Most of those units are wired in parallel, so you can just change the connections to wire them from one connector. The connectors illustrated below by the red arrows and the yellow arrows are wired in parallel, all the pins on one connector are the same on the matching connector. So, running the motor from one connector would solve the problem of the wiring.
Ah, didn't see your post as I was in the basement clipping in that rectifier experimental pair into a diesel. So, I just clipped the ends of the yellow wires to it as above. Your method would have been easier, I think. Future mods.
In any case, it's working. I'm going to try it on the track to test it out. As I mentioned, I'm experimenting with the 4amp rectifiers I had on hand. When the 6 amp ones arrive, should I swap them out?
Going to use my Fluke infra red meter to get a heat reading on these rectifiers.
Roger
The 4A ones will be fine, I can't imagine you drawing anything close that that current with a single locomotive with a couple of can motors. That's probably more than the stall current! My measured stall current of the Mabuchi RS385 with 10 VDC on it is less than 2 amps, so unless you stall it and apply full throttle until the diodes cook, 4A will be plenty.
Ok......just tested out the experimental rectifier pair in an F3 I had previously wired in series. NICE. I'm liking it so far. Nice, smooth slow startups under load without the jackrabbit thing. And it seems to have more pulling power than the series situation.
Now, about temps.......I ran it for about 5 minutes pulling a consist and popped the shell off and took a measurement. I got an accurate measurement of 130 degrees. When I first hit it, it showed about 140.....may have been the fact that the shell was on. FYI.....before I put it on the track with a consist, I had it clipped to my test transformer and took an amp reading........ .7amps
1. At what temp do I have to be concerned about the shell?
2. These are 4 amp rectifiers.....will the 6amp versions run cooler?
3. How much more heat will there be if I decide to use 3? 3 might work out better for the smoke unit in the Hudson. I'll have to see after they are installed.
Roger
140F is 60C, well within the range of the diodes. It's hard to believe they even got to that with .7 amps. If they're not right on the shell, I can't imagine it being a problem. 6A bridges would be a bit cooler, just because they have more mass. Of course, remember than any bridge that has a mounting provision like a hole is designed to be on a heatsink, and that's how they're rated.
If you want to dig deeper, take a look at the bridge rectifier datasheet. Here's what a typical one might look like.
What you're trying to do is insure that the semiconductor junction of the diode(s) operates within its limits which is essentially 150C (not F) for most if not all train electronics. All datasheets will tell you the thermal resistance of the package/body. You measured the case at 60C. If the bridge carried zero current; then it is dissipating 0 Watts and the semiconductor junction would be the same temp as the case. In your case, you're carrying say 1 Amp of current across 2 diodes. Watts = Voltage x Current...so at just under 1V per diode, your device is dissipating at most 2 Watts.
Using the thermal resistance of 2.4 C/W, that means if the case is at 60C, the junction is operating about 5C warmer or ~65C...well below the 150C limit. Some liberties have been taken here with the numbers and definitions but the point is you're nowhere near pushing the envelope.
Stan and John,
You guys are fonts of information and indispensable to this Forum. Thanks! Also........would 3 rectifiers wired together actually be cooler than 2 because of the greater surface area?
John,
The .7amp reading was taken clipped to my test transformer with zero load. The warm temps were recorded after pulling a consist. Should I bother with heat sinks? I definitely won't with the Hudson as the tender shell is die cast. I was also thinking that the smoke unit in a diesel with a plastic shell has to get a lot hotter than these diodes.
If I were a betting man, I would put money on the belief that most folks who have Williams engines and do a mod on them to slow them down, do series wiring. After seeing how these rectifiers work, to me it's night and day.
Roger
You don't really need a heatsink, I'd just keep the bridge away from contacting the shell. A couple of watts power dissipation in there is nothing.
Yes, the smoke unit gets hotter. I regularly use hot glue on the shell, and I can tell you for sure it's hotter than 60C. 
Ok, sounds like a plan.
But speaking of heat, I did a dumb thing a couple of years ago. I was upgrading a dummy SW-1500. Put a smoke unit in it and was going to add a variety of LED lighting. One of them would go in the cab. Ordinarily, I rarely use incandescents (except for interiors of buildings) but I had a grain of wheat bulb in my electrical draw and thought......why not use that in the cab? Mistake. I installed it and it looked great. Ran the engine and it's new dummy around the layout for about 10 minutes and glanced at the cab. Couldn't believe my eyes......the roof was dimpled. $%#@! Well, what I did was.....I installed a red flasher LED in the center of the dimple, which just happened to be centered at the rear of the cab. Instant strobe. Lucked out with that, but I've been very wary of heat ever since.
Roger
Bulbs can get a little hot, especially if you run them over their ratings. Here's one I got in for repair...
No, I couldn't rescue that cab interior, I ended up getting a replacement and rewiring the new one.
Now that's a nasty one.
ROGER1 posted:...would 3 rectifiers wired together actually be cooler than 2 because of the greater surface area?
