Mitsubishi 3000GT & Dodge Stealth Forum banner

181 - 197 of 197 Posts

·
Registered
Joined
·
20 Posts
Wow, you clearly know a lot about electronics! I recently tried to fix one but kept getting the red lines and overheating R38. Is it true that it's possible it is beyond repair if the VFD goes bad? Information on repairing these is very scarce and sources often have differing information. Thanks!

Sent from my SM-G960U1 using Tapatalk
If the VFD is actually bad, then it is beyond practical repair. The VFD is sealed. The red lines are the heater for the VFD. The heater usually runs on a lower voltage than the VFD itself. So if the VFD has red lines, the heater filament is drawing more current than it is supposed to. That means the heater filaments themselves or bad or the voltage driving the filaments is too high. One of those circuits on the climate control is probably in control of that.

In my experience with a few of these, most problems come from C14 and C15's electrolyte eating traces. If you don't clean the corrosion and repair the traces, you will never have a working display.
 

·
Registered
Joined
·
1,156 Posts
unless you use the wrong transistor as a replacement.
I am guessing the replacement transistors I used are operating differently. My first attempt worked fine for a couple years using the NTE378 until about half of the VFD dropped. My latest attempt with TIP42C heats up the resistor until it is too hot to touch. The VFD is completely lit. Possibly too bright. Maybe the TL494 is being over driven by the replacement TR2? Thanks for your advice, we definitely need people who can shed some light on this issue. Everything I have seen about VFD's indicates that they rarely fail, even with age. I ordered replacements that claim to be the same as the original Toshiba TR2.

Would love to know if anyone knew the correct check-point voltages on the board, Vss, F, gd+ etc... Would love to know how to check the VFD for something other than just resistance.
 

·
I feel the need for speed
Joined
·
740 Posts
If the VFD is actually bad, then it is beyond practical repair. The VFD is sealed. The red lines are the heater for the VFD. The heater usually runs on a lower voltage than the VFD itself. So if the VFD has red lines, the heater filament is drawing more current than it is supposed to. That means the heater filaments themselves or bad or the voltage driving the filaments is too high. One of those circuits on the climate control is probably in control of that.

In my experience with a few of these, most problems come from C14 and C15's electrolyte eating traces. If you don't clean the corrosion and repair the traces, you will never have a working display.
That's great info, I wish I knew more about the theory to be able to check it out. I'm an electrician by trade and I've tinkered with electronics often, but I'm no engineer.

My board had very little, if any, corrosion so I don't think that's it. I'm guessing it's a component but I lack the knowledge of how to diagnose it. I had 3 units I was working with, 2 I bought condition unknown, the third came out of my car which was working intermittently until one day I smelled smoke and that was that. I rebuilt it but it has the red lines. I can't remember what the second one was doing right now. The third looked terrible internally but still functioned fine, after a thorough cleaning and capacitor swap it's been working in the car with no trouble.

I want to learn how to fix these so I can pass the knowledge along to the community. Is there any way that you could walk me through figuring it out? I understand it may be time consuming and I'm okay if you can't take on the project, you've already been helpful as is!

Sent from my SM-G960U1 using Tapatalk
 

·
Registered
Joined
·
20 Posts
I am guessing the replacement transistors I used are operating differently. My first attempt worked fine for a couple years using the NTE378 until about half of the VFD dropped. My latest attempt with TIP42C heats up the resistor until it is too hot to touch. The VFD is completely lit. Possibly too bright. Maybe the TL494 is being over driven by the replacement TR2? Thanks for your advice, we definitely need people who can shed some light on this issue. Everything I have seen about VFD's indicates that they rarely fail, even with age. I ordered replacements that claim to be the same as the original Toshiba TR2.

Would love to know if anyone knew the correct check-point voltages on the board, Vss, F, gd+ etc... Would love to know how to check the VFD for something other than just resistance.
I haven't actually had a dead TR2 in the couple I've looked at. In my quick look of what was available, I have 2SA2222SG on my list, but I've never had to use it. The 2SA2222SG can handle more current, but less voltage than your TIP42C. The Vce of the TIP42C is also pretty high. The gain bandwidth product and DC current gain are also much higher on the 2SA2222SG. So take that substitute with a grain of salt as I have NOT used it in one, but it was the best match I could find available for the OEM one at the time.

That's great info, I wish I knew more about the theory to be able to check it out. I'm an electrician by trade and I've tinkered with electronics often, but I'm no engineer.

My board had very little, if any, corrosion so I don't think that's it. I'm guessing it's a component but I lack the knowledge of how to diagnose it. I had 3 units I was working with, 2 I bought condition unknown, the third came out of my car which was working intermittently until one day I smelled smoke and that was that. I rebuilt it but it has the red lines. I can't remember what the second one was doing right now. The third looked terrible internally but still functioned fine, after a thorough cleaning and capacitor swap it's been working in the car with no trouble.

