interest, I did observe the timing of the illumination of the indicators compared to the warning light and found that they were in fact in phase on my bike as it currently stands, not out of phase as the first quote suggested they would be .. but I can't profess to currently have an understanding of why either should be the case.

Also, I think I have determined that the original warning light bulb is (probably) 1.2W.
I have found some which are rated at 2.3W and I might just try one of those (at under£4 for two, its money well wasted).
I'm just curious to see (ie prove) what effect they might have.

Finally , its unfortunate eh ?, that the quotes use the word "indicator" to describe the warning light .. thus creating an obvious potential for confusion with the turn signals (or indicators) themselves.
Here's hoping that the author is better at electrics than they are at the use of language.

Anyway, thanks muchly to everyone for your help.
And please feel free to add further comments if you feel the need.
Meanwhile, I'll ponder my next move.


ps. The above was written prior to reading the last post .. #8
utopia is offline   Reply With Quote
Old 01-08-2020, 08:04 PM   #10
utopia
No turn left unstoned
 
utopia's Avatar
 
Join Date: Jun 2010
Location: leicester
Bike: M750
Posts: 4,543
Quote:
Originally Posted by spuggy View Post
With non-LED flasher relays and no load resistors, the relay hyper-flashes because the forward resistance of the LED - even with the ballast resistor - is nothing compared to a bulb. So it behaves as though a bulb is out.
That's the bit I've never really understood.
I've always reasoned that if the system behaves as if a bulb is out, that would tally with an ultra high or infinite resistance, not the ultra low resistance of an LED.
In other words, it would behave as if the bulb had been short circuited (low resistance path) rather than blown (infinite resistance path).
utopia is offline   Reply With Quote
Old 01-08-2020, 08:18 PM   #11
spuggy
Registered User
 
spuggy's Avatar
 
Join Date: Apr 2016
Location: Farnborough
Bike: M900sie
Posts: 171
Quote:
Originally Posted by utopia View Post
Perhaps of interest, I did observe the timing of the illumination of the indicators compared to the warning light and found that they were in fact in phase on my bike as it currently stands, not out of phase as the first quote suggested they would be .. but I can't profess to currently have an understanding of why either should be the case.
That is interesting... I've never really paid any attention to my Ducatis, but most cars are out-of-phase.

What voltage do you actually see on the lead for the tell-tale light?

Maybe it's some interaction with the flasher relay - how many wires?

Quote:
Originally Posted by utopia View Post
Also, I think I have determined that the original warning light bulb is (probably) 1.2W.
I have found some which are rated at 2.3W and I might just try one of those (at under£4 for two, its money well wasted).
I'm just curious to see (ie prove) what effect they might have.
I'd caution to check to see just how hot they get before installing them... Think those were used in one of my cars as panel illumination for the heater - and the plastic tubes that held them in place were deformed and melted from the stock bulbs when I swapped LEDs in...

Quote:
Originally Posted by utopia View Post
Finally , its unfortunate eh ?, that the quotes use the word "indicator" to describe the warning light .. thus creating an obvious potential for confusion with the turn signals (or indicators) themselves.
Here's hoping that the author is better at electrics than they are at the use of language.
They're 'Merkins. They call the blinky lights that aren't in the dash "the flashers"...
spuggy is offline   Reply With Quote
Old 01-08-2020, 09:21 PM   #12
spuggy
Registered User
 
spuggy's Avatar
 
Join Date: Apr 2016
Location: Farnborough
Bike: M900sie
Posts: 171
Quote:
Originally Posted by utopia View Post
That's the bit I've never really understood.
I've always reasoned that if the system behaves as if a bulb is out, that would tally with an ultra high or infinite resistance, not the ultra low resistance of an LED.
In other words, it would behave as if the bulb had been short circuited (low resistance path) rather than blown (infinite resistance path).
Good lord, I'm too young to understand all that analog stuff!

Heh. As-if...

https://auto.howstuffworks.com/turn-...he%20resistor.

Quote:
When you push the turn-signal stalk down, the thermal flasher connects to the turn-signal bulbs by way of the turn-signal switch. This completes the circuit, allowing current to flow. Initially, the spring steel does not touch the contact, so the only thing that draws power is the resistor. Current flows through the resistive wire, heating up the smaller piece of spring steel and then continuing on to the turn-signal lights. At this point, the current is so small that the lights won't even glow dimly.

After less than a second, the small piece of spring steel heats up enough that it expands and straightens out the larger, curved piece of spring steel. This forces the curved spring steel into the contact so that current flows to the signal lights unimpeded by the resistor. With almost no current passing through the resistor, the spring steel quickly cools, bending back away from the contact and breaking the circuit. The cycle then starts over. This happens at a rate of one to two times per second.
Typical spec for these relays is 90 +/- 5 cycles/minute - when the load is 2x 21W bulbs wired in parallel, as per this example diagram:



So the resistance from the bulbs in parallel is lower than two bulbs in series (parallel resistance is calculated as 1/Rt = 1/R1 + 1/R2 + 1/R3 + ...).

A modern one will still operate with one bulb out - but if you've blown a 21W bulb, it'll hyperflash - because a single bulb means more resistance, so with less current flowing, bi-metallic strip doesn't get as hot - and cools down faster.

