What Component in a PS2 Engine Can Overheat?

Most assuredly it is quite possible to not only overheat it, but kill it as well.  For the early PS/2 5V board, my guess would be the motor driver FET's first.   That would be the two chips at the end of those long heatsink strips.  I've seen those cooked on a number of locomotives almost instantly for a motor stall, so clearly they are running near their max ratings.

@Scott J posted:

Is it possible to overheat an engine pulling an excessive load?YES

I have had a couple of early model PS2's do sudden stops and read somewhere that this will happen if the engine overheats. Is it the actual motor that overheats or something in the electronics stack? Thanks

The entire PS2 5V board...

There's more than one reason they moved to PS2 3V and repair and reliability most definitely were factors.

An over heated motor still spins until the insulation on the windings breaks down shorts to AC frame, and then it blows the electronics. I've seen it, it's ugly. Your engine doesn't work later even if the motor cools off. Again, cooked windings, I've seen first hand.

John and Vernon, thanks for the no-doubt-about-it answers! Both PS2's are pulling about 12 cars on mostly 036 curves at about 15 smph. I can see where that's a tough pull for these engines over a long period of time. If I can figure out how I'll measure motor housing and FET temperatures directly and report back.

If your talking under a motion load you really only have the rectifier, motor fets, wiring and motor itself carrying the current load.  What is the amp draw when doing this.  Other than a stall that can damage the FET, I would think the motor windings are the weak link when heavy load placed on the motor.  G

I know there are a broad range of answers but how much power (watts) does a typical engine draw? Torque x rpm at the wheels is the work (power) coming out. Torque power out minus power in is heat, mostly in the FET, Rectifier and motor windings. Does DCS report current? Of course I can measure with the Fluke current meter I have. I am an electronics thermal/reliability engineer by trade and John's comment about parts running near "max ratings" is ominous to me. Part temperatures during normal operation should never approach or exceed this limit.

The ps 2 5 volt board were run above the chip current limit design that's  why ps 2–5-volt board were done away with , ps 2 3-volt boards were ran below max current ratings!

Alan

@Scott J posted:

I know there are a broad range of answers but how much power (watts) does a typical engine draw? Torque x rpm at the wheels is the work (power) coming out. Torque power out minus power in is heat, mostly in the FET, Rectifier and motor windings.

Does DCS report current? No, it has no way of measuring. Not in any engine board nor the TIU.

Of course I can measure with the Fluke current meter I have. I am an electronics thermal/reliability engineer by trade and John's comment about parts running near "max ratings" is ominous to me. Part temperatures during normal operation should never approach or exceed this limit.

Most MTH engines at the track level draw about 1 to maybe 1.5A running down the track. More if smoke is turned on.

Obviously that does not directly translate into what the motor is seeing, however you can turn smoke off and get a baseline idea. Even under load the most I see a single MTH engine drawing is 3A at 18V.

The big problem with the PS2 5V was the stacked components and double sided PCBs being so close. In addition, electronics have advanced since then with lower resistance MOSFETs and so forth, so losses are higher in that older PS2 5V construction and design.

That's why it said it all gets hot rather than listing components. I wasn't trying to be funny, it's a practical fact of the design. The diodes might be a heat source- but given their location, it just heats the entire board, processor, memory, capacitors, and so on. They all generate some heat under load.

5V were not run above limits, but I was told ran near limits based on factors involved with constructing that board in late 90s.

5V will draw more current mostly based on Audio.  Watch the amp meter in neutral of a 5V engine and the sounds.  Smoke can add .5 or more amps to load depending on the sound file.  Sometime you get more heat especially at startup to get that big puff of smoke.

PS-2 3V run at .2 amps in neutral and in the .5amp range in motion (unless a diesel with lots of light bulbs, that is higher).  Smoke on adds the same current or less.  Usually in motion smoke on is in 1 to 1.2 amp range.  As voltage goes up amp tend to come down.  But as far as heat, if not damaged the rectifier gets warm.

Now PS-3 diesel boards seem to generate more heat in my experience.  They do have a heat sense near the motor drive section.  those boards are always warm after loading a sound file.  10-20mins of on time.  G

If the electronics don't fail first, it's the motors that will suffer.  At 15 scale mph under load, IMO they are just barely in the linear part of their voltage-rpm curve.  Without speed control, they probably couldn't maintain that slow of a speed all the way around your layout without stalling.  To make it possible, the on-board electronics are battering the motors with short, sharp-edged pulses of perhaps 18V DC.  At that speed, and with a heavy load, much of it is being converted into heat instead of motion.

