Dashboard Clock goes off? - Page 4 - Toyota Yaris Forums

TeeThree · 03-04-2019, 06:33 AM

Hello Sven, I was wondering if you found a solution to the transistors getting hot and burning out?

sven337 · 04-24-2019, 10:15 AM

I do not know why my repair failed. I would assume the transistors are getting hot too, which makes me wonder if maybe the drive circuit is to blame. I wouldn't be able to fix that.
I'm planning to replace these transistors with P-MOSFETs, to see if they behave any better. They should generate less heat.

I've disassembled the clock again, and clearly the transistors were getting hot, and continued to burn off parts of the boards. A bunch of traces were torn apart when I removed the transistors. I've connected P-MOS the best way possible (had to use a few wires where the traces were gone), will test in a few hours and report back. Hopefully MOSFETs will have fewer thermal problems. Next step if *that* fails is to get 2 or 3 big MOSFETs instead of 6 small ones.

I do not have time to test more, but I went with PCP1302 MOSFETs and the result was a very very hot board after 3 minutes of displaying the clock. How much current does that thing pull?!

sven337 · 05-04-2019, 02:17 PM

Just replacing the transistors doesn't seem to work (because they'll die again soon), so let's figure out what they're for exactly and take a higher level look: what is being implemented here?
The way they're connected, the six of them are basically in parallel and drive the cathode (the horizontal filaments) of the VFD. Their job is to get a steady 300mA through these filaments (full brightness, figure empirically found with a bench power supply).

In fact, you can remove the 6 transistors, and connect a constant-current power supply set to 300mA between the + and - (leftmost and rightmost) pins of the VFD, while it's plugged into the board - it will work normally.

I couldn't fully work out how the transistors were driven (=what their base is connected to), but it all goes back to the mainboard and a, on the power board, a pin that seems to be used to dim the VFD. I don't remember it dimming when turning on the headlamps but maybe that's actually what happens. I'm about to lose that feature anyway.

So, instead of trying to fix this broken cheap power supply with such a terrible efficiency that the transistors dissipate multiple watts (about 500mW each, which is not nothing for SOT-89), let's just replace it with something that works better. My theory for why the repair isn't working is that this design relies on the board helping to dissipate some heat, and when it's as badly burnt as mine it doesn't dissipate as well. Of course as soon as 1 transistor dies the rest will follow due to the parallel design. Crappy power supply design that saved maybe 2 bucks.

My plan is to find a buck module with a 2.5V output that I can use as a constant-current device, or if I can, add a series resistor to limit the current to about 300mA.
Then I'll rip out the 6 transistors and connect the buck module where I need it.

sven337 · 05-05-2019, 09:59 AM

OK, this is fixed! Hopefully this time for good.

I had an adjustable MP1584EN-based buck module ("D-Sun" chinese brand).
These super common modules you can find for a few dollars.
For example:
https://www.ebay.fr/itm/Ultra-Small-...-/322408214136

I used my poor man's lab power supply to test what were the minimal voltage and currents needed to obtain full brightness, and I found out that a constant voltage 1.5V supply would yield a steady 300mA current that would correspond to full brightness. Going to 2.5V would increase current to 450mA without any visible change in brightness, and starting at 4V the filaments would glow red.
(Sorry about the quality of my pictures, I was focused on solving the problem and only took the pics as an afterthought).
https://perso.aquilenet.fr/~sven337/...test_setup.jpg
https://perso.aquilenet.fr/~sven337/...k/halfwatt.jpg
https://perso.aquilenet.fr/~sven337/.../2halfvolt.jpg

Once this was proven working, I cleared some space on the board by removing the transistors and their base and emitter resistors.

https://perso.aquilenet.fr/~sven337/...cleanboard.jpg

Laid out a layer of tape to insulate, soldered the module and taped it in place. This fits with room to spare when reassembling everything.
https://perso.aquilenet.fr/~sven337/...lock/final.jpg

Sorry that final picture is so anti-climactic.. Everything works quite well for now. I hope it lasts, this time, but see no reason why it wouldn't.

TeeThree · 05-06-2019, 02:54 AM

Thanks for the feedback Sven. I'm confident you may have found a lasting solution to the problem. I will look into the DC-DC converter - a nice neat solution.
Would something like the old LM317 work as well? I already have that in my box of spare parts.
If i set the LM317 to 1.5v this should be ok from what you have discovered from measurements.
Thank you for your effort and feedback.

