I've done a ton of searching and haven't come up with much. I have an 05 ES330 with stock HID's and I'm looking for more light output. I replaced the bulbs with Osram Nightbreaker laser and they are better, but my daughter's 04 camry with a halogen projector and cheap amazon LED's is brighter and has better visibility.

I can keep going down the path of trying to make my HID's better, next step is to replace the ballast and see if that helps, but I don't want to keep putting money into it if it's a losing battle.

I can try and wire up a 55w ballast, but worried about the heat.

I can convert to LED, most likely the Morimoto 2 stroke 3.0 D2 and bypass the factory ballast.

I don't know if I can do a full retrofit and change to a morimoto D2S 5.0

Thoughts?

 

55 watt HIDs are terrible don't do it the bulb gets far too hot and the ballasts are highly failure prone. Check out this channel he tests a bunch of LED bulbs and also has a spreadsheet with them all compared very useful
https://www.youtube.com/@CarLightReviews

If it was me I would try and go LED, problem is if your car has DRLs this complicates things. I have a 2006 Highlander I solved this buy having a friend with techstream disable DRL function, instead the car always has the headlights on when the car is running. Not ideal but does the job.

 

Hello,

I wonder if there is a way to make the conversion without loosing DRL function, what caused you to disable yours in a Highlander?

 

DRL function as you know runs the high beam bulbs in series for half brightness this doesn't play nice with LED bulbs. Morimoto makes a DRL module but from what I gather it still needs an LED that runs at 6 volts. In retrospect I should have simply got the Morimoto kit for high beam, 20/20 hindsight.

When I get around to it I'm going with LEDs on my ES300 I'm not happy with the HID light output especially after running LEDs in my other cars.

 

Thank you, that's great information! My high beams are separate from the low and I'll keep those halogen so hopefully switching to LED's won't cause any issues. I'll follow up with my experience after. I'm going with the Morimoto 2 stroke because they are 5500k. I wish I could find 5000k or even 4500k, but no luck.

 

Please keep us updated. 5500K is nice for sure close to ideal. 4500K may seem better but once you get used to a higher Kelvin you won't go back. It's a common misconception that the human eye is more sensitive to warmer light at night, our eyes have evolved to be sensitive to moonlight which is quite blue.

 

Can you link the module you mentioned?

Funnily enough, I couldn't find any Morimoto modules other than the one for a newer Tundra. It appears like the DRLs in Toyota vehicles was a bit of an afterthought until they moved on to PWM from about 2007, and even then they made a fault of using a slow MOSFET to dim the lights, so the LED bulbs are still affected by it, causing some flickering, and a need for a "CANBus decoder" as they call it to smooth it all out.

5th-gen Camry ran a Series Resistor along with connecting the bulbs in series to dim them even further, something that a 3rd-gen ES doesn't have, I wonder if that could be the cause for your lights not working that well in series, as LEDs shouldn't be affected by that alone as much..

If that's not the case, I don't see any other way but to eliminate both DRL No.3 and No.4 relays and place a good quality LED driver like this one instead. By monitoring the power on DRL fuse, you can know when you need PWM signal to dim the bulbs, and when you need the full power for High Beams. With that, there is no constant 12V power that comes into that separate fuse box, so you will have to drag the power wire in there and take the whole fuse box apart, removing all but DRL No.2 relay and modifying the wiring, which likely won't be reversible, and it will never work with regular bulbs again as those demand a lot more power than LED driver can handle.

 

This one
https://www.morimotohid.com/morimoto...ses?quantity=1

I figured out a much easier way to make my own DRL controller. I monitor the brake light in the cluster which is only on when the engine parking brake is set, or the key is in the ON position but not running. With some simple circuitry what happens is if you start the car with the parking brake set, DRL relay (my relay) not active. Release the parking brake relay latches and you have DRLs until the car is shot off. Or if you start the car without parking brake set DRL relay (again, my relay) turns on and latches, you have DRL power. Basically a simple way to know when the engine is running, not as easy as you'd think. Another way is look at the charge warning signal coming from the alternator.