I want to learn how to fix these so I can pass the knowledge along to the community. Is there any way that you could walk me through figuring it out? I understand it may be time consuming and I'm okay if you can't take on the project, you've already been helpful as is!

Sent from my SM-G960U1 using Tapatalk
I don't have the pinout of the VFD and never bothered trying to bug it out. Some of the youtube videos linked earlier in the post show the basic components and how to potentially find the right pins.

In terms of fixing them, I fixed my own which was kinda like yours. Nothing obvious wrong with caps, but I did a full refurb anyway. I have a previous post with my list of parts (including ECU and ECS). Updated Digital Climate Control, ECS, and ECU Component List Mine worked after that and has been working ever since. I got 2 other displays and refurbed them too the same way, but I have not tested them out. I did not feel like ripping my dash back apart just to test them. The other 2 I bought had definite capacitor issues where the electrolyte ate traces. So I had to clean that off and repair the traces. I've found it is more time efficient to just replace everything on that list than hunt individual problems. The circuit really isn't that large and on top of that you extend the longevity of the entire circuit with newer and better parts. If you really want to learn about the power supplies, you can look at Perry Babin's tutorials at Basic Switching Power Supply Design Tutorial

The first steps to really digging into this would be to trace the circuit starting at R38. Specifics of a circuit can be explained (if I can) as you get stuck on them.
 

·
Registered
Joined
·
1,156 Posts
Just an update. I bought a "direct" new replacement transistor for TR2 Toshiba B1019. It has the same number printed on the front. Found it in England on EBAY. After I pulled out the Tip42, I decided to check all of my transistors OFF the board. My multi-meter does not have a transistor checker, so I can only check resistance.

With the top side of the transistor (with the model number) facing you and the leads pointing toward you, the leads from left to right are Base(B) Collector(C) and Emitter(E). On a PNP transistor, it should conduct when you put the negative lead on the base and the positive lead on the collector or emitter. It should not conduct n the opposite direct. You can use a diode checker to hear the beep or measure the resistance value to confirm conduction. If you reverse the leads it should not conduct and the meter will read Over Load ( 0L).

Here are the resistance results:
Original B1019-Neg-Pos B-C 2.84 Mohm, Neg-Pos B-E 2.86 Mohm, Pos-Neg B-C 0L, Pos-Neg B-E 0L.
Ebay B1019-----Neg-Pos B-C 2.48 Mohm, Neg-Pos B-E 5.31 Kohm, Pos-Neg B-C 0L, Pos-Neg B-E 5.3 Kohm
Tip42C----------Neg-Pos B-C 3.72 Mohm, Neg-Pos B-E 3.76 Mohm, Pos-Neg B-C 0L, Pos-Neg B-E 0L.

Observations:
-Ebay has a short between the B-E both ways and the resistance is lower than all of the others.
-The Original is testing normal.

Conclusion:
Putting the original back on the board.

Going to look at TR3 next it is a Voltage regulator transistor.
 

·
Registered
Joined
·
20 Posts
As a general rule of thumb, you should always check BJT's using the diode checker and post the voltage drops of that as opposed to resistance. On a MOSFET, you can use resistance check. You have to be careful checking a MOSFET though because you can introduce enough charge if you initially check gate to source to actually short the drain to source. That might be beyond what you are trying to do, but just be aware of "false" readings when checking a MOSFET out of circuit unless you remove the gate charge first.

Good luck with the rest of your investigation.
 

·
Registered
Joined
·
1,156 Posts
So, I tried using NTE378 and TIP42C replacement for PNP transistor 2SB1019 (Big green one) TR2. Did a bit of research and the equivalents have differences. I don't really know but it makes sense that the transistor could be the problem since R38 is heating up.

Model Hfe Frequency
2SB1019 Y (120-240) 10
TIP42C (15-76) 3
NTE378 (40) 40
I bought a component tester 16.00 shipped on ebay. It measures transistors gain (Hfe).
Model Hfe
TIP42C 159 (Previously installed on hot board) The gain is out of spec. for this transistor, but within spec for the original.
TIP42C 160 (New) Vbe 612mV Ic=6.2mA
2SB1019Y 47(New Ebay Replica) Vbe=1.14mV Ic=5.4mA
2SB1019Y 45(Previously installed on hot board, Ebay Replica) Vbe=1.13mV Ic=5.4mA

TR3
2SA1015 Hfe=429, Vbe=669mv Ic=6.1mA (New) Spec. says Hfe (70-400) so again it is reading on the high end.