Kaanktm · 05-07-2019, 04:23 AM

Quote:

Originally Posted by sven337

OK, this is fixed! Hopefully this time for good.

I had an adjustable MP1584EN-based buck module ("D-Sun" chinese brand).
These super common modules you can find for a few dollars.
For example:
https://www.ebay.fr/itm/Ultra-Small-...-/322408214136

I used my poor man's lab power supply to test what were the minimal voltage and currents needed to obtain full brightness, and I found out that a constant voltage 1.5V supply would yield a steady 300mA current that would correspond to full brightness. Going to 2.5V would increase current to 450mA without any visible change in brightness, and starting at 4V the filaments would glow red.
(Sorry about the quality of my pictures, I was focused on solving the problem and only took the pics as an afterthought).
https://perso.aquilenet.fr/~sven337/...test_setup.jpg
https://perso.aquilenet.fr/~sven337/...k/halfwatt.jpg
https://perso.aquilenet.fr/~sven337/.../2halfvolt.jpg

Once this was proven working, I cleared some space on the board by removing the transistors and their base and emitter resistors.

https://perso.aquilenet.fr/~sven337/...cleanboard.jpg

Laid out a layer of tape to insulate, soldered the module and taped it in place. This fits with room to spare when reassembling everything.
https://perso.aquilenet.fr/~sven337/...lock/final.jpg

Sorry that final picture is so anti-climactic.. Everything works quite well for now. I hope it lasts, this time, but see no reason why it wouldn't.

Hi, thank you for sharing this. It is really helpful. Hope it lasts lifelong. :)

Could you please inform about the circuit connections on the board. I’m not bad with electronics but apparently not as good as you. Also i couldn’t find about detailed module connection schematics.

So i will appreciate and be very grateful for your reply. Thanks again. This forum rocks. :)

sven337 · 05-07-2019, 03:05 PM

Quote:

Originally Posted by TeeThree

Would something like the old LM317 work as well? I already have that in my box of spare parts.
If i set the LM317 to 1.5v this should be ok from what you have discovered from measurements.

Electrically it will work. Thermally I'm not so sure. LM317 is a linear regulator so in order to deliver 300mA@1.5V it will draw 300mA@12V and dissipate the difference, which is 3.1W.
The TO220 LM317 datasheet mentions a thermal resistance of about 40°C/W, so you're looking at 120°C over ambient temperature for a device with a maximum operating temperature of 125°C. That's why the original board had 6 transistors...
I wouldn't attempt it. A buck regulator has a better efficiency.
The LM317 is supposed to have thermal protection so it will likely shut down without burning too many things, but if your board is like mine any heat is too much heat.

sven337 · 05-07-2019, 03:09 PM

Quote:

Originally Posted by Kaanktm

Could you please inform about the circuit connections on the board.

Are you asking about how to connect the buck module I had to the board? Assuming you have the same, but all buck modules work the same way anyway, you have three leads to connect: Vin, Vout, and GND.
If you look at my very blurry (sorry..) last picture, you can see two red wires and one black wire (at the bottom)*come out of the module. The black wire is GND, and I've taken it at the GND of the capacitor that is next to the blue inductor.
Vin was the + of this capacitor (I think the capacitor/inductor/diode here are an input filter, all I really care about is that there's a constant 12V power supply available there and it was the input of the NPN transistors).
Vout is the leftmost pin of the VFD connector (all 4 left pins are connected together).
I hope this helps you

Kaanktm · 05-07-2019, 03:56 PM

Quote:

Originally Posted by sven337

Are you asking about how to connect the buck module I had to the board? Assuming you have the same, but all buck modules work the same way anyway, you have three leads to connect: Vin, Vout, and GND.
If you look at my very blurry (sorry..) last picture, you can see two red wires and one black wire (at the bottom)*come out of the module. The black wire is GND, and I've taken it at the GND of the capacitor that is next to the blue inductor.
Vin was the + of this capacitor (I think the capacitor/inductor/diode here are an input filter, all I really care about is that there's a constant 12V power supply available there and it was the input of the NPN transistors).
Vout is the leftmost pin of the VFD connector (all 4 left pins are connected together).
I hope this helps you

Actually i was quite able to clear up my mind after asking the question. With your answer it is now crystal clear. Thank you so much again. Hope it would last long.

adsb · 05-12-2019, 01:17 PM

Has this repair held out for you so far sven337 ?

sven337 · 05-12-2019, 02:01 PM

Because I got the question in private multiple times:
- buck modules are not insulated so they have a single ground - no need to connect both ground wires, any of the two will do
- yes, the repair is still holding out, and I've got a good hope that it will hold forever because I think it's "clean"

Do note that you need to remove the transistors (and while you're at it the emitter and base resistors to make room) before connecting the buck module.

adsb · 05-12-2019, 02:43 PM

Thanks sven337,

So the wiring is like this

The 12v feed to the module derived from smd capacitor c535 (470uf/25v) and the dc, aprox 1.5v out (adjust via potentiometer) goes to the left hand pin 1 of the VFD.