IF the loco didn't have rubber tires OR if the gear ratio were lower, you might see wheelspin.  First, this would let you know for sure that the loco is overloaded.  Second, it would however briefly let the motors rev up, changing more of the input power into motion.  It would also get some air moving around the windings.  (There's a reason the motors have those oval holes in their cases!)

Lionel "solved" this problem on its Legacy locos by building in overcurrent protection.  But I don't believe the early PS2's had any built in protection against excess current draw or high temps.

That's some good feedback. I like the early PS2 steam engines because they are fairly spartan in detail and often a good deal on eBay. But didn't realize that these are very early versions of electronics that have evolved over time with newer versions. I know I've seen scope shots before but I'm guessing that the current and voltage forms at the track are not exactly sinusoidal.

The voltage at the track is most likely sinusoidal.  I was referring to the waveform going from the PS2 board stack to the motors.

The way all of these speed control systems work, is that they start with a high amplitude waveform, perhaps 18V or more, and "chop" the wave.  By providing a  high-voltage pulse for a short duty cycle, the motor interprets it the same as a full cycle or straight line DC at a much lower voltage.  The sharp leading edge of the waveform provides a lot of inductive kick.  It's like driving a nail into a board with a series of blows, instead of just pushing really hard with the head of the hammer.  This, combined with a feedback loop, is how speed control forces a motor to turn much more slowly than it would by just lowering the track voltage.

The problem is, if the motor is partially restricted from turning while all this is happening, some of that power ends up being dissipated as heat.  You might be ok at 15 scale mph, but I personally wouldn't run any slower than that when pulling heavy loads for an extended period of time.

I may have to take some temperature measurements. I have three different engines doing the abrupt stop and all three are early PS2. Last night the oldest ran for a full hour before it happened. That seems like an awfully long time to overheat but that depends on how the heat is being dissipated. Could it be a TIU issue?

@Scott J posted:

Could it be a TIU issue?

I can't imagine how the TIU would be doing this.

Wow, 12 cars at 15smph doesn’t seem like very much load.

Scott, wondering if your loco's revived after a cool-down?          j

Thanks for the feedback. What happens is a dead stop. But then when I roll the thumb wheel on the hand held remote they start moving again. The premier engine stops and immediately starts up again at full volume like I did a feature reset. I would shut it down and start it back up (it starts up normally) but if I hit the smoke button it would stop and repeat. I could see where the extra power for smoke would cause overheat if the FET was already hot. The track is mostly 036 curves and that adds drag but yeah, 12 cars and 15 smph doesn't seem that difficult. Tonight I'll run my late model PS2 (Railking P47) and see if it stops.

I've heard of infrared sensing "guns" that would allow you to read temps without burning your fingers.  Of course if you run with the shell off, it'll run cooler so that's not exactly a controlled experiment.  I have a feeling your issue is caused by some form of PS2 board failure, rather than motor heating.  Let us know what you find out.

@romiller49 posted:

Wow, 12 cars at 15smph doesn’t seem like very much load.

I agree, if the motor(s) aren't defective, I can't see this being the motor.  The fact that the board starts up with changes suggests a board problem.

How clean are the wheels, track and pick-up rollers on your layout ?

what is the power source to the track?

What gauge wire runs to the track?

I have Atlas track and keep it pretty spotless with period cleaning using Deoxit D5. Wiring is 8 gauge (maybe 16) in a pretty good star pattern around the track. Power is from a Lionel 180 watt (I think) power brick and there is a pretty expensive power conditioner between that and the wall outlet. One interesting detail is that during a track upgrade last year I blundered and created a dead short in the track that was somehow intermittent. It killed the diodes in the TIU and finally blew the fuses. I have replacement diodes but haven't soldered them in yet SO the dead diodes have been removed. GRJ said to replace the diodes as soon as you can BUT they are not needed to operate DCS. I'm trying to work up the courage to take a soldering iron to my TIU. It doesn't make sense though that this is the problem.

