DTS Brake Lights - Third brake light voltage, and should brake lights work with ignition off?

[Chris]

The third brake light doesnt work on my 'new' 2007 DTS .

The voltage at the CMHSL is approximately 14.5V when the ignition is on and the brakes are depressed. Is this the normal voltage? I just want to make rule out electrical issues before replacing the CMHSL.

I also noticed during testing that the brake lights don't light up with the ignition off and brake pedal depressed. Is this normal?

Thanks... photos of the new ride to come...

[OldCadTech]

One of the maladies of battery deaths in Cadillacs is the brake light switch mis adjustment keeping the brake lights on. I don't believe 2007 would be any different.  I would check the brake switch adjustment.

[Logan]

Those lamps are known for going bad.

Too cheap new to bother with trying to fix them. A little over $200 for a brand new one. The other major problem with the 2007 and newer DTS lamp is they crack. About 1/2 of them have a crack in the lens.

One of my cars... a 2010 MINI S....no brake lights....no full running lights...unless the car is running.

 

[rockfangd]

Seems to be a issue with GM in general lately. 

They changed the circuit so that they only work with the key on. 

I have seen many newer cars with multiple symptoms

from no brake lights, brake lights always on, no third, only third. You name it.

But as soon as the key is turned off lights go out.

If you are getting at least 12v to one of the wires and ground to the other then likely the lamp itself is your failure

[Cadillac Jim]

Looking at the 2007 FSM, power to the stop/turn lamps and the CHMSL are through the BCM, which has power to those circuits through three separate fuses that is always ON if the corresponding fuse is present and good.

Since the lights are powered through a module, firmware in the module may be used to change the logic from simple switch/relay connections.  The BCM has signals for key in ignition and the key position, so it may not light the CHMSL with the key off and brake depressed if it is so programmed.  But it would seem to me that safety dictates that all three of the stop lights always turn on when the brake is depressed, in case the key is accidentally turned off while the car is being driven, or the key is turned off with the transmission not in Park.

If the BCM is in sleep state, it will wake up if any door is opened, any other module puts anything on the car's network the key is put in the ignition, the headlights or parking lights are ON, or you press any button on the fob.  If it's in the sleep state when you press the brake, the lights may come on or may not but I think that they will because the brake switch will activate the EBCM and such and put signals on the network, which will wake up the BCM.  There may or may not be a noticeable delay when you do this but if there is it will be under half a second.

[OldCadTech]

I agree with @Cadillac Jim , the FSM information would support the theory that the brake lights should have the ability to illuminate with ignition off.

This would be a safety feature in IMO.......   BUT ------\/

Lets assume that the BCM controls the circuit. This means the BCM is controlling the ground circuit which would require a relay. It still needs a signal from someplace, I'm thinking that would be the brake switch. See my previous post...

Since the CHMSL is getting 14v (the engine must be running or a battery charger is connected) it still needs a path to ground. Usually the CHMSL is the first in the circuit to illuminate. SO, if 14v is present supply a redundant ground and see if the lamp illuminates. Is the "BRAKE" relay working/clicking? You can usually "feel" them to verify.

@Cadillac Jim there should be a relay prior to the CHMSL is it ID'd in the manual you have?

[Logan]

This is from the GM electronic service manual for the 2007 DTS. Searching term 'chmsl'.

No brake lights unless key is on.

 

[Cadillac Jim]

The schematics show hard grounds on both the CHMSL and the stop lamps.  Power is supplied directly by the BCM.

The usual procedure for switching high power such as headlights is to use a relay near the bulb(s) and to switch it by grounding the relay from a module.  I think that the reason this isn't done for the stoplights is that the relay would be switched many times per drive cycle.  A solid-state relay would work and that is probably what is done inside the module.  A solid state relay is a MOSFET or IGFET that has protection circuitry and is driven by a circuit that insures rapid, complete switching.  The plug-in relays are $3 throw-away parts; solid state relays are more expensive but rarely fail.  I just did a check and solid state relays that can handle most automotive applications are out there for under $10 retail, which means that the car of your future will have them instead of the throwaway plug-in mechanical relays eventually.

[Logan]

GM service electronic service manual

[Logan]

GM electronic service manual

[Logan]

GM electronic service manual

[OldCadTech]

Thanks @Logan & @Cadillac Jim very good information.

The ability of the BCM/Module(s) to handle that kind of heat load is baffling to say the least. There is a higher fail rate in SSRs and they do generate a lot of heat in use and heat is the worst enemy of electronics. Where did they put the heatsink? Technology constantly evolves!

I don't agree with GM's decision to rely on only the ignition key position especially in light of the ignition cylinder fiasco and brake lights are an important "feature" but they don't care what I think anyway.

We did get a little off track and I think the OP is out working on the vehicle  

[Cadillac Jim]

The electronic relay doesn't generate any more heat than the mechanical relay.  The key is fast switching and low ON resistance.

The description provided by Logan tells us what the BCM is programmed to do.