With a dissipation of 2 watts for each bridge, which is probably a reasonable figure for normal running, other than keeping the bridge from being directly in contact with the shell, I don't see any issues at all. The smoke unit dissipates from 5 to 7 watts, depending on how hard you're pushing it, it's right next to the shell, and it's not an issue. Sure, it has more mass and surface area to spread the heat, but the heat is still mostly inside the shell. If you're pushing an amp across the motors, they're also dissipating quite a bit of heat, but that's not causing any issues, right?
Roger, this really is a non-issue. Just keep the bridge(s) away from the shell and run the trains.
I mounted the pair in the diesel last night, attached to a small piece of wood that holds them upright with airflow around them. I'll test it out later.
My thread drifted from smoke units to this more interesting topic of controlling the voltage in Williams engines. Wish I could rename it so more people can take advantage of the tips that are coming out here.
Roger
You can rename it, just edit the first post and change the title.
Title updated
Just ran the F3 with the rectifier pair in it, all wired up and shell back on. Wonderful. Going to do this with my other two series wired engines. But as for the Hudson, I may have to use 3 bridges. I've found that it has to run at about 12V to get significant smoke and two might not do it. I'll decide when the new ones arrive. But if smoke is not an issue, I think 2 is perfect for this.
Roger
If you're trying to fine-tune the smoke-speed relation, you can tap the bridge at the half-way point (where + and - are connected) to get "half" a bridge of voltage drop so to speak. So rather than just under 2V of drop per bridge you can get just under 1V of drop. Perhaps obvious but I don't think it's been mentioned in this thread. Depending on load that seemingly small voltage difference can be 5 scale MPH or more.
Good point Stan as 1 volt sometimes makes a significant difference in smoke output as well.
Well.....my rectifiers arrived this afternoon and I wired up a pair of them. Clipped them to the reverse board on the open tender and ran it around the layout. Major problem. I think I fried the reverse board. I tested the rectifier pair before I clipped it in and it was reducing the voltage by at least 3 volts. Put the engine and the tender on the layout and it made it about 1/2 way around and stopped. Lights, smoke ,sound all worked but no motion. I tested the voltage OUT of the reverse board. 0 volts. My mistake, I think. I laid the rectifier pair lightly on top of the tender boards figuring that the shrink for the rectifier leads covered most of the "legs" of the diode. I'm thinking it probably shorted the board. Or......coincidentally burned out (there is no smell, nor is there evidence of anything burned). I clipped the Hudson harness into the reverse board of another Williams engine and powered it up. The Hudson's motor worked. So.....what I may do is take the reverse unit out of that other engine and use it for the Hudson and try to find another reverse on the Bay. Is there anything else that may have burned out the reverse unit other than the diodes making contact with it?
Roger
If you shorted either motor lead to the frame, you very likely cooked something, they MUST be isolated. Either the drivers or the input bridge on the reverse board are likely suspects to be dead. There's nothing about putting the series diodes in the motor lead that should adversely affect the reverse board.
John,
Something strange is going on here. I used the reverse unit from the other engine and was back in operation. Ran it without the rec. pair and then clipped in the diodes (only this time I taped around them to keep them from hitting anything. It ran about 10 feet and then my breaker tripped. Checked the wheels.....all ok. Ran it again.....another trip. Put my meter on it and it was running at 10 amps! Immediately removed the diodes and tested it again.....no trips and it ran at about 2.5amps. ?????
I think I just thought of a possibility. I've got the pair connected to the output from the reverse board to the motor leads. There are 4 wire nuts connected to the board. 2 are AC in from the rollers. I tapped into those for the ERR RS unit. The other two are DC out to the motor......blue wire in one, yellow in the other. But there's a black wire in both. I'm thinking about that and what you mentioned about ground isolation. That black wire is probably ground. Should I just connect the diode pair to the yellow wire and the blue wire without the black ones?
Roger
Well, I fixed it. Had it wired incorrectly. Accounted for the big amp draw. Expensive mistake.....time to find a new reverse unit for the other engine. In any case, it's running at 2.5amp pulling a load and.....the smoke works as it requires at least 12V to pull that consist (with the diode pair in place). I may try the half diode that Stan suggested. Would I connect that second bride AC out to the AC in the third diode and pig tail the + and - together and to a wire OUT . And the AC out on that diode would be disconnected. Correct?
I think you answered your own question, the black wire should NOT be wired to the motors in any way.
The diode pairs go in ONE of the motor wires, either one. They are just in series with the motors to drop the applied voltage.
John,
Yes, something didn't look right and I figured it out. Don't know why I lost the reverse board with my 91 breakers (instantaneous) and TVS. Anyway, I'll find a replacement. Lesson learned.
Any downside to using 3 rectifiers or 2 1/2? I checked the voltage drop on the 2 1/2 setup (clipped it) and it was 3.25V. I'll have to run it a bit more with 2 to get a feel for how the smoke looks. It smoked well with a fairly heavy consist. I'll have to try something lighter.
Roger
You can use any combination of diode pairs, one... ten... etc. Obviously, the more diode pairs you stick in the motor leads, the less top-end you'll have with a heavy load. That may not be a factor, just something to keep in mind.