Downside is this means you're doing hacking of the factory headlight bulb wiring like you're talking about, or you have to add dedicated DRL bulbs somewhere. Bunch of ways to do this, open up the headlight housing and add an LED strip, add a bulb and socket to one of the corner lights etc.

I have some pictures of all of this somewhere I'll post if I find them.

A note on PWM and LEDs, it kills LEDs. I did some testing on one of my cars (which has custom tail lights) I ran a PWM module on one lamp (dropped the voltage down to about 11) none on the other. Did this for over a year the LEDs on the PWM circuit kept burning out the other ones did not. I've noticed the same issue using LEDs in a cluster that uses a PWM dimmer circuit which is common on most any car after about 1988. LEDs keep burning out I think this is because the LEDs have their own driver circuit which doesn't like the PWM input. Solution is to add a schottky diode and capacitor after the PWM circuit.

 

Yep, those are the CANBus Decoders, I have no idea why so many people call it that though..

The thing is that this circuitry is already in your car, either in an Integration Relay, or in a Body ECU, depending on the year that we are talking about. It monitors the Input from the Alternator, Light Stalk and a Parking Brake switch, and uses DRL No.2 relay to turn DRLs on.

DRL No.3 and No.4 are energized from the same source, such that they always engage at the same time. When they are off, both High Beams are powered from one and the same fuse, and are connected in Series to Ground through a 5th terminal on the DRL No.4 relay. When the HEAD relay is On, it powers the DRL fuse and turns both No.3 and No.4 relays On, causing the High Beams to run in Parallel, Integration Relay also turns Off the DRL No.2 relay, from now on it acts as a High Beam relay.

The idea is to use the DRL Fuse as a signal, when the Fuse is powered, DRLs must be off. You can eliminate DRL No.4 completely, and run an LED Driver in place of the DRL No.3 relay, that way the High Beams will never be wired in Series again, instead, when there is no power on DRL fuse, feed a PWM signal to the bulbs, when the DRL fuse is powered, run full power to the bulbs when High Beams are On.

The problem with this setup, as mentioned above, is that the dedicated fuse boxes in both Highlander and ES300 don't have any constant 12V supply, there is no way to power the LED driver, so a separate wire will need to be added for it to function. Another problem is that 2 out of 3 relays will have to be cut away, and a new watertight enclosure will have to be designed to install the driver and hide all the wiring.

That's one of the moral hurdles that you will have to face when doing something like that..

Hence the dedicated LED Driver instead of raw PWM signal from a transistor. There is really no other way to dim an LED as far as I am aware, that's why you see more or less all of them flicker on camera, it all depends on how you implement it. Not all the bulbs will like it either, so it is important to select one that will function correctly for the application, I am pretty sure it is something you can learn from a manufacturer.

Hope this helps and best of luck!

 

You can, linear voltage regulator or if you like even more heat use resistors. Actually if you put resistors on each LED it works well you see this on things like instrument clusters. LEDs simply don't like AC current they are very low current diodes and in a tiny physical package my theory is that's why they fail long before the rated life. I see LED failures on cars only a few years old, the Lexus DRL swoosh is notorious for going bad some of the LEDs stop working or sometimes the entire thing fails. I saw a 3-4 year old RX the other day one of the brake lights was flickering, LEDs failing.
Most if not all LEDs now have some sort of driver circuit even things like T10's.

BTW great explanation on how the convoluted DRL relay system works. The custom circuit I'm talking about, using it on cars that have no DRL capability. It doesn't help me at all to mod it on my Lexus because as you say you end up with full power to the high beams blinding other drivers. Or if I go for a more extensive mod again you're right creates other problems.

That Morimoto kit I linked to does work in cars that use 6 volt PWM a member on this forum bought them and verified this. Will it work on my Lexus? Tempted to buy and find out.

 

Coming to think of it, this driver has an Analog dimming function, reducing current supply to the light, the precise amount of current is set via a series of resistors that can be calculated using the datasheet. It is a lot more efficient than running a resistor to the light, and it doesn't use PWM, win-win. Now the question is how well the onboard circuitry of the LED light will enjoy it, unfortunately there is only one way to find out.

