CR400AF - Interior light overloading the power pack - SOLVED

[cteno4]

Hopefully a lot of leds at very low amperage to get an even fill without blasting the car out. But could explain the amperage draw if they start drawing a lot of amps at higher throttle settings.

 

jeff

[SwallowAngel]

The side lights do light up, just like @disturbman mentioned. I have a light set for my 25B coaches from the same manufacturer and all LED's light up. They really are a nice feature, since it nicely diffuses the light without an actual diffuser. I did initially order a set for myself, but I decided to cancel my order (budgetary reasons 😅).

I don't actually know how much amperage the set for my 25B's draw but my 4-car set behaves as you'd expect with the only "odd behavior" being the first 2-3 seconds of slow running because of the ginormous capacitors that need to charge up first.

 

[cteno4]

Thanks @SwallowAngel for the info. Im always the fan of more leds used very low as opposed to one or a few mini suns!

 

38 is pretty massive though, hopefully all set to very low current. So the coin cells are super caps. Ive only seen one other board using super caps, but they were very pricy, super caps are not cheap and probably more circuitry needed.

 

Any idea of what the two switches do? My guess would be to turn off and/or dim the 2 sets of lights. 

 

From the components on the board it looks to maybe use a constant current chip and then the leds in parallel and with different sized leds in the 2 sets there would need to be separate circuits for each. With those usually they pop on at around 4v then tend to stay the same brightness even with increasing voltage. Also nifty as you can change the output current by just changing the resistor value across two of the chip’s pins.

 

jeff

[SwallowAngel]

1 hour ago, cteno4 said:

Any idea of what the two switches do? My guess would be to turn off and/or dim the 2 sets of lights. 

My 25B boards actually have those too, but tbh I feel a tad too stupid to exactly know what they do 😅

 

IIRC some combinations just turned them off and some just turned them on at a set brightness. I assume the different combinations allow for different capacitor setups so that the manufacturer can reuse the same board for different trains. Unlike Lessingen's board mine have one big fat electrolytic capacitor that plugs into the board via pins. But the switches shouldn't do much besides turning the lights on and off

[cteno4]

Ok thanks, just curious, there are a number of ways to do these lighting boards and always curious the outcomes. I guess they went to the coin cell super caps over electrolytic. electrolytic are cheap and higher voltage, but way bigger per charge. I found the coin super caps a few years ago when looking at super caps. Biggest issue is they are usually lower voltages so you need to have a voltage regulator in your circuit to use them downstream of that.

 

Im guessing the switch is to turn one or both sets on or off or have a high/low setting that probably just sets a different resistance on the current regulator chip to set the amperage output to the leds.

 

jeff

[Lessigen]

Hey all, I got proper alligator clip leads and was able to test it. I ran two tests for each car, one where I peaked the throttle (cranked it to max quickly), and another where I "averaged" it (half throttle). I did another two sets for an increasing number of cars. Here are my findings from today's test

• Each of the cars pulls a max of ~280 mA each peak, and around ~223 mA on average
• The stall test on the motor + interior lighting pulls a peak of 610 mA, and 320 mA on average
  • Without lighting, the peak is 370 mA
• The motor with lighting runs 330 mA peak, and 300 mA average
  • Without lighting, this goes down to 40-60 mA for the motor only
• None of the cars seems to be short circuiting, the lighting just seems very current-hungry
• In the increasing number of cars test, 1-3 cars seems to be OK in terms of peak amperage. 6 cars (without motor) seems to be the maximum if I slowly bring up the throttle, and that only allows me to go to 3/4 throttle.
  • 4 cars or more at peak will cause the fuse in the Power Pack to trip. Doesn't matter if has the motor or not.
  • For numbers: 2 cars peaked at 580 mA/averaged 390 mA, 3 cars at 820 mA/averaged 500 mA, 4 cars (incl. motor) at peak 1.4A/averaged 620 mA. Any amount of cars above this trips the fuse when peaking, or (probably) warms it up if ran over a few minutes.
• The power pack is designed for 1.2A, but will trip the fuse at about 1.75A.

