Author Topic: Programming a decoder  (Read 9480 times)

0 Members and 1 Guest are viewing this topic.

Chris333

  • Crew
  • *
  • Posts: 18396
  • Respect: +5668
Re: Programming a decoder
« Reply #60 on: March 14, 2018, 04:57:41 PM »
0
Thanks, I think I'll just leave 29 alone. Instead of setting 29 why aren't there just 7 other CVs to set? And there is a button on the controller to change the speed steps on the fly.

My speaker box showed up today so I re-wired everything back up in the Heisler. The current keeper is not working in that loco either.

How did the KA3 stop working? It is just 4 caps with a few small bits.
« Last Edit: March 14, 2018, 05:00:13 PM by Chris333 »

DKS

  • The Pitt
  • Crew
  • *
  • Posts: 13424
  • Respect: +7026
Re: Programming a decoder
« Reply #61 on: March 14, 2018, 05:04:53 PM »
0
I probably did a stupid thing.

No comment... :trollface:

peteski

  • Crew
  • *
  • Posts: 32958
  • Gender: Male
  • Honorary Resident Curmudgeon
  • Respect: +5343
    • Coming (not so) soon...
Re: Programming a decoder
« Reply #62 on: March 14, 2018, 05:05:45 PM »
0
Thanks, I think I'll just leave 29 alone.

My speaker box showed up today so I re-wired everything back up in the Heisler. The current keeper is not working in that loco either.

How did the KA3 stop working? It is just 4 caps with a few small bits.

If the KA3 is defective then the most likely failed components are those "small bits" (diodes).  I  suppose SuperCaps can fail too, but I would expect them to short out (not become open circuit).  Maybe your track voltage is too high (but if you are using the stock NCE Power Cab wall-wart then the voltage should be fine).

You can give the KA3 a quick test.  Take a 9V battery and hold the blue wire to the positive battery terminal, and the other wire to the negative terminal. Polarity is important. Hold it like that for about 30 seconds (to allow the KA3 to charge up). Then take a 12V light bulb (not LED) or a small N scale size motor and hook it up to the KA3 terminals. Polarity doesn't matter. The KA3 should be able to power those devices for several seconds.
. . . 42 . . .

Chris333

  • Crew
  • *
  • Posts: 18396
  • Respect: +5668
Re: Programming a decoder
« Reply #63 on: March 15, 2018, 03:56:56 AM »
0
So this is what the KA3 looks like:


I don't have light bulbs so I tested with a meter. I get 1.5V from the KA3. Right after the 9V charge and 10 min later same 1.5V.

Any clue what is wrong with it. Do I need the components other than the caps?

Chris333

  • Crew
  • *
  • Posts: 18396
  • Respect: +5668
Re: Programming a decoder
« Reply #64 on: March 15, 2018, 04:46:30 AM »
0
I discharged the cap down to 0.5V with a small motor. Wired it back to the decoder, didn't work as needed. Removed it and now got 3V out of it. Was able to power a LED with it.

peteski

  • Crew
  • *
  • Posts: 32958
  • Gender: Male
  • Honorary Resident Curmudgeon
  • Respect: +5343
    • Coming (not so) soon...
Re: Programming a decoder
« Reply #65 on: March 15, 2018, 05:32:41 AM »
0
That KA3 is sure acting strangely.  I have one of these TCS KA units at home but I haven't done much experimenting with it. I think mine has 5 caps.

The "151" component is a 150 ohm resistor which limits the charging current.  The 3-lead component connected to that resistor is probably a CMOS transistor which along with the tiny resistor and the other black component (Zener Diode?) are probably some sort of voltage-limiting circuit.  What is the voltage rating on those caps?  That circuit will limit the voltage to less than the cap voltage times 4.  Then the other 3-lead component (SH1?) is probably just a Shottky diode through which the KA3 supplied power back to the decoder when the track voltage is not there.
. . . 42 . . .

Chris333

  • Crew
  • *
  • Posts: 18396
  • Respect: +5668
Re: Programming a decoder
« Reply #66 on: March 15, 2018, 05:37:01 AM »
0
They are 2.7 volts each.  The thing was $25 bucks  :|

jdcolombo

  • Crew
  • *
  • Posts: 2265
  • Respect: +973
Re: Programming a decoder
« Reply #67 on: March 15, 2018, 09:09:41 AM »
0
FWIW, I did Peteski's test on a KA-3 that I had lying around.  Charged it on a 9v battery for 1 minute, then attached my Fluke DMM to read DC volts.

Got 7.5, with very slow decay (about 1/100th of a volt per second), which I would expect without a big current-drawing load.

So maybe you do have a faulty KA-3, though I've used at least a dozen of these in the past and never encountered a failure . . .

John C.

Lemosteam

  • Crew
  • *
  • Posts: 5919
  • Gender: Male
  • PRR, The Standard Railroad of my World
  