Many Thanks again for your help.

sven337 · 05-12-2019, 05:03 PM

Correct

Kaanktm · 05-14-2019, 01:30 AM

Did it yesterday works perfect for now. Thanks again. Btw, be sure that frame of VFD which is the white plastic part, is holding it. There are small guides on it, mine was broken by the one who disassembled it before i guess. Because after i did everything it just worked but when i put it together it stopped working.

Then i just disassembled again and stick the VFD to white plastic frame, now it is perfect.

sven337 · 05-14-2019, 09:22 AM

That is correct, my instructions didn't cover it but the VFD is held in place inside the white plastic part by its top, you have to slide the top first and then press down the part with the pins to get it back in the white plastic part.
But, you don't actually need to remove the VFD from the white plastic part for this fix, what you need to do is separate the white plastic part from the board which my instructions do cover, albeit without drawing attention to the need to ensure that the VFD pins are correctly connected upon reassembly (it's just a matter of pressing hard on the VFD itself).

adsb · 05-18-2019, 03:27 PM

Hi all,
I have the buck module on route from China and hope to complete this modification/repair in the next few weeks.
Can someone explain what this fragile VFD white plastic frame looks like.
I take it that we are dealing with the small VFD and is this "frame" part is the frame which holds the whole clock/warning light pcb to the rest of the instrument assembly ? Hence "many clips" in svens dis assembly guide.
I am looking on this thread for a photo of the frame to identify it but all i can see is a small black frame /surround fitted to the small clock VFD.
Thanks guys.

Great fix btw !

sven337 · 05-20-2019, 07:45 AM

Don't worry about it. Can't miss it when you disassemble, it's the largest plastic part and the only one colored white. And there's nothing to be careful about really, just to know that when you separate it from the black part at its back, it will disconnect the VFD, so don't pull sideways. And after you clip the black part back on (don't push sideways...), you'll need to push on the VFD (it's glass, you can push hard) to make sure its pins are connected.

adsb · 05-27-2019, 05:46 AM

Thanks sven337, I see the white part now, it is pictured in post #48 of this thread, it looks like a support frame to hold the main pcb board.
From the descriptions the white frame holds the VFD display module and the main pcb and when you separate the frame from the main PCB the VFD stays in place on the white frame as it has a slot and a pair of clips to hold the VFD module.
Then its a case of unclipping the VFD from the frame and plugging it into the bare socket for testing/wiring up. Plugging in the VFD provides a load so that the buck module output voltage can be adjusted properly (with a load attached).
I take that the buck output wire is stripped back a few mm at the VFD socket end and the bare end is pushed into pin #1 of the socket receptacle? (The male pin of the VFD holds the wire in place when the VFD is placed back onto the white frame and its all pushed back together).

Did you have any diffculty in removing the SMD components ? Lead free solder is a pain sometimes.

I can see this as a really good fix as the MP1584 based modules run at much lower temperatures than others.

Still waiting for my package from China (buck module)

03-04-2019, 06:33 AM	#55
TeeThree Drives: Yaris T3 5 dr 2006 Join Date: Mar 2019 Location: South Africa Posts: 2	Dashboard Clock VFD circuit Hello Sven, I was wondering if you found a solution to the transistors getting hot and burning out?

04-24-2019, 10:15 AM	#56
sven337 Drives: Yaris Join Date: Feb 2014 Location: France Posts: 33	I do not know why my repair failed. I would assume the transistors are getting hot too, which makes me wonder if maybe the drive circuit is to blame. I wouldn't be able to fix that. I'm planning to replace these transistors with P-MOSFETs, to see if they behave any better. They should generate less heat. I've disassembled the clock again, and clearly the transistors were getting hot, and continued to burn off parts of the boards. A bunch of traces were torn apart when I removed the transistors. I've connected P-MOS the best way possible (had to use a few wires where the traces were gone), will test in a few hours and report back. Hopefully MOSFETs will have fewer thermal problems. Next step if that fails is to get 2 or 3 big MOSFETs instead of 6 small ones. I do not have time to test more, but I went with PCP1302 MOSFETs and the result was a very very hot board after 3 minutes of displaying the clock. How much current does that thing pull?! Last edited by sven337; 05-01-2019 at 08:46 AM.

