Diode Strings vs. Series Wiring

Well, I rewired in the diode string. Ran it on the test stand. Diodes got a bit warm, but not hot, so I put it on the layout with the big consist behind it (no shell on it and alligator clipped in). It pulled well and smoothly so I ran it for about 6 or 7 minutes around the loop. Stopped it and hit it with my infra red meter. Both motors were around 83 degrees (within a degree of one another). The highest temp I got in scanning down the chain was 115. Much better than the 160 I had. How warm are these supposed to get? I don't trust it enough to put the shell back on (it will get warmer in there with that). The consist was running at 3.7amp (also a big improvement).   Any advice as to how warm these things are supposed to get and what temp would be dangerous to the shell and components.

Roger

They will dissipate some heat, so they will have a temperature rise above ambient.  I wouldn't worry until about 130-140F, but any hotter than that and I'd be looking for the cause.  They are dropping around .6-.7 volts at whatever the actual motor current is, so they will get warm.

Roger just wondering if you changed the diodes or just rewired them as you got some

big changes in both heat and amp draw for just a rewire ???

I didn't change the diodes......just put the chains as they were back in. Unfortunately, now I'm getting some reading of 144 in spots along the chains. And that's only with a few minutes run time. Both motors are still in the 80s, so I don't think it's a motor problem or bind.  But I don't trust this.

Roger

Ok......here's a very curious thing I just discovered. I ran the engine by itself and it drew about 1.4amps. I could hear the rail sounds of the consist sitting in the siding (which was at that point powered by the main loop). So, I cut the power to the siding and the engine seemed to speed up. I measured it's speed around the loop.......11 seconds by itself, 13 at the same voltage, but power going to the siding and consist (albeit not moving). That's a noticeable difference. Why should it slow down even though it's not pulling the consist? This is powered by a ZW.....plenty of power.

Roger

I think it's time to test the motors individually with a DC supply and measure the current when they're running without any load, i.e. trucks suspended.

In order to diagnose this, you need to start eliminating possibilities.

I just ran it some more by itself and the diodes barely got warm. As soon as that big losd is behind it, they get hot.

I don't have a DC transformer. Any other options to do that?

Roger

John,

Wouldn't a bad motor get warm under load? These don't get warm.

I'll be away for a few hours...so, I'm not ignoring the responses.  Thanks

Roger

John.....last quick message before I head out. I called Fluke to find out how to set up my 115 for DC amps. Never used that function before. Easier than I anticipated. And how would this work for a DC supply?   Hook my test transformer (1033) to one of my DC to DC adjustable buck modules with a wave rectifier in front. Yellow and blue motor wires to the outputs?

Roger

The bridge should be fine, you just connect the motor and the meter in series with each other and then check the current at a reasonable speed.  If it's drawing more than 200-300ma free-running, I'd be somewhat concerned.

Again,the string of 4 paired diodes will drop 2.4 to 2.8 (volts) times the amperage drawn by the motors. The string acts like a variable resistor,always dropping a consistent voltage to maintain the drop.  Voltage drop times amperage equals wattage consumed. So if the motors draw 3 amps,the diode strings would consume about 8.4 watts,the 2 motors would draw 45 watts at 15 volts. Each diode would have to dissipate 2 watts. Assuming it is not crammed in and breathes a little it should not be a problem. Only 4 diodes work at a time depending on train direction.

I can not imagine 2 motors pulling more than 4 amps or 60 watts at 15 volts. The reverse board I think uses 6 amp bridge rectifiers. It drops the circuit also by .6 to .7 volts and also uses 2.8 watts at 4 amps,just like each of 4 diodes. I have not seen an instance of this chip failing. (On MTH PS1 boards I have seen some melting along solder joints on the 6 amp diodes used,they may have been soldered inadequately.) Then the string would shed around 11 watts, or less than 3 watts per diode.  This would be a worst case scenario IMO unless the motors had a problem. Perhaps a bad winding or something. The only other thing I could think of that would heat the string is if there were a bad or cold  solder in the string that had resistance.  

How hot would they get? With 8 watts,warm but not boiling hot. I have never burned up a 6 amp diode this way, but I do not run my trains real hard either. Take a Williams passenger car with stock lighting. They have 2, 4 watt bulbs rated at 18 volts. Add a third bulb for the 4 amp draw example.  Yes they put out a bit of heat,but I do not think they would do damage unless smothered and against plastic. In my link I do instruct not to jam them in,but to leave a bit of space. The heat has to be considered. 

(Corrected, sorry I was not clear. The current flow is the same through the circuit.  thanks John)

Dale H

Dale, I hope you meant to say voltage.  NO way the diodes handle more current than the motors, they're all in series.

I still can't see why this engine is drawing that kind of power with no load.  My Vision Line Big Boy with all four smoke units running draws less than four amps peak moving at a good clip.  It has two audio systems, lights everywhere, etc.  It also has a honkin' big motor, and the sucker weighs 16 pounds engine and tender!

