I'll put one in later and see what happens. Why did the resistor heat up and not the switch coils? I kept feeling the coils of that switch while I was working on my wiring and it was cool. Then I went under the layout on that side to put some wire "staples" on my work and.....felt heat under there without even touching the resistor.
It's all about the DC resistance and power dissipation if you're feeding the circuit DC. The coils are large enough that they'll dissipate more power without overheating than the puny 1W resistor.
The resistor getting hot is what I expected, I was really surprised that others are remaining cool, if the coils are energized through them, I can't imagine why they didn't overheat as well.
With the bigger resistor, do you anticipate some heating of the coils now? Also, when I measured the coil resistance in the switches and got 9.6 ohms. Can that be accurate? Another poster said 14 (from one of Bob Nelson's posts).
I got 24 ohms on an 1122, so who knows. 
The larger wattage resistor will make no difference in the circuit current, it'll just dissipate the heat from the power dropped in the resistor better. You're not changing the circuit current at all.
Ok, I put one of the bigger resistors in a switch and reconnected the light bulb to a nearby switch to compare. With the 10ohm resistor, the toggling was almost instantaneous. Now, with the bigger resistor, it's alot slower. In comparing to the light bulb switch....it's faster than that but maybe a half second or so. I guess bottom line is....it's bit of improvement over the light bulb and it's also cheaper to build (no bulbs or sockets to buy).
Well, you can lower the value of the resistor, it's the wattage that was in question. Try a 27 ohm 5W resistor, for instance.
John,
I'm going to order 8 of these if you think this one is ok. http://www.digikey.com/product...B-27R/27W-5-ND/18665
Since I have to experiment anyway, any other size you suggest I try?
Roger, I'd probably grab a few values so you can truly tinker with the response. Also, don't be afraid to consider a higher value for switches where you won't have two trains coming through very close together. The lower value resistors are only needed if you're going to be switching the switch quickly successive times.
Values from 10 ohms up to 100 would be nice to have. Perhaps 10, 27, 47, 68, and 100? Those are all standard values and should be readily available.
John,
Well.....I've already tried 50 and 10 (the resistor gets hot). Going to 68 will make the response slower than the 50.....no point in that. And 100....that will be slower than the bulb. So, I think I'll order the 27s and maybe 30 something if that exists. I don't want to go low (like the 10s) and run the risk of a fire under my layout. It's very easy to have a train sit on a switch. My layout isn't that long, so that's going to occur alot.
John,
Going to place an order for 27s, 33s and 47s. Does my reasoning make sense?
Chris....I hope all this experimenting will help you out. I think we need to find the proper balance between recharge time and safety (not burning out coils or electronics).
I'd probably do the 22, 33, and 47 in that case. Gives you a nice spread, the 27 and 33 won't be much different.
Alright.....ordered the 22,33,47s. I guess the only way I will know which ones are safe is to park a consist over a switch and see what happens. The 50s are not bad, but not much faster than the bulbs (which I suppose are safer because they bleed out the heat easier).
I'll post results when my order arrives.
Roger;
Just thought I would add that if you decide as you are testing that you need faster response time than the 22 ohm resistors give you, you can always combine the resistors you are ordering in parrallel as follows:
22 + 33 = 13 ohms
22 + 47 = 15 ohms
33 + 47 = 19 ohms
This gives you a few more options.
Remember they must be wired in parrallel to do this.
Rod
To all, thanks for adding to this topic. It's really helping me learn alot. And Roger, thanks for doing the actual field work. My only fear is that we may find out that the #53 bulb ends up being the best solution. But in this hobby, trying to push the envelope is what it's all about.
In the meantime, I'll start working on converting my switches to accessory power this weekend. And I've got a banjo signal to fix, and a Type V transformer that I'm working on.
Thanks again to all, and keep me posted!
Chris
Rod,
Thanks! I was wondering about doing that, but didn't know how to calculate that. What is the formula, by the way? I've been using the "bulb option" for 2 years now and thought I was happy with it until I tried the 10 ohm resistor and it was instantaneous in it's response. Little did I know what was going on under the layout, though. So, I'll have to let these trials sit with consists over them for awhile so I don't get any surprises. Still don't know why my switch coils aren't heating. When I switched over to DCS, I initially put 18V bulbs in those switches. Melted all the switch covers. I ended up finding new covers and replaced the bulbs with LEDs. Heat is the enemy of those 027s, so I don't want to melt my coils.