Chris:  If the CHMSL doesn't light up when there is voltage to it, it's likely a bad ground, ground connector, or connector to the CHMSL ground circuit.  The connector is A10, circuit 1320, metric wire size 0.5, color L-BU.  Ground is directly to G402, circuit 850, metric wire size 0.5, color BK.

G402 comes out of the wiring harness to the right taillight and is located "(CKT 850) In the lower left of the rear compartment, behind the left wheelhouse panel."  If there is voltage anywhere on this ground, connection from there to the body is bad.

And, yes, it is normal for the brake lights not to have voltage when the key is OFF.  Logan looked that up for us.

[OldCadTech]

Mosfets are bi-directional so they require two to prevent back feed and generate more heat in the closed or ON state, since brake application is temporary in nature maybe the heat factor doesn't apply, except MAYBE the loooong stop light, that is my understanding of their operation....

[Chris]

8 hours ago, Cadillac Jim said:

Chris:  If the CHMSL doesn't light up when there is voltage to it, it's likely a bad ground, ground connector, or connector to the CHMSL ground circuit.  The connector is A10, circuit 1320, metric wire size 0.5, color L-BU.  Ground is directly to G402, circuit 850, metric wire size 0.5, color BK.

G402 comes out of the wiring harness to the right taillight and is located "(CKT 850) In the lower left of the rear compartment, behind the left wheelhouse panel."  If there is voltage anywhere on this ground, connection from there to the body is bad.

Thanks for all the replies!

I'm glad that the brake lights aren't supposed to operate with the ignition in the off position!

Jim - Just to clarify, I disconnected the stop light and measured voltage across the connector. With the brake pedal released, there is 0V, with the pedal depressed, there is 14.5V. Based on that, I don't believe there is any problem with the ground, and that there is an internal issue with the CMHSL.

[Cadillac Jim]

Getting voltage across the terminals of the CHMSL connector pretty much nails it.  However, if the ground has a high resistance, it can still pass that test with a voltmeter.  Use a test light and if it lights up, then you know for sure.

Another thing to check is the connector.  If you can put 12 Volts across the CHMSL itself and get it to light, the connector isn't making good contact.

Sorry to be picky but I see that the CHMSL for the 2007 DTS in L trim is $217 and change.

[Cadillac Jim]

OldCadTech - high-current MOSFETs that are intended for switching where there may be overvoltage or back-current is to incorporate diode protection into the MOSFET right at the die, and zener overvoltage protection.  They will need a heat sink if there is any amount of that.  In an automative lighting application, I would not think that there would be much of that.  The package allows for a heat sink but unless there is a motor or a really long wire, you won't see much heat from the protection circuitry.

The LED array in the CHMSL is a bunch of resistor-LED circuits all in parallel.

[rockfangd]

10 hours ago, Cadillac Jim said:

Getting voltage across the terminals of the CHMSL connector pretty much nails it.  However, if the ground has a high resistance, it can still pass that test with a voltmeter.  Use a test light and if it lights up, then you know for sure.

Another thing to check is the connector.  If you can put 12 Volts across the CHMSL itself and get it to light, the connector isn't making good contact.

Sorry to be picky but I see that the CHMSL for the 2007 DTS in L trim is $217 and change.

Preferably not a led test light but rather a regular bulb type. Led uses so little power you may not recognize any resistance in the circuit if there is any. 

Where as a standard test light will be likely dim or not at all

[Cadillac Jim]

Yeah, what rockfangd says.

[OldCadTech]

Cadillac Jim, I think the key here is the LEDs, THAT's how they can get by with no or very low heat. Didn't quite click until the last two posts by you and Rockfangd. If the brake lights were 1157s or even 194's THAT would be a different picture. That brought it into focus thanks. 

Side note: No diode is needed, the mosfets will become unidirectional if two are used and the opposite ends are connected together. I can't think of the name of the terminals that have to be connected, but for this conversation it doesn't really matter anyway. Thanks!

[Chris]

I have a 12V DC external power source and hooked it up to the CMHSL. Nothing.

I've ordered a new unit and surprisingly, prices are much cheaper up in Canada for once ($339cad is the standard price at stores or $295cad including shipping at dealership online stores) whereas I was quoted over $300usd by gmotors.com and on ebay.

Thanks again for the replies and interesting to read some of the conversation!

[Cadillac Jim]

Let us know when you are done and the CHMSL lights up again.

Logan, do you sell refurbished CHMSL assemblies?

[Logan]

Not for the 2007 and newer DTS..for a couple of reasons.

They are fairly cheap...little over $200. They used to be about $145 shipped to your door.

And...most will have a crack in the lens. Very common problem. The crack cannot be fixed.

The part # is 25844853

 

 

[Chris]

The new brake light arrived today. I swapped them (wow that was fast & easy) and the new one works great.

The date sticker on the faulty one indicated a January 2007 build date so I assume it was the original unit. Out of interest, the new one had a build date of November 2015.

Thanks again for everyone's feedback - definitely made me more comfortable spending $300cad on an LED strip

[rockfangd]

Glad you got it fixed