John,

A couple of things.......the engine's diodes are only getting hot/warm when the consist is behind it. Without the load, they run cool.   I set up the Buck converter with the rectifier hooked to my test transformer and attached the leads from a motor to it. Ran well. Then I up the Fluke in series with it and got no reading. Discovered that unfortunately, the fuse was dead. Getting one of these fuses is a problem. No place local has them so I'll have to order one/some. It will take a few days before I can run that amp test. Anything else I can test while I'm waiting for a fuse?   Thing is.......both motors run cool even under load. If I had a bad motor, wouldn't that motor get warm/hot along with the diodes?

Roger

Those Fluke fuses are a pain. Order more than one. 

Look at the size of the motors vs. the size of the diodes.  The motors can dissipate a LOT more heat before they get hot.

My problem with your results is they don't make a lot of sense for what I've experienced.  I'm sure I haven't "done it all", but I have worked on a lot of locomotives of all descriptions, and many Williams dual-motored ones as well.  My experience with the motors is they don't draw nearly enough current to overheat 6A series diodes.

Funny thing Chuck, I have never blown the fuse in any of my meters.  I know they're there, it says "fused" right on the front.

John,

I found a place nearby here on Long Island that has a fuse. Going there this afternoon to get it. I don't know why that fuse was blown as I don't use the ammeter function (I've got a clamp-on for that). So I'll get some results on the motors later.  These cans run cooler than my Pullmors. When I'm testing an F3 with the meter, I typically get 95-97 degrees after 10 minutes of running. These are at 83 or so.

Anything else I can check on?

Roger

The fuses in the Fluke meters are quite fast. Yours is the 11 amp, 1000 volt fuse and when you move the lead to A and the other lead is of course on COM, you will find that unless you are extremely careful not to then try to use it to measure voltage across a circuit, the fuse will blow. Saves the meter though!

It's so fast, you will not realize that it's blown, but you will realize that you have the red lead in A and then move it to Volts. The fuses cost $7 to $30 each depending on where you buy them. Amazon has them for $7 and up. I would buy one if it's a high priced store and get a couple spares from Amazon.

Chuck,

Just got back from a Grainger for the fuse. 19 bucks there, but needed it right away. Here's where I'm going to buy a few from later......

http://www.ebay.com/itm/251899...e=STRK%3AMEBIDX%3AIT

In reading your post, that sequence is exactly what I must have done to lose the fuse. Have to remember that.

R

John,

Just popped the new fuse in the meter and measured both motors. The both came out identical. At 9.46VDC output on the Buck (with no load), they both drew 1.12 amp (individually).    So, what's the verdict?

Roger 

Correction.......one motor drew .248amp and the other .282amp. I realized they were not completely suspended. That seems to be within your stated range.    

Roger

Thanks for the link. I should get a couple spares...even tho I have been blowing these fuses since the '70s I still forget.

I have no problem with the 1/4A figure, but if you actually draw over an amp from a Williams can motor that is running with no load other than the truck, something is wrong.

So, the question remains.....if the motors are ok, why are the diodes getting hot?  

I have no idea.  I think I'd have to have a closer look to figure it out.  Obviously, there is a reason...

The only thing left in there is the reverse board. The Tru-Blast is gone.  Took it out and put an ERR Sound Commander in one of the B units. How many amps do those diodes have to pull to go from warm to hot?

I can't think of anything the reverse board could do to make the diodes heat up like that.  In point of fact, the reverse board has a smaller bridge rectifier that is processing all the motor current, so if those diodes are getting hot, the bridge rectifier must really be cookin'!

Here's a possible.........I just revisited the diode tests. Fluke states that diodes should test out between .5 and .8V.    When I tested mine the other day, they all came out almost exactly the same.......but looking back on it, I think there may be a problem. They were all approx.   .49V.   I assumed (perhaps wrongly).....how could they all be bad? That's very close to the .5 value.  But, I just tested a smaller RadioShack diode (the ones in my chain are RS) and it was .56.......significantly higher.  Could those lower values contribute to the heat?

John,

Another thing.....what you just wrote about the rectifier. When I did my sweeps of the chain with my infra red meter,  when it got warm/hot, it wasn't uniform along the length. There was definitely a hotter spot.....and perhaps the rectifier was under the beam at that spot. Conjecture.

The larger diode will drop less voltage for the same current, and the lower voltage drop actually means less power dissipation and less heat, not more.

You might get a hotter spot right at the junction near the middle of the diode body, that's perfectly normal.

The diodes aren't bad.

That's too bad......would have been an easy fix.

Even cheaper.......http://www.ebay.com/itm/171758915812?_trksid=p2060353.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

Just ordered a few of them. Good deal

Those sure are expensive fuses, even at the reduced prices.

I don't think the size is standard. Also they are fast and resist arcs with a 1000 v rating when they open. I guess Fluke was trying to protect the meters, s ome of which did cost hundreds of dollars. 

I can say that some decades ago in the lab someone put a meter set up for amps across the power line carbonizing the circuit board and blowing traces off the board. If it had been a Fluke, it would have just blown that not so expensive of a fuse, all things considered.