Roger
Roger
Chris,
A suggestion.....don't remove the bottoms of those switches. Cut a window over the spot you need. I can send you the measurements if you want to go that way.
Roger
Originally Posted by ROGER1:
Rod,
Thanks! I was wondering about doing that, but didn't know how to calculate that. What is the formula, by the way?
Roger, try this page: http://www.1728.org/resistrs.htm
Not only does it have the formula, it also has a handy calculator so you don't have to strain your brain. 
John,
Now THAT is VERY cool. And when it goes in a bookmarked thread.....perfect.
Roger
John,
I've got a question for you. My order from Digikey is delayed. Thought I'd have it Monday, but it won't arrive till Wednesday now. So....I've been experimenting a bit based on Rod's post and your chart for parallel resistors. I figured....let's see what happens when I hook the two 50s together (to get 25ohms). I ordered 22s, so I thought this would at least be close in performance to when I get those. Anyway, I alligator clipped them up and fired up the switch. It's speed performance was very nice. I would be happy with that. The heat is where my question lies. Again, the switch coils do not heat at all (at least as of 15 minutes so far with a consist over it). The two resistors are definitely warm, but not hot. But what do you think will happen when there's only one resistor there? I'm thinking it will be hotter than each of the two resistors (even if I connected a 25ohm resistor rather than the 22 that's coming). With two resistors, I'm thinking they split the heat better, but I could be wrong about that. What do you think?
Roger
And yet another scenario.......I reclipped to the light bulb and ran a 50 ohm resistor in parallel to the bulb. Nice snap! The resistor does get warm. I'm not sure about the resistance measurements, though. The bulb and socket by itself measured 11 ohms. That doesn't seem right (seems way low). The resistor and the bulb together were measuring 9 and change. That's what the chart says as well, but.....with the light bulb not being lit....not sure that's all correct. But it does provide another option.
Again, the power dissipation of the resistor will determine how warm they get. If you parallel two 1 watt 50 ohm resistors, you get a 2 watt 25 ohm resistor. If you're buying 5 or 10 watt resistors, they will have the same resistance, but they'll dissipate the heat better and not get as warm.
The reason the bulb and resistor measure low is you're only measuring the cold resistance of the bulb. As the bulb lights, the resistance goes up a bunch.
Example:
I have a 433 bulb, cold resistance is 7 ohms. I put 10 volts DC across it, and it draws 150ma. That would indicate it has a hot resistance of 67 ohms, quite a bit different than the cold resistance.
That's what I was thinking with the bulb. It wasn't lit. But....that may be an option.....hooking a resistor in parallel to the bulbs I already have in place.....nice snap. Close to what the 10 ohm resistor was giving (but without the heat). Any downsides to doing that?
The point of the resistor value is to not dissipate much power in the coils, but to charge the cap as fast as possible so it's ready for another switch movement. With the bulb, you get high resistance very quickly and slow the charge. Personally, I'd just go with fixed resistors and forget any "gimmick" installations, but that's just me.
Ok, that sounds like a plan to me.
John,
The resistors arrived and later I will be setting up some experimental trials. I've got a question for you. I want better snap (faster recharge than my bulbs) but I want it to be safe (no fires under the layout). Will these resistors hit a max temp after awhile or will they continue to get hotter as a consist sits on the switch? I'm leaning towards using two resistors in parallel because in my previous trials with the ones I had at the time, they didn't appear to get hotter than they were after about 20 minutes. But that's subjective and possibly dangerous.
Roger
And one more question.....10W vs. 5W At the same resistance, does that mean that the 10W can handle more heat?
The greater the wattage, the better they handle the heat. It's simply a matter of power dissipation, if you're dropping three watts in the resistor, a 5W will get pretty warm, but the 10W will run much cooler. Parallel resistors of equal values split the power load equally, unequal parallel resistors will dissipate whatever share of the current (and power) they're carrying.
Once you reach a thermal equilibrium of power dissipation vs. heat loss, the temperature will remain constant. Also, the resistors should be in free air, not pressed against anything flammable.
John,
Initial results.....nothing (singly) less than 50(and possibly even a single 50) is acceptable. 22,33,47.....all get too hot. Started doubling. 100/100 I think is too slow. 100/50 and 100/47 snap better than bulb and seem to be (so far) acceptably warm.
All of the ones from 22-47 are 5 watt (Digikey). My 100 and 50s are 10W. Going to try a 100/50....should be ok heatwise and possibly speed. Worried about heat more
Be sure to monitor the heat in the coils as well! If the resistors get hot, there's power being dissipated by the switch coils, don't forget that.
