Another strange problem!

Dirty wheels, dirty rollers, dirty track.

.....

Dennis

Possible a wire is making sufficient contact at a connection point to give you the full volt read. Is it giving the same reading while the track is under load?  Just a thought. 

Yes, the dirty track idea is my main focus.  I scrubbed and scrubbed an area of track, (discovered that I'm out of track cleaner), and neither caboose lit up, but lit up on the other main again just fine.  The reading i get on the multi-meter is the same under load as it is without.  My friend has been running exclusively on that inner main, and he's complained that his wheels are pretty dirty.  I'm going to get some track cleaner and super clean both mains; I just couldn't believe it had gotten that dirty.  I clean it fairly regularly with a good cleaning car, but it's mainly done dry.  I think I have to change my procedures.  

Bob Blomberg

If you have smoking engines and cabooses with smoke, you need to clean your track very often. The smoke fluid doesn't all burn off and some goes onto the tracks.

Also for what it's worth, postwar rolling stock wheels need a good cleaning more often then the newer wheelsets.

Lee F.

Well, I don't believe the track needs to be any cleaner than you can get it with a dry scrubber.  Mine works fine that way.  Check to be sure the roller and especially the roller axles are clean.  Also the wheels should be checked and cleaned.

.....

Dennis

  Double check the power connections to the track. Although you may get a good voltage reading, a bad connection will not allow the current to flow properly. Put a caboose on the track with the lockon and jiggle the wires. Pull off the lockon and check that it is not dirty or rusted. Disconnect the wires from the lockon and hold them on the track directly. Finally, touch the wires directly to the pickup roller and wheels to see if the caboose lights.

Larry

Wasn't the track, or any connections; for the second time in a year, channel 4 of my TIU went haywire.  I checked again with a multi-meter, and lo and behold, it was reading 48.7 volts!  The reading on the transformer was 18 volts.  So, I went below and quickly found that the input to the TIU was at 18 volts, and the output was reading 48 volts.  It just means that I won't be running conventional stuff on that main, which is OK, because I don't do much of that anyway.  The last time, the channel completely blew out.  

Bob Blomberg

48 volts!  Wow!!!!!  I don't like to go near anything over 20 volts, ever . . . .

48 volts? Sounds like a diode may have gone bad, at least that is my guess.

I put a multi-meter to a 20 volt A.C. wall type transformer and it read 48 volts D.C. on the D.C. scale, interesing eh? This is one that works correctly.

Lee F.

That just doesn't sound right as a TIU can't make more voltage than what's being fed to it. There is no voltage doubler/ tripler or quadrupler circuit in it. I think your cheap multimeter is giving you a false reading.

The reading on the transformer was 18 volts.  So, I went below and quickly found that the input to the TIU was at 18 volts, and the output was reading 48 volts.

It's not the TIU, period. As Lee mentioned above, the TIU simply cannot manufacture a higher voltage than what goes into it.

Are you measuring the voltage at the TIU outputs with the wires to the outputs connected or disconnected?

If connected, the problem lies with the power already in the tracks. This can be due to a track connection to a second power source or due to something bridging the power between two different transformers, possibly out of phase with respect to each other.

If the wires are not connected, your meter may be having a problem with some transient voltage or DC in the AC.

Try swapping the channel's inputs and outputs with another channel. If the other channel now reads insanely high, the problem is in the tracks or the wiring.

No, sorry, but a multimeter that reads everything else correctly won't just read something incorrectly.  And it's hardly a cheap multimeter; it's the same kind I've always used in the past, including while working for an airline and a railroad.  Now try saying something constructive, Chuck.

I did all that, Barry.  Input to the TIU was 18 volts; output was 48, at the terminals.  Everything now works fine using the other outputs.

Bob Blomberg

I reported this same exact problem a few weeks ago.   Hate to tell you guys but it is the TIU.  I have the exact problem.  Everybody told me it was a bad meter but thats baloney as I have used a high end Fluke with the same reading  I repeat   Its the TIU   This is the third person that had this.

Many years ago when the HO club I use to belong to installed DCC on the layout we had a problem where the booster would push out move vlotage than was inputed and at the time it was discovered the distance from the booster to track feeders and where the locomotive was would cause the booster to send out higher voltage.  this was discovered by an electrical engineer in our club at the time

Something strange is going on here.  In your original post, you reported 18-19 volts at the track, not 42 volts.  I suspect 42 volts would have destroyed your lights and probably damaged your engines.  But then you said you put the cabooses on another track and they worked fine, so you didn't destroy the lights which leads me to believe there never was 42 volts on the track.

Are you sure you are checking the voltage under load?

Earl

Can anyone post a link to a schematic of the TIU?

Dale H

Could be 42 volts with no amperage.

Dale H

Here is what has happened on our TIU  We had a problem in our yard  DCS engines would only run in conventional and run very slow.  That area is on Variable one of TIU 3 in our layout  We have all of the variables to fixed.  When I measure track voltage it was 10 volts.  When I measure voltage at the TIU the Variable 1 input read 19 volts and variable one output read 10 volts.  When I disconnected the output the voltage measured 57 volts.  So yes its high voltage with no amps because as soon as you connect a load it drops down to 10 volts

Voltages measured with no load are pretty much meaningless.

Earl