It's been a while since I've seen a blown LED light, the most recent one being the household one out of my desk lamp, though it weren't the LEDs themselves that failed. Gotta wonder if there is any way to get a bare bones light with no internal circuitry and have all the flexibility.

The Integration Relay should be the same in cars that never came with DRLs, gotta wonder if it is as simple as adding a couple of missing wires with respective relays to get it working.. I am yet to come across a car that was not equipped with DRLs to take a deeper look though, who knows, maybe all the wiring is already in place, like it is with TEMS system for example.

It's likely the question of finding the right driver to function with bulbs you got, I am fairly certain it is possible to make a custom PCB that replaces the relays for a Lot cheaper than what they charge for the harness, that is as long as you are ready to get your hands dirty and sacrifice the factory wiring for the LED setup..

I wonder what year that was though.. Below is a quote from their website, it appears that those are just a beefy capacitor in a metal box, they can only help with newer cars that use PWM to dim the bulbs instead of running them in Series, cars from about 2007 and beyond.

Hope this helps and best of luck!

 

That's not a true analog chip, same thing that is being used here.
Switching frequency that drives a MOSFET that is the part they consider analog which technically it is. I think you're right that Morimoto box is a fancy capacitor which is fine you can't make one yourself that is packaged that nicely. Integration relays for 80's and 90's Toyota's didn't have DRL function that was done in separate box. U.S. spec cars were missing the box and associated relays.

You think dealing with LED headlight bulbs is a pain try going all LED on the entire car.

edit - found a picture this is a modded DRL module from a '90 Camry all it did was turn all the lights on any time the engine was running. Pretty useless so I removed all electronic components wiped out most of the traces then added my own circuit. The market lamp you see is a modded to have LEDs

 

It's a little more complicated, there are multitudes of ways to design a controller like this, here is a very useful document from Texas Instruments that describes different ways of dimming the LED, their advantages and drawbacks.

For whatever reason, I just ended up sitting until about 4AM trying to work out what are the ways to dim LEDs, which ways are better, and how they are implemented. Try as I might, I wasn't able to find any information on PWM actually killing the LED diode itself, with it being quite the opposite if anything as there is a lot less load on the light the lower duty cycle you select. Here is a very good research article on the subject, which concluded that it's actually a preferred dimming method over Constant Current Reduction in terms of light clarity and output temperatures, which is not something I would expect.

My guess is that the issues you ran into when trying to dim your lights were caused by the Internal Circuitry used in the light, as it doesn't appear that they considered a possibility that someone in their right mind would go out of their way to dim their fresh new LED bulb. Not a lot of ready to use LED lights will play nicely with attempts at dimming them, as it requires a lot more of a complicated setup that's usually not justified. I am pretty sure it is also the cause of your lights not faring all that well to being connected in Series, as just the LED light itself is not all that affected by it, in fact that's usually how they wire them to a driver.

As for OEM implementation failing prematurely, I am not sure, it is a subject to further investigation, my best guess would be that they are trying to push the smallest lights they have much beyond their safe operating limit in hopes that it will look good until the warranty runs out. It appears that it's the temperature that gets them as the quickest way to fry an LED..

One of the simpler ways to get an analog output out of a controller is by opening the MOSFET gate only partially, similar to the way Toyota used to regulate speed of the blower motor while they were transitioning from load resistors. With that, I can definitely say that no integrated driver will like the current limit, as all the filtering and decoupling components are calculated for a very specific current, so there is a chance it either won't work, or the light will simply fry itself in a near future, which is likely what happened to your taillights when you ran an experiment.

Coming back to one of my questions, whether they sell just the light itself, with no driver circuit to have complete flexibility, or at least the lights that provide the ways of dimming them, since it is always preferred to have a driver located as close to the light as possible. Considering that this is not the first time this issue came up, some companies may be very much interested in getting it to work. For now though, the only way to pull something like this off is by tearing into the light and bypassing the existing driver to implement one of your owns

Well, that's precisely why I never even tried to dig into it, at least until yesterday that is..