The 5 W resistors are a no go. Too hot. So, it's either going to be 100/50 or 100/100 (10W). The 100/100 is not that much different from the bulb setup. If the heat is acceptable in the 100/50, that will be the choice. What would you say is the limit of "hot/warm" when you touch the resistor with your fingers. That's what it's going to come down to.
Roger
The coils.....they never go above room temp. Bizarre. My fail safe mechanism for dealing with the switch issues was/is knife switches that cut the power to them. I put them on before I had the cap system to get rid of buzzing and also to stop some of my MTH engines from stopping on some of the switches. That would eliminate the heating problem, but I'd have to remember to do that if I walk away from the layout for awhile. I'd rather not do that. So far, the 100/50 combo is not getting beyond warm (the smaller resistor). I'm going to go to RS now to get more of those resistors. I'll check the heat when I get back and that should tell the tale.
If you can comfortably hold your fingers on the resistors, they're fine. For the coils, I'd want them to stay pretty much room temperature, since internal heat could be greater.
Well, here's the problem. I can hold them in my hand. Definitely warm, but not where I have to let go of it. The only other choice I have is 75 ohm. Raises the overall resistance from 33 to 42. It would slow it down. By how much I don't know. It would probably be safer.
Well, as long as they're in free air, and they don't get any hotter sitting there for 10-15 minutes, that shouldn't be a problem.
I don't trust them, John. They are kind of on the edge of comfortable. So.....I'd have to get more 100s if I was to do 100/100 for all. So I think I'll order 8 75s and do 100/75. It will be less warm than the 100/50 and have faster snap than the 100/100.
Roger,
Have you tried a triple combo? Per the calculator 100-100-100 would yield 33 ohms. I finally got my accessory transformer fixed yesterday (new rollers, cord were easy, readjusting the circuit breaker was tedious). Looking forward to hearing back.
Hi Chris,
I haven't forgotten about you. The major problem here is the waiting for components to arrive. As you can see above, I finally got my Digikey resistors and they didn't work out. Singles are not going to be the way to go (unless we use a monster resistor that can handle the heat). As John suggested, he thinks the heat in the 100/50 combo would probably be ok. I respect John's vast knowledge of electronics, but what I was feeling with my fingers under there.....I just don't feel comfortable trusting it. It was barely comfortable to the touch. So.....I had previously tried 100/100.....acceptable heat, speed only slightly faster than the bulb. Yesterday, I ordered 10W 75 ohm resistors and they are to arrive tomorrow. They will work. In combo with the 100, it will be faster than the 100/100 and to me, safer than the 100/50. I will let you know as soon as I hook one up tomorrow. Last night, I tried a 50 in combo with the bulb setup that's still under there. Very nice! John frowned on that option, but I have to say, it works nicely. Last night, I also ran numbers through the calculator for "triples". The 100/50 is also 33. Nice snap, too much heat. I'm sure the triple would be good for heat. And I'm sure there's other resistor combos that would work, but I've already invested quite a bit here with these experiments and I can be satisfied with one I've tried or tomorrow's 100/75 combo.
I'll be back to you when I get the one I want. Meantime, tap into a switch and try the fixed voltage wiring.
Roger
Well, the resistors would be rated to run at least at 85C (185F), which is a lot hotter than you'd want to hold onto. Even the cheapest power resistors would have a 70C (158F) rating. If you can keep your fingers on a power resistor for 15 seconds and not come away with a burn, it's not too hot. They're designed to dissipate the power, and they usually expect them to get pretty warm near the limits. Obviously, I'm talking about after they have reached a thermal equilibrium, if they ramp up to that hot right away, that's different.
John,
It was very close to being untouchable....especially the leads coming out. However....I just went to the RadioShack website to see if there were any specs on them. I figured...I've got a mix of resistors....RS and Digikey. The RS ones are probably cheap. Well.....turns out they are rated to 275C. That's 525F. Apparently they are more resistant than we thought. I have to say.....when I touched the smaller ones.....I thought I had "branded" myself. I expected to see a band pattern.
Here's another question for you John. Small resistor vs. larger resistor. As far as the COIL windings are concerned. Isn't there a lesser chance of them taking heat with a larger resistor rather them in tandem with smaller ones? I keep waiting for the other shoe to drop with the coils. Keep expecting to see switch housings melting. But, they've been staying at room temp.