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
GI: "YarGauge" replacement clock module (Digital Tach, etc.)	CTScott	Items for Sale by private party	756	03-21-2013 08:22 PM
05 Echo changing time on clock	jambo101	DIY / Maintenance / Service	3	11-12-2011 04:34 AM
Dashboard mount for iPhone 4S?	shima	In Car Entertainment + Electronics (audio / video / alarm)	3	10-25-2011 09:22 AM
Dashboard remove ... ALL of it	Toji	Cosmetic Modifications (Exterior/Interior)	7	02-26-2011 06:30 AM
Setting the clock after DST change	ToddR	DIY / Maintenance / Service	10	03-09-2009 11:33 PM

05-04-2019, 02:17 PM	#57
sven337 Drives: Yaris Join Date: Feb 2014 Location: France Posts: 33	Just replacing the transistors doesn't seem to work (because they'll die again soon), so let's figure out what they're for exactly and take a higher level look: what is being implemented here? The way they're connected, the six of them are basically in parallel and drive the cathode (the horizontal filaments) of the VFD. Their job is to get a steady 300mA through these filaments (full brightness, figure empirically found with a bench power supply). In fact, you can remove the 6 transistors, and connect a constant-current power supply set to 300mA between the + and - (leftmost and rightmost) pins of the VFD, while it's plugged into the board - it will work normally. I couldn't fully work out how the transistors were driven (=what their base is connected to), but it all goes back to the mainboard and a, on the power board, a pin that seems to be used to dim the VFD. I don't remember it dimming when turning on the headlamps but maybe that's actually what happens. I'm about to lose that feature anyway. So, instead of trying to fix this broken cheap power supply with such a terrible efficiency that the transistors dissipate multiple watts (about 500mW each, which is not nothing for SOT-89), let's just replace it with something that works better. My theory for why the repair isn't working is that this design relies on the board helping to dissipate some heat, and when it's as badly burnt as mine it doesn't dissipate as well. Of course as soon as 1 transistor dies the rest will follow due to the parallel design. Crappy power supply design that saved maybe 2 bucks. My plan is to find a buck module with a 2.5V output that I can use as a constant-current device, or if I can, add a series resistor to limit the current to about 300mA. Then I'll rip out the 6 transistors and connect the buck module where I need it.

05-05-2019, 09:59 AM	#58
sven337 Drives: Yaris Join Date: Feb 2014 Location: France Posts: 33	OK, this is fixed! Hopefully this time for good. I had an adjustable MP1584EN-based buck module ("D-Sun" chinese brand). These super common modules you can find for a few dollars. For example: https://www.ebay.fr/itm/Ultra-Small-...-/322408214136 I used my poor man's lab power supply to test what were the minimal voltage and currents needed to obtain full brightness, and I found out that a constant voltage 1.5V supply would yield a steady 300mA current that would correspond to full brightness. Going to 2.5V would increase current to 450mA without any visible change in brightness, and starting at 4V the filaments would glow red. (Sorry about the quality of my pictures, I was focused on solving the problem and only took the pics as an afterthought). https://perso.aquilenet.fr/~sven337/...test_setup.jpg https://perso.aquilenet.fr/~sven337/...k/halfwatt.jpg https://perso.aquilenet.fr/~sven337/.../2halfvolt.jpg Once this was proven working, I cleared some space on the board by removing the transistors and their base and emitter resistors. https://perso.aquilenet.fr/~sven337/...cleanboard.jpg Laid out a layer of tape to insulate, soldered the module and taped it in place. This fits with room to spare when reassembling everything. https://perso.aquilenet.fr/~sven337/...lock/final.jpg Sorry that final picture is so anti-climactic.. Everything works quite well for now. I hope it lasts, this time, but see no reason why it wouldn't.

