Reverse board won't go forward, bad transistor or some thing else?

Another bad board?

I've never seen one of these but I'd guess a better than even odds one of the power transistors popped. This is undoubtedly a so-called "bridge" drive which has 4-transistors, 2-NPN and 2-PNP or 2 each of the MJF44H11 and MJF45H11. In one direction, one N and one P are "on". In the other direction the other pair are "on". If one or both in a pair fail, the motor will not spin for that direction. These are carried by Mouser and DigiKey for less than $2 each though DigiKey is out of one of them.

Do you have a voltmeter? If so and you have a steady hand (to poke around with power on) I think I can help you isolate which transistor is bad. Or one of the Lionel repair techs might step up with a direct plan of action.

I'd recommend replacing these in sets, ie. both of the ones for forward operation. Truthfully, they're only $1.62 and 1.63 from Mouser, so I'd probably just replace all four.


MJF44H11G

MJF45H11G

It looks like both F44H11's are blown. If I'm understanding how to test them correctly (Testing a transistor) there is not supposed to be continuity between the collector and emitter (pins 2 and 3) but both of mine have continuity. The other pair checks out OK.

Unless someone thinks I've missed something, I'll fist see if I can desolder them without destroying the PCB and then order the parts with my whistle project.

Since I don't have a desoldering iron, my idea for removing each transistor is to cut the three pins and then remove them one at a time.

quote:

Originally posted by stan2004:
What are you using to determine continuity?

Fluke set on Diode test.

quote:

Something's off here. If both your NPNs are shorted, I can't imagine how it can work even in one direction.

I'm assuming the motor is disconnected when you made these measurements.

Yes

quote:

Confirm there IS continuity between the emitters of the two NPN devices.

There is continuity between the two NPN emitters, likewise on the PNPs. I rechecked everything and I have continuity between pins 2-3 on both the NPNs. The PNPs check out fine, no continuity on pins 2-3.

quote:

gunrunnerjohn's idea to just be done with it and replace them all has merit. But if you want to tinker more, tolerate 20 questions, and explore the wonders of E-units along the way I'll try to help you. A photo of the bottom of the board would be handy. Is there a separate bridge rectifier off the board?

I don't think there's a separate rectifier off the board. The motors plug in directly to this board. Considering how cheap the transistors are I'll probably order one extra set of all of them, just in case. However, I like fooling around with my trains (chronic tinkerer) and learning how it works, or doesn't, which is simply another chapter in life on how to fix stuff. When I was a kid, I remember being fascinated by taking apart the open wound motor in my steam engine, cleaning and reassembling it. I suppose not much has changed.

So to be clear, with no motor connected, using Diode mode you measure each NPN between pins 2-3 (and then 3-2 backwards) and in each case you measure 0.0V (like when you touch the test leads together).

You do the same for the PNPs and in each case (pins 2-3 and 3-2) you measure "OL" or whatever the Fluke says when the test leads are not touching anything.

And if you measure, say, that diode on top of your pic of the burned K-line board you measure 0.7V (or so) in one direction and "OL" in the other direction.

Let the games begin!

Disclaimer: I don't own one. The 6 pin chips are probably opto isolators. If one is bad you will lose a direction. One is turned on for forward and the other reverse. It could also be the chip that provides the direction signal, most likely the 8 pin one with the label on it. Peal off the label and try and determine if its 555 type or microcontroller.

Pete

I'm having trouble with a 555 providing the direction logic, never seen that before. I'd be guessing that's the uP.

555 were used on some designs. I doubt its used here because of the neutral positions. More likely its a microcontroller given the label. If it is that chip you will be pretty much out of luck doing a component level repair. You could verify if its that chip if you have and identical board. It appears to be in a socket so just swap them out.
If you have to replace the boards I'd suggest one from a Williams engine. The relay design is much more bulletproof than the transistor or triac types.

Pete

I've had good luck with the Dallee electronic E-Units, put a few in. I never had an issue with them.