For temperature measurement I have a gauge that I can put on a trailing car. But need to find a 30 gauge FeCn (iron constantan) thermocouple to sneak under the shell or in the tender near or on the FET. Then I can run laps and read the gauge on each lap. Honestly though I can't see how it would take the board an hour to reach a threshold that causes the problem. For now, I'll just limit run time to 30-45 minutes.

Thanks for all the great feedback!

Sorry about your Bills Joe. I really didn't want another KC Superbowl.

The missing TVS diodes are NOT the problem.  They only come into play when a voltage spike is present at the TIU, they serve to damp it out.  However, they have no effect on the DCS signal, other than slightly degrading it due to capacitance.

@Scott J posted:

I'm trying to work up the courage to take a soldering iron to my TIU.

The TVS diodes can also be put across the output terminals of the TIU until you solder them back to the board.

@H1000 posted:

The TVS diodes can also be put across the output terminals of the TIU until you solder them back to the board.

Awesome! I would be so screwed without this forum. "For sale, entire train collection, layout is broken and owner is too dumb to fix it".

@H1000 posted:

The TVS diodes can also be put across the output terminals of the TIU until you solder them back to the board.

Excellent point and 100% true.

Gents, I have had similar issues to what Scott has described but have a different question: Are the stacks, with their FETs and other electronic components only for command operation, or are those parts of the brain likewise  required for conventional  operations as well? I ask this because my particular subject loco does run fine in conventional but has lost all ability to see, be seen or commanded by the TUI. It ran fine when I bought it but I have tended to run heavy trains, so this thread piqued my curiosity. I may have overheated the brains...???

I do not know if my loco is a 3V or  5V, which begs another question: is it possible to predict, when buying a pre-owned loco, to tell which voltage it has just by year of manufacture?

Thanks, Bob

@endless tracks posted:

I do not know if my loco is a 3V or  5V, which begs another question: is it possible to predict, when buying a pre-owned loco, to tell which voltage it has just by year of manufacture?

Thanks, Bob

I've seen different answers posted. 2003 was the transition year of PS2 from 5v to 3v. So by 2004 3v boards were the thing.

PS2 5 volt ran from 2000-2003 I believe. They were phased out.

No charging jack or a round one went with the 5 volt. Rectangle charging jack meant later 3 volt.

@endless tracks posted:

Gents, I have had similar issues to what Scott has described but have a different question: Are the stacks, with their FETs and other electronic components only for command operation, or are those parts of the brain likewise  required for conventional  operations as well? I ask this because my particular subject loco does run fine in conventional but has lost all ability to see, be seen or commanded by the TUI. It ran fine when I bought it but I have tended to run heavy trains, so this thread piqued my curiosity. I may have overheated the brains...???

Most of the components on the board are needed for any mode of operation.  There are specific parts that are not missed in conventional operation, but it's unlikely that just running it would affect those.

@endless tracks posted:

I do not know if my loco is a 3V or  5V, which begs another question: is it possible to predict, when buying a pre-owned loco, to tell which voltage it has just by year of manufacture?

Generally, anything made from 2000 through 2003 will be 5V boards. 2004 was a transition year, and by 2005, all the manufacture was the 3V board. There may have been one or two models in 2005 that still had the 5V board, but certainly not many. You can check to see which board is in by looking at the charging jack on the bottom. The round connector with a single pin is a 5V board, the rectangular connector with two pins is a 3V board.

Thanks John! I did have a separate thread about this sick loco a few months back calledl "Why did my loco eat my TIU"...Maybe you remember...which I never resolved and had to move on to other things. I was just surprised it ran on conventional but burbs on command mode. AND made the TIU non-functional for any purpose until I reset it.

I'm renovating the house now (and my eyes with cataracts next week) so a new train room is in the offing but I'm afraid a few weeks or even months away.

Thanks again,

Bob

I have replacement TVS diodes but it occurred to me that they are directional correct? The 12 I bought are arranged between a red strip of tape and a white strip. Does the red strip connect to the red terminal? Thanks!

@Scott J posted:

I have replacement TVS diodes but it occurred to me that they are directional correct? The 12 I bought are arranged between a red strip of tape and a white strip. Does the red strip connect to the red terminal? Thanks!

Being it's AC, using a single direction TVS is how you royally mess up and cause problems.

You bought bi-directional TVS diodes right?? AKA non polarized right?

Ahhh. It's been a while since I got them but yes, pretty sure they are bi-directional. GRJ gave me the part number.

Thanks!