I think there is one side effect I noticed the other night when I installed the 4amp pair in my F3 (although I do need to verify it). As I mentioned, I typically run my transformers with 91 breakers to protect them. Extremely fast, but the one downside is their upper end is not that high (somewhere close to 6amps but probably lower). Normally, it's not an issue (and I do have a toggle that can switch to a larger amperage thermal). With the pair of rectifiers in that F3 and pulling a consist of incandescent illuminated passenger cars, it put me very close to tripping the 91 (especially at startup of the pull). I'm pretty sure that if I added another rectifier to that particular engine, I'd be in tripping territory. Does that seem logical? Of course, I could toggle to the other breaker, albeit it isn't anywhere near as fast.
Roger
Exactly what reverse unit are you using? Or do you have a photo? Is it DOA in both directions?
It appears you're able/willing to work at the component level and if you indeed "fried" the reverse unit by short to the frame it is likely the one (or more) transistors on the reverse unit. Sure it's easier just to replace the entire unit and reverse units are inexpensive, but if you're of the DIY ilk there's a sense of satisfaction in repairing something. In all likelihood you can replace the transistors (requires soldering) and be back in business. Transistors used in reverse units should be less than $1 each. Just saying....
Stan, I think he's using the Williams Reverse Unit. That's the one we were discussing.
Hi Stan,
I love tinkering......to me I like working on this stuff more than actually running them. I generally only work on boards when they are toast as I have nothing to lose. However, what I can diagnose is a bit limited. I can test resistors and capacitors, but the more complex things in there.....I don't know how to go about it. What would be the most likely culprit to be fried?
I started tinkering/testing some things on the bad board. The solder used on this board is very hard and I would have to crank up my Hakko higher than the 600 degrees I use for the solder type I have. I know that it's best to remove the components but I seem to be getting pretty accurate readings on the caps and resistors. They all seem ok. Then I turned to diodes. When I do a diode test with my Fluke on a fresh diode, I get OL in one direction and a low reading in the other. The two diodes on the right side of the board produce the same results as the fresh diode. But there's a diode at the top to the right of the white rectangular component and another at the bottom....same relative position. I get identical readings for both diodes in both directions (equal to the reading I got on the other two, but no OL in the other direction). So, I took a chance that they are bad and alligator clipped a large fresh diode to each side (without removing the old diode). I still got no motor activity when I turned it on. Is that because the old diode has to be removed? Or am I totally off base with what I attempted to do.
Remove the old diodes, I suspect grounding one lead of the motor may have killed those. FWIW, for lead-free solder, I run the iron at 650F, for tin/lead, 600F. For circuit board work, I use 550F.
I was going to ask you about those temps. When I got my Hakko years ago, I followed one of your recommendations about the 63/37 solder. Great stuff. A pleasure to work with. And I use 600 for that. But the solder on this circuit board is very hard and at 600 I couldn't seem to soften it. I suspect lead free.
I've decided to use the 2 1/2 rectifier option on the Hudson. I do get smoke with 2, but with the other 1/2 I get real good output through most engine speeds. And I come nowhere near top end speed of that loco. So, while I had a setback with that reverse board, I'm really liking the results here.
IMO, as far as usability, the 63/37 rosin core solder is the best. The lead-free stuff may save the planet, but it's a much bigger PITA to use. 
If you didn't get a calibrator for your Hakko, I'd consider it. When I first checked after getting the calibrator, my iron was almost 50F off the correct temperature.
After I read the thread on the calibrator, I bought one of those (what a deal!) and a second iron for a finer tip.
Took out the two old diodes and clipped fresh ones in (larger, 3A.....all I had). No go. Ordered a new one.
I've got a transistor question for you guys. I ordered a new reverse unit, but I'm going to try and fix this fried board. I'm getting a Digi-Key order ready. Ordering some N4002 diodes to replace the ones I removed and I want to swap out the three transistors I found on the board. 2 of them are ZTX692B. I found them on the DK search feature (ordered one extra). The third one is larger and the whole number is ST2N2222AVNC. I had a lot of trouble finding it until I dropped off the first 2 letters and came up with this......
https://www.digikey.com/produc...2N2222ACS-ND/4806845
The three transistors on my board had the typical half round profile (viewed from above). The one in the link appears to be round. That might be a problem for me in putting it back in the board as I know which way the flat side of the half round original was facing. Does it matter and is this the correct transistor?
Roger
I hope you mean 1N4002 diodes. FWIW, I keep a lot of 1N4003 diodes around, they're handy in a lot of places, and they're dirt cheap in quantity. By the same token, I keep garden variety transistors around as well, and the 2N2222A is a pretty universal NPN transistor, good for lots of things.
For the transistor base diagrams, does this help? Note the little tab on the metal can near pin 1.
I saw that tab in the photo and had a feeling it had to do with orientation. That diagram was perfect. This whole transistor thing is new territory. Looking forward to work on that board. Thanks
ROGER1 posted:...The third one is larger and the whole number is ST2N2222AVNC. I had a lot of trouble finding it until I dropped off the first 2 letters and came up with this......
Access to this requires an OGR Forum Supporting Membership