THis is device is similiar in design to the basic 103 DC reverse unit.

Could be the Tranistors, could be the 6 pin MOC, but they seem pretty hardy and are isolated from a derailment. They do normally use a transistor to turn the MOC on, that may be the 8 pin chip.

If the Power transistors are shorted, I would replace them and the test.

You can test the moc for continuity with a 5 volt test signal, but you will need to lift pins.

This arrangement is on most of lionels AC and DC driver boards in some variation. G

quote:

Originally posted by stan2004:
So to be clear, with no motor connected, using Diode mode you measure each NPN between pins 2-3 (and then 3-2 backwards) and in each case you measure 0.0V (like when you touch the test leads together).

To be precise it's 0.6V black on 2, red on 3; and then 0L red on 2, black on 3

quote:

You do the same for the PNPs and in each case (pins 2-3 and 3-2) you measure "OL" or whatever the Fluke says when the test leads are not touching anything.

Yes

quote:

And if you measure, say, that diode on top of your pic of the burned K-line board you measure 0.7V (or so) in one direction and "OL" in the other direction.

YEs

quote:

Let the games begin!

Indeed

quote:

Originally posted by Norton:
Disclaimer: Peal off the label and try and determine if its 555 type or microcontroller.

Pete

Here's the numbers 12C509A 04/P4K6 0351

quote:

Originally posted by Norton:
You could verify if its that chip if you have and identical board. It appears to be in a socket so just swap them out.
Pete

I do have an identical board. I'll tear into the engine that has it tomorrow and report back.

quote:

Originally posted by GGG:
You can test the moc for continuity with a 5 volt test signal, but you will need to lift pins.

DO you mean it will need to be disconnected from the board before I can do the continuity test?

quote:

Originally posted by iakobos:

quote:

Originally posted by stan2004:
So to be clear, with no motor connected, using Diode mode you measure each NPN between pins 2-3 (and then 3-2 backwards) and in each case you measure 0.0V (like when you touch the test leads together).

To be precise it's 0.6V black on 2, red on 3; and then 0L red on 2, black on 3

"That's a relief" says Detective Thomson (without a p). The NPNs aren't shorted and order is restored to the universe. You're measuring the protection diodes placed externally across the transistors - standard practice in a motor bridge.

I don't know which path you want to pursue (swap processors, testing optoisolators, just changing all transistors, etc.) but here's what I'd do:

Using the DC-Volts mode, with the engine powered up and motor connected, somehow hold or affix the black test lead to one of the NPN emitters (the two emitters are shorted as established before). With the engine commanded to forward and reverse, measure the red test lead voltage at each motor terminal. Obviously this will be trickier in reverse since it wants to move but perhaps not so bad with a low enough track voltage.

Here's how it works. In Fwd, one PNP switches positive voltage onto the "+" motor terminal and its mating NPN puts a less positive voltage onto the "-" motor terminal. The other two transistors are off. The motor sees a net positive voltage and spins forward. In Rev, the first pair of transistors are off and the other PNP puts positive voltage onto the "-" terminal and its mating NPN puts a less positive voltage onto the "+" terminal. The motor sees a net negative voltage and spins backwards.

In making the measurements as suggested, I believe you will be able to see which transistor has failed (if it's just one). That is, if either the PNP or NPN does not switch on, both motor terminals will read the same voltage set by the working transistor.

Then we can work back to confirm it is in fact a bad transistor or if whatever is driving the transistor (the MOC optoisolator, and before that the microprocessor) is busted.

quote:

You can test the moc for continuity with a 5 volt test signal, but you will need to lift pins.

Be very careful if you test the optocoupler. The input side is an LED that must have the same current limiting resistance that other LEDs require.

I test opto couplers in the circuit. Can you read a part number? With that you or one of us can determine which pin is the output. With the engine on the track or the board powered you can determine if the outputs toggle when you hit the direction button. This will tell you if the problem is the optos or circuitry before the optos or if the problem is after the optos like the power devices.

Pete