I was inspired and began working on a project which necessitates converting AC to DC and I went to my parts bin and pulled out a few KBP307 rectifiers. I take the middle two pins and connect them to AC of 7-18vac out of my TW A-U terminals and I get nada out of the outer pins.
1) Are the KBP307 a viable rectifier for dealing with Lionel AC voltages and up to 3 amps?
2) Is there a better rectifier I should be using?
Replies sorted oldest to newest
you want to monitor the ac input to the full wave bridge rectifier , make sure your reading some ac voltage, maybe your transformer can't handle the load of the bridge rectifier or maybe the rectifier has shorted diodes causing no voltage. do you have another transformer to use. don't forget the out side pins of the rectifier will be vdc volts dc out put so your meter must be on dc volts to read voltage!
let me know how you make out beb glad to help you any way I can!
Alan
Alan, my TW transformer is in working order and puts out 7-18vac. When connected directly (and only) to the KBP307 rectifier, I get readings of under 1 vdc.
what reading do you get on the ac side while outputting 1 vdc
Alan
The AC side still shows the input voltage with no noticeable drop or load.
Ive done this many times, and have not had a problem. I would be suspicious the rectifier is bad. Radio shack sells some pretty cheap ones.
What size breaker does your transformer have? I would use a bridge rectifier capable of 20% more load then your transformer puts out.
Maybe a 6 amp 50 volt rectifier would work better and not give out under load. I suspect your bridge rectifier is too weak, a few amps more rating is not that expensive, so I would get a higher rated bridge rectifier.
FYI; I am using 6 amp 50 volt bridge rectifiers on my Williams engines.
Lee Fritz
Those rectifiers should work fine, I'd suspect something else is in play here.
Here is a visual representation of my setup - for testing, it is just a Lionel TW transformer (175 Watts across two coils, so really < 90 watts on the AU side) , wires, rectifier and volt meters. There is no load currently. I have about 20 or so of the KBP307 rectifiers in my parts bins and are rated for 3 amps and 700 volts so I think they are well within range for this test. For those saying the rectifier is bad, I've tried 5-7 of them and I cannot image they are all bad (but who knows!)
Try putting a resistor across the DC outputs, maybe 470 to 1,000 ohms. This will give the meter a small load to work from.
Presumably you tested your meter's DC function by measuring a 1.5V or 9V battery or whatever? Be sure to reverse the leads and confirm whatever DC reading value changes sign (from + to -).
Assuming you have a digital multimeter, most have a "diode test" function. A bridge is just a packaging of 4 diodes...draw out the diagram of a bridge and you can go thru one by one testing the internal diodes. Here's a "freebie" Harbor Freight meter doing so - showing about 1/2 Volt for one diode of the bridge. Each diode will read ~1/2 Volt (473mV shown in pic), the same 4 pairs with meter leads reversed should read nothing.
Well, this is embarrassing... turns out after all this fuss I discovered one of my test wires was tested for snackiness by one of my feline companions. Once I replaced the lead, things started working as expected! I have been able to build upon my inspiration and unveil my final plan, but it appears I'm getting a short because the DC ground and AC common are connected through the - terminal of the battery adapter...
In which case perhaps consider just using a single-diode (half-wave) to convert AC to DC so that you can share AC-common with DC-ground for the battery adapter. Going from track voltage down to 1.5V DC is quite a drop so your regulator IC is wasting a lot of power (and get hot). Looks like you have some space on the chassis; perhaps consider one of those 99 cent eBay DC-to-DC converter modules to convert the half-wave DC to 1.5V DC.
I'd just use a diode, fairly large capacitor, and the LM317T. Bolt the LM317T to the chassis for heatsinking and you're good to go.
FWIW, most of my electronic projects for interfacing with locomotive electronics uses a single diode for the reasons Stan stated, I need a common AC and DC ground.
gunrunnerjohn posted:I'd just use a diode, fairly large capacitor, and the LM317T. Bolt the LM317T to the chassis for heatsinking and you're good to go.
Metaphorically that is! The metal tab (with the screw hole) on a LM317T is connected to the "output" pin. If directly screwed to the chassis, that would short the LM317T output with the AC-common for most engine chassis. So use insulated mounting parts or use the version of the LM317 that has a non-metallic tab with screw-hole.
Excellent points Stan, I make assumptions I shouldn't make! Probably look at the LM317P for ease of use.