05-06-2019, 02:54 AM	#59
TeeThree Drives: Yaris T3 5 dr 2006 Join Date: Mar 2019 Location: South Africa Posts: 2	Thanks for the feedback Sven. I'm confident you may have found a lasting solution to the problem. I will look into the DC-DC converter - a nice neat solution. Would something like the old LM317 work as well? I already have that in my box of spare parts. If i set the LM317 to 1.5v this should be ok from what you have discovered from measurements. Thank you for your effort and feedback.

05-12-2019, 01:17 PM	#64
adsb Drives: Yaris 1.0 t3 Join Date: May 2019 Location: UK Posts: 5	Has this repair held out for you so far sven337 ?

05-12-2019, 02:01 PM	#65
sven337 Drives: Yaris Join Date: Feb 2014 Location: France Posts: 33	Because I got the question in private multiple times: - buck modules are not insulated so they have a single ground - no need to connect both ground wires, any of the two will do - yes, the repair is still holding out, and I've got a good hope that it will hold forever because I think it's "clean" Do note that you need to remove the transistors (and while you're at it the emitter and base resistors to make room) before connecting the buck module.

05-12-2019, 05:03 PM	#67
sven337 Drives: Yaris Join Date: Feb 2014 Location: France Posts: 33	Correct

05-14-2019, 01:30 AM	#68
Kaanktm Drives: Yaris 1.4 D-4D 2006 Join Date: May 2019 Location: Izmir, Turkey Posts: 5	Did it yesterday works perfect for now. Thanks again. Btw, be sure that frame of VFD which is the white plastic part, is holding it. There are small guides on it, mine was broken by the one who disassembled it before i guess. Because after i did everything it just worked but when i put it together it stopped working. Then i just disassembled again and stick the VFD to white plastic frame, now it is perfect.

05-14-2019, 09:22 AM	#69
sven337 Drives: Yaris Join Date: Feb 2014 Location: France Posts: 33	That is correct, my instructions didn't cover it but the VFD is held in place inside the white plastic part by its top, you have to slide the top first and then press down the part with the pins to get it back in the white plastic part. But, you don't actually need to remove the VFD from the white plastic part for this fix, what you need to do is separate the white plastic part from the board which my instructions do cover, albeit without drawing attention to the need to ensure that the VFD pins are correctly connected upon reassembly (it's just a matter of pressing hard on the VFD itself).

05-18-2019, 03:27 PM	#70
adsb Drives: Yaris 1.0 t3 Join Date: May 2019 Location: UK Posts: 5	Hi all, I have the buck module on route from China and hope to complete this modification/repair in the next few weeks. Can someone explain what this fragile VFD white plastic frame looks like. I take it that we are dealing with the small VFD and is this "frame" part is the frame which holds the whole clock/warning light pcb to the rest of the instrument assembly ? Hence "many clips" in svens dis assembly guide. I am looking on this thread for a photo of the frame to identify it but all i can see is a small black frame /surround fitted to the small clock VFD. Thanks guys. Great fix btw !

05-20-2019, 07:45 AM	#71
sven337 Drives: Yaris Join Date: Feb 2014 Location: France Posts: 33	Don't worry about it. Can't miss it when you disassemble, it's the largest plastic part and the only one colored white. And there's nothing to be careful about really, just to know that when you separate it from the black part at its back, it will disconnect the VFD, so don't pull sideways. And after you clip the black part back on (don't push sideways...), you'll need to push on the VFD (it's glass, you can push hard) to make sure its pins are connected.

05-27-2019, 05:46 AM	#72
adsb Drives: Yaris 1.0 t3 Join Date: May 2019 Location: UK Posts: 5	Thanks sven337, I see the white part now, it is pictured in post #48 of this thread, it looks like a support frame to hold the main pcb board. From the descriptions the white frame holds the VFD display module and the main pcb and when you separate the frame from the main PCB the VFD stays in place on the white frame as it has a slot and a pair of clips to hold the VFD module. Then its a case of unclipping the VFD from the frame and plugging it into the bare socket for testing/wiring up. Plugging in the VFD provides a load so that the buck module output voltage can be adjusted properly (with a load attached). I take that the buck output wire is stripped back a few mm at the VFD socket end and the bare end is pushed into pin #1 of the socket receptacle? (The male pin of the VFD holds the wire in place when the VFD is placed back onto the white frame and its all pushed back together). Did you have any diffculty in removing the SMD components ? Lead free solder is a pain sometimes. I can see this as a really good fix as the MP1584 based modules run at much lower temperatures than others. Still waiting for my package from China (buck module)