In conclusion, the lighting boards just seem too current-hungry for my power pack to run all 8 of them at once. I could maybe run 4 of them if I really inch the throttle forward so that the antiflicker caps charge up fully. What I believe is happening is that the spike (and/or overall provisioning) in current from the lighting board and the antiflicker caps causes the Power Pack fuse to trip because it shoots way past 1.75A. Me having it at a throttle point that doesn't cause it the fuse to trip means it's likely over the 1.2A design spec of the transformer and is why it starts feeling warm and smelling "magic smoke-y". I don't think the lighting boards are defective - unless they all came from the same bad batch - but they really suck up power. Going to reach out to Jacky to see what he can suggest or propose.

 

Edited Friday at 09:47 PM by Lessigen

[cteno4]

Wow thanks Lessigen! Yeah it just appears that the 38 leds are running at about 5-7ma each [sure some is sucked up by the power regulation circuit]. Thats around where seems like an awful large amount of light for a train car! Cars with only a few leds running at 10-15ma can make a car blaze. 38 at 5ma is a lot of light, unless they have something in the circuit that is leaking current thru. Just seemes odd. Usually i cut leds down to less than 5ma and usually below 2ma for structures, so im just wondering where all that current is going. Thats like 10 leds running full blast which would make a car glow like a nuclear meltdown.

 

similar results with the simple n600 throttle?

 

just to be thorough the car with out the lighting board didn’t draw any current, correct? I would not expect it to but always good to touch all the bases.

 

But yeah it makes sense why the the breaker is tripping. With those throttles you will just need to live without the lighting boards. Never want that magic smoke smell, it means something is really unhappily hot and usually smelled when something goes pop. May be the breaker as im guessing it uses the thermal breakers and i sort of remember the ones we had on the old layout smelling some when they tripped.

 

Motor itself seems normal

 

Hard to say with the lights stopping much increase at half throttle, sometimes when things get bright you cant notice any more changes or it could be using a current limiting chip which would pop on around 3-4v and max out at around 5-6v and not get brighter past that. Does the current peak out at half throttle for a lighting board or just keep going up? If it maxes out half way up thats a good sign it’s using a current limiting chip.

 

Is Chanming selling a high amp power supply? Usually N scale throttles max out at around 2a

 

jeff

[Lessigen]

Hey Jeff, yeah I just tried with the TOMIX N-600 and have very similar results. No flicker because of what I assume to be no PWM.

The cars that don't have any other electrical draw aside from the lighting board (i.e. not the motor or end cars) read 0.00A when I tested them without the lighting board.

The current does peak out halfway - by how I'm understanding your question, you mean that the current goes down after hitting a maximum amount of current (the aforementioned 280-610 mA depending on the car), which it does because of I believe that antiflicker cap probably filling up alongside the current limiting chip.

 

I'm not sure about Changming selling a high-amp power pack, perhaps @disturbman has some insight on that. Though that said, mine only tops out at 1.2A, getting a 2A one would probably yield very positive results thought it might still complain if I ramp it up very quickly. I was looking at an KATO 22-017 (outputs 15V2.4A, but as I understand it should do 12V too) or a Hitec HTC-100 (12V2A) to upgrade to. Of course, another option would be to get a less powerful lighting kit.

 

Edited Saturday at 01:06 AM by Lessigen

[Madsing]

There must be something wrong with that lighting kit. The power consumption of the lights should not be more than a few mA.

For comparison, I just tested the Popondetta 1501, its power consumption on the "low" setting is 4.5mA. On the "high" setting it is 17mA (but that one is too bright).

 

 

The side LEDs of the CR400AF look interesting, though. It's the first time I see that. I'd be interest to know if they really help the light look more even.

 

Marc

 

[cteno4]

Thanks, good info. 

 

On the n600 does the current level out and not increase or decrease after you hit half throttle? There it should be pure dc and the cap not have any effect once a current limiting circuit is fully powered. The cap should only fill in when there is a disruption in the track current.

 

still odd so much current drawn to make 10 leds blaze at 20ma ea which would just blast out a car, but it sounds like its not going nuclear on you in terms of light. My only thought is something odd going on with the super cap passing a lot of current. But im not an ee and my electronics training was decades ago!

 

Yeah most 2a and above throttles are more like 14-16v throttles for larger scales.

 

jeff

 

 

[cteno4]

Thanks @Madsing Yeah I agree, something like 5-10ma of led light should be the right amount to fill a car, its diffusing it well so its not coming in a series of spot lights that is the trick. With 38 leds you certainly spread the light out but would run the leds really dim and thus very low current. Even loosing some current to the circuitry, I would not expect that to be 250ma per board.