Hmm, that's a lot more lights than I am used to seeing in a corner lamp.. One day I will get my hands on corner lights off of Windom though, it appears that they have done something similar to your mod, but straight from the factory.

Hope this helps and best of luck!

 

In theory as you say less current should mean longer life. Persistence of vision means we see the light having about the same brightness as well. But my year+ long experiment proves that PWM kills LED bulbs but the test is not only testing the LED emitter itself as you know. BTW, in my city they converted to LED streetlights a few years ago and you guessed it, they keep burning out. Sometimes they develop a seizure inducing flicker.
I am certain aftermarket LED bulbs die because of heat they are pushed hard and get very hot. Headlight bulbs can run at 250 degrees Fahrenheit it's amazing they survive as long as they do. Can see the same issue on OEM lighting. Plus people may not realize this, the LEDs themselves all come out of the same factories in China. All of them. Philips is making less and less components (their overall revenue has dropped drastically over the years) so whatever advantage the vaunted Philips lights may have enjoyed is going away.
Toyota did not use a MOSFET in a linear way, not that I know of. When you do that means all of the heat is sunk across the MOSFET for a blower motor that can be 15+ amps a MOSFET will not survive that unless on a massive heatsink. Pretty sure the circuit is PWM.

I'm using a MOSFET switchback circuit for my front signal lights, they fade in after the turn signal is off. Interesting thing here is the T10's do fade in but do so in a digital stepped way, the built in circuit is trying to output a constant current.
I've been experimenting with these

Want to go deep red for my tail lights, but again the problem is voltage. These run best efficiency at exactly 2 volts so I need a regulator. I tried running them in series didn't go so well they burned out quickly because each LED does not have the identical voltage drop, so one LED gets more current it gets hot voltage drop changes. Run away effect until failure. Took about 10 minutes that's it.

One LED for signal, two for DRL. One is turned off when headlights are turned on to prevent blinding people at night. There is a switchback circuit that kills the white LEDs when turn signal is active. At night, one white LED is always on this is so the corner of the car can always be located at night this is a safety thing.

In fact since that picture I have improved the lamp further, the side firing LED is now a dual mode amber LED so now the corner lamp has both amber and white running light. All these different modes/conditions gets complicated wiring wise.

 

The thing is that PWM doesn't reduce current, it stays the same, it only changes the amount of time the bulb is on, hence why it is a lot more effective at preserving the color of the White, as opposed to current control that reduces the temperature along with brightness.

I think that's one of the major causes, trying to adjust the brightness of the light that was already adjusted by the internal driver. Instead of feeding the PWM signal to a dedicated port, it is basically switching the entire power to the circuit on and off, and that plays a major role in the effectiveness and longevity of the setup. LED itself is not that much affected by the PWM, it's the driver that has no idea what to do with it that suffers the most.

My DeWalt LED light just kicked the bucket recently, I took it apart and to my surprise, the light itself is fully intact, it's the driver that got fried for one reason or another, maybe just age, it's been a while since I got it, and I it's not like I even tried to preserve it..

Not sure why would they try reducing their brightness though, there may be another reason..

That's a big part of the problem as well.

Isn't everyone falling in the pits of the abyss by now, that's pretty sad to think about..

It does seem weird, but this is what I presume they mean by adjusting the voltage at the Base of the transistor, not sending PWM signal, I don't have a 2nd-gen ES to hook up an oscilloscope. The power transistor in that car is huge though, it sits in the same location as load resistors, being cooled by the blower motor. This was only used in that generation I think, as in 3rd-gen they thankfully moved to a completely separate driver that does utilize PWM.



That's actually one of the main reasons to use a driver with a feedback loop, which can adjust the current on the fly if it goes out of limits, and a good reason to look for some higher quality LED lights, maybe that could make a difference.

Sounds like it's about time to use a microcontroller like an ATtiny chip to conduct it all..

Hope this helps and best of luck!