LEDs draw so little current without load resistors that an electro-mechanical relay might not operate at all - but if it does, it hyperflashes, same as if you've got a bulb out.
spuggy is offline   Reply With Quote
Old 01-08-2020, 09:25 PM   #13
utopia
No turn left unstoned
 
utopia's Avatar
 
Join Date: Jun 2010
Location: leicester
Bike: M750
Posts: 4,543
I couldn't tell you what voltage I see on the lead for the tell tale light cos my multimeter has gone intermittent after too many winters in a damp shed.
Flasher relay is three-wire.

So then I checked my modified Dommie (Honda) which is fitted with monster indicators bearing LED bulbs, an aftermarket, LED-specific flasher unit (also 3-wire) and a Koso dash unit.
Those flash out of phase with the warning light.

So then I checked my car (I weakened and bought one last winter).
Bog standard Citroen C1 (so just half a car, really).
I was expecting out of phase but they flash in phase with the warning light.


Good point about the heat in the 2.3W replacement bulbs ... I'll check.
Then again, its an intermittent flash rather than a continuous illumination so maybe the odds are in my favour as the same bulbholders hold continuously lit warning lights for lights on, main beam etc.
utopia is offline   Reply With Quote
Old 01-08-2020, 09:50 PM   #14
Mr Gazza
Lord of the Rings
 
Mr Gazza's Avatar
 
Join Date: May 2013
Location: Norwich
Bike: M900sie
Posts: 5,766
Quote:
Originally Posted by Mr Gazza View Post
I've probably exposed my slender grasp of what's happening here, but someone could probably explain it better or blow it out of the water?
Blimey Spuggy you have certainly explained it better! And not really blown me out of the water either.
I figured out the redundant side of the circuit creating the earth whilst out in the workshop, but I was kind of half way there in my post.

Jeff I have some super bright green LEDs which need to work in sets of three and need 12V, you're welcome if they will work for you.

Request to admin... Can we please make this sticky as it's such a good explanation?
__________________
Mr Gazza is offline   Reply With Quote
Old 02-08-2020, 01:33 PM   #15
spuggy
Registered User
 
spuggy's Avatar
 
Join Date: Apr 2016
Location: Farnborough
Bike: M900sie
Posts: 171
Quote:
Originally Posted by utopia View Post
I couldn't tell you what voltage I see on the lead for the tell tale light cos my multimeter has gone intermittent after too many winters in a damp shed.
Flasher relay is three-wire.
I'd swap the stock flasher relay back in temporarily and see how bright the tell-tale bulb was then. I suspect this may be the issue, but have no idea why. Diagnosing is key, though...

Are the grounds good? Difficult to see how that'd be an issue with LEDs but not regular bulbs. But always good to check...

I looked at a 1999 Monster current flow diagram; certainly looks to me like the repeater bulb is bridging the left/right indicator circuits (one terminal, F, is connected to the W/G (White/Green?) wire on the left direction indicator circuit, the other, E is connected to the W/BK (White/Black?) wire on the right direction indicator circuit.

Which kind of raises the interesting questions
  • why don't you have the "all my flashers flash at once" issue, and
  • WTF is it getting an ground path from - as the LEDs shouldn't allow one, and the bulb won't work at all without one, the way it's wired...

These question probably both have the same answer. Something is providing a path to ground... I suspect not a great path - hence the dim bulb. And you don't get both sides flashing at once. Although I'd have expected the LEDs to be more tolerant/work on lower voltage/current before the tell-tale bulb, so it's kind of a puzzle...

You would be able to measure voltage between the indicator feed and ground anywhere in the circuit and see what the voltage is when you blag a DMM... If you've got a happy 12.5-14V on the indicator circuit, you're laughing, just a bad connection; trace it back through the harness to the bulb base until you see what's there. I doubt it's that simple though.

It sounds like the flasher may be restricting the voltage and/or current available to the whole indicator circuit - and thus the tell-tale bulb - somehow. The LEDs are probably more tolerant of lower voltage and won't care, but a voltmeter would make it pretty obvious. If that's true, then a higher-wattage bulb likely won't make any difference. Perhaps the relay is in some way incorrect or faulty?

LED flasher relays with a dedicated "pilot" (tell-tale) bulb lead exist, but you'd still need to re-wire the bulb and provide a traditional power/ground arrangement for it. In which case, seems to me you might as well fit a couple of diodes and be done with it.

The "really easy" fix would be a T5 LED bulb that auto-sensed voltage polarity and Just Worked (TM) either way round. No wiring mods. Unfortunately, while these exist for things like the "festoon" bulbs I don't see any offered in a T5 package. Yet.

(fun fact: in some older cars with two-positions for the interior light switch, current flows a different direction through the festoon bulb depending on whether the door is open or you switch the light on manually; use a "dumb" LED festoon and you get to choose which function works, depending on which way round you fit it...)

I think I'd be leaning towards the dual-diode mod, providing a real ground and fitting a T5 LED. The LED should be much brighter than the bulb anyway. Win-win

Should be able to de-pin the connectors somewhere convenient, re-arrange what you need in a little extra mini-harness and re-connect back into the original connector, so no cutting factory wiring.

Quote:
Originally Posted by utopia View Post
Bog standard Citroen C1 (so just half a car, really).
I was expecting out of phase but they flash in phase with the warning light.
lol - Mate of mine had a Peugeot he never managed to get running. Kind of ironic, as he was earning a living as a mechanic at the time... He told me to never trust French electrics, and I believed him...
spuggy is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT. The time now is 01:18 AM.

vBulletin Skins by vBmode.com. Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.