 

Wonder what they would do if the super cap was popped out the holder. I dont think it should harm the rest of the circuitry just loose the stabilization ability. Something seems to leaking a lot of current and I can’t believe they would have lighting boards drawing way more current than a motor car.

 

jeff

[disturbman]

5 hours ago, Lessigen said:

I'm not sure about Changming selling a high-amp power pack, perhaps @disturbman has some insight on that.

As far as I know, there no homegrown powerpacks or track system. Kunter made moves in that direction, but the track of choice is Unitrak.

@cteno4 these specific lights weren't made by CM, but by an interior light manufacturer. Either on their own, or contracted by Jacky/CNRailModels.

I'm also wondering about that cap. I have had interior lights with large caps (or perhaps a battery) and it took a rather long time to charge. @Lessigenhow long did you test a single line? I'm wondering if draw could have lessen after 2 or 5 minutes of usage.

[cteno4]

That level of current draw is just crazy and requiring a long charge time should have been a notice in any instructions. Even then its pretty crazy as most n scale throttles are like around 1-1.5a so not able to even charge a 8 car train safely. Something is off here.

 

jeff

[disturbman]

Yes. It's my last attempt at trying to see if it's a design issue or simply a bad batch.

Next step is anyway, asking Jacky for replacements/reimbursement.

[Lessigen]

I tried testing for about a minute for each car. The current draw pretty much started going down after 3-4 seconds from the initial 280 mA to around 200.

 

That said, I think I finally found a proper solution/root cause. There are two switches on each of the boards, one controls the big LEDs the other the small ones on the sides. It toggles between low and high, but if you finagle the switch to be in-between those two toggles it turns into a pseudo potentiometer and causes the LED array to output its equivalent of the Sun which causes the humongous 280 mA draw. Meanwhile, if I selected one of the two normal states on the switches it goes between like <10 and <30 mA.

 

Here are some pictures showing the difference between "high" and "the Sun" - the 4th car (far left) is the one with the unchanged "the Sun" setting and the rest are on "high." You can see the large difference that ~200 mA will do in this case.

 

I haven't tested this with a full 8-car train yet, but I imagine it'll be fine running on the Power Pack. 25 mA times 8 cars plus whatever for the head/tail lights and motor should be plenty in my case. I do wish the LEDs on high settings were a bit more powerful though without having to potentially melt the transformer in my power pack doing the "hack." I think in this case it's more a design issue with causing the LED output to go crazy if it isn't in one of the two positions, though perhaps that would be more warmly welcomed if I had a more beefy power pack or even a DCC-EX setup to take advantage of it.

 

 

Edited Saturday at 07:46 AM by Lessigen

[disturbman]

So all the light units were on that "Sun" setting? If that's the case, it could be an oversight. All the pictures I have seen from Jacky of a lighted CR400 were on the "High" setting.

HKTILC lights also have a middle, non-notched setting, which is an in between the "Sun" and "Low" setting and which fits models the best. Unfortunately, you can't toggle it properly due to the lack of notch.

Happy you found the solution.

[Lessigen]

Indeed, all of them were set like that. I had asked Jacky about adjusting the light and he mentioned that the boards were set to the "Sun" setting because you couldn't see the effect very well.

 

Thanks everyone for your insight.

Edited Saturday at 12:41 PM by Lessigen

[Lessigen]

By the way, here's the documentation for the lightbar. The brightness adjustment was set for the 1000 +/- 100 lux, the only difference being that the bottom switch was set between the + and - points.

[cteno4]

Ok that now makes sense! So with both small and big leds set to low its like 10ma and with them both set to high its 30ma. Then if you put it in the middle it’s the 280ma sun, correct? I would guess when set to the its engaging power to the high and low settings circuits and causing something weird to happen. If it is a constant current chip then maybe it’s an odd resistance on the current setting pin on the chip and setting the current to like 7ma to 38leds! But i would expect that to look even brighter than your sun setting picture, but photos are even worse to see effects. But led brightness is hard to really gauge once they start getting really bright, you can change the current some and not really notice any change in brightness the way you dan when they get set lower.

 

The second or two extra current pull is the super cap charging i would guess.

 

yes at 25ma per car should be no issue with the full train on your power supplies.

 

jeff