I think I have some axial diodes that I could dig out and some ~500uf capacitors and I should be able to package that up in the space allocated by the battery adapter holder. I'll just need one lead to go to the hot side of the AC and I'll have a self contained unit that converts track power to 1.5vdc to operate the horn. No more worries about battery leaks!
As for heat dissipation, I'm not worried about it as the horn usage is not a consistent draw.
Shouldn't the LM317 have a filter capacitor on the pulsating DC input from the diode to protect it from possible voltage spikes?
Apparently bmoran4 is planning to use ~500uF caps.
you stated I'll just need one lead to go to the hot side of the AC and I'll have a self contained unit that converts track power to 1.5vac to operate the horn. No more worries about battery leaks!
the horn on older Lionel conventional engines operate on 1.5 vdc not 1.5 vac as you stated! but as Stan suggested installing a 500 uf capacitor will help to maintain the current draw on the horn! so the dc supply won't drop off as much!
Good luck
Alan
repair technician posted:you stated I'll just need one lead to go to the hot side of the AC and I'll have a self contained unit that converts track power to 1.5vac to operate the horn. No more worries about battery leaks!
the horn on older Lionel conventional engines operate on 1.5 vdc not 1.5 vac as you stated! but as Stan suggested installing a 500 uf capacitor will help to maintain the current draw on the horn! so the dc supply won't drop off as much!
Good luck
Alan
Alan, Yes, I meant 1.5vdc... I'll share photos and success once I get there!
Just out of curiosity, Does anyone know how much current those horns use? I don't have one or I would measure it myself.
Yes, the DC horn current would be interesting to know for the record ....hint hint to bmoran4! 
There have been several threads about battery replacement...many suggesting a higher voltage is better. Apparently even 9V DC works, though something closer to 3V DC seems more popular. For example:
https://ogrforum.com/t...replacement-question
https://ogrforum.com/t...06#42398304065657306
I think going more than about 3V is likely to quickly burn up the contacts in the horn.
stan2004 posted:the DC horn current would be interesting to know for the record ....hint hint to bmoran4!
My measurements show roughly 440-550ma on what appears to be an original horn in an PW 602 NW2 . I did end up using 470uF capacitor and a half wave single diode.
So a couple bucks in parts, some wire, solder, steady-hands, good eyesight...
bmoran4 posted:stan2004 posted:the DC horn current would be interesting to know for the record ....hint hint to bmoran4!
My measurements show roughly 440-550ma on what appears to be an original horn in an PW 602 NW2 . I did end up using 470uF capacitor and a half wave single diode.
Thank you for measuring that. I was expecting more current, to be honest, but I have no reason to doubt your measurement!
Yes, without turning this into a math quiz, I wanted to be sure 470uF was large enough (it is) to smooth out the DC side since only 1/2 of the AC cycle is available with the single diode method. That's about 3/4 Watts power and though I don't have one of these horns either, it seems that's enough power to make noise.
He could use the LM317P so that he could use the chassis as a heatsink, he'll likely want some head dissipation capability.
GunRunner, yes I could use the LM317P if I were to be mounting it to the frame However, I am not ordering new parts for this project - just using what I have around. Additionally, in my situation, I am putting all these components into the battery adapter as a self contained unit (and a lead connecting to the center rail AC line). In my testing, I don't have any heat dissipation issues. However, I suspect with a constant duty cycle, one would want a heatsink of some sort (frame, or otherwise).
I'd consider an open-air heatsink at least. You can, of course, simply use the insulating hardware and shim to mount the standard LM317T onto a heatsinking surface. I can assure you, if you are pushing 1/2 an amp through the LM317T for any amount of time, it'll get red hot and go into thermal shutdown rather quickly. I've used hundreds of these parts, so I have a smidgen of experience with them. 
gunrunnerjohn posted:I'd consider an open-air heatsink at least. You can, of course, simply use the insulating hardware and shim to mount the standard LM317T onto a heatsinking surface. I can assure you, if you are pushing 1/2 an amp through the LM317T for any amount of time, it'll get red hot and go into thermal shutdown rather quickly. I've used hundreds of these parts, so I have a smidgen of experience with them.
I'll have to go with GRJ on this one! It's not just the 1/2 Amp...but the whopping voltage drop from the capacitor voltage to 1.6V DC. So if blowing the horn with, say, 15V AC on the track, even with only half-wave rectification, that part is dissipating north of 5 Watts every time the button is pressed. Sure, it may be on for only seconds, but a TO-220 package without a heatsink heats up quite quickly.
Access to this requires an OGR Forum Supporting Membership