No conclusions here except the transistors tested pretty well even the one that I thought tested bad using only resistance. The ebay replica looks like the Tip42 and measures within it's spec but out of spec for the transistor it is supposed to be. All of the variations of transistors seem to heat up too hot on my bad board.
 

·
Registered
Joined
·
20 Posts
Your ebay replica's look like they have a base to emitter short. I would expect Vbe to be in the 400-700mV range. Use a multimeter on the diode function to check base to emitter (black on base, red on emitter) If it still reads super low, they are bad.

If the new transistor has at least the same amount of gain as the original, it does not matter if it has more or not. It was designed for a certain gain, so it should only have problems if you have less gain than OEM. Also. if your replacement transistor has a higher Vce(sat), it is lossier and therefore will run hotter. The TIP42C has a Vce(sat) max of 1.5V. The 2SB1019Y has a Vce(max) of 0.4V. The 2SB1019Y will run cooler than the TIP42C. Does your transistor tester give you Vce?
 

·
Registered
Joined
·
1,156 Posts
I checked the ebay replica and it tested bad using a diode checker on a multi-meter. Thank You. The transistor tester does not measure Vce.
 

·
Registered
Joined
·
20 Posts
You absolutely must repair any damaged traces to have any sort of reliable repair. If you have the right tools, I still like removing the display first. You have much easier access to removing the capacitors and you can ensure there is no debris on the display side after repair. It makes removing all the other parts a lot easier too. Using braid and flux can get pretty time consuming if you only have access to one side of the board.

The video certainly does a good job of showing some of the repair process though.
 

·
Registered
Joined
·
1,156 Posts
I bought a component tester 16.00 shipped on ebay. It measures transistors gain (Hfe).
Model Hfe
TIP42C 159 (Previously installed on hot board) The gain is out of spec. for this transistor, but within spec for the original.
TIP42C 160 (New) Vbe 612mV Ic=6.2mA
2SB1019Y 47(New Ebay Replica) Vbe=1.14mV Ic=5.4mA
2SB1019Y 45(Previously installed on hot board, Ebay Replica) Vbe=1.13mV Ic=5.4mA
2SB1019Y 183 (OEM supplied not running hot) Vbe=539mV Ic=6.3mA *** Added a new measurement
NTE378 168 Vbe=555mv Ic=6.3
TR3
2SA1015 Hfe=429, Vbe=669mv Ic=6.1mA (New) Spec. says Hfe (70-400) so again it is reading on the high end.

No conclusions here except the transistors tested pretty well even the one that I thought tested bad using only resistance. The ebay replica looks like the Tip42 and measures within it's spec but out of spec for the transistor it is supposed to be. All of the variations of transistors seem to heat up too hot on my bad board.
Added new measurements above. Looks like the tip42C and NTE378 are pretty decent match to the OEM. Also TR2 really is not an issue even on my hot board.
 

·
Registered
Joined
·
1,156 Posts
Fixed both of my displays! It boiled down to checking every trace and connection. The video helped immensely because shows where traces tend to break. They break near the test points and on the edge of the component solder pad and the trace. Seems like the capacitors leak and they attack the un-protected surfaces like the test points and pad edges. It had NOTHING to do with TR2 or the VFD display. One of my displays was only partially working. The other unit had a R38 that burned off the top layer of the board.
 

·
I feel the need for speed
Joined
·
740 Posts
Fixed both of my displays! It boiled down to checking every trace and connection. The video helped immensely because shows where traces tend to break. They break near the test points and on the edge of the component solder pad and the trace. Seems like the capacitors leak and they attack the un-protected surfaces like the test points and pad edges. It had NOTHING to do with TR2 or the VFD display. One of my displays was only partially working. The other unit had a R38 that burned off the top layer of the board.
Were they non functional completely? Curious if they had the famous red lines after everything was replaced most specifically.

Sent from my SM-G960U1 using Tapatalk
 

·
Registered
Joined
·
1,156 Posts
Were they non functional completely? Curious if they had the famous red lines after everything was replaced most specifically.

Sent from my SM-G960U1 using Tapatalk
One started out completely non functional. I replaced components (Main resistor large capactitors and diodes) and it started working but the resistor and transistor would get very hot very quickly. Repairing the the traces helped stop it from heating up. The other one was partially functional. It did have the red lines and then only part of the display would light up. I replaced the diodes on that one and the capacitors, but had no improvement until, repairing the traces.
 

·
Registered
Joined
·
1,156 Posts
Just took the car out for a ride. Display worked great and is fully installed. Dimmer function was even working. For some reason I missed seeing that the light bulb illuminating the left buttons is burned out. An, old Indian practice is to leave an imperfection in your work as a sign of humility. Probably will pull it apart again to replace the bulb, just not today.
 
181 - 197 of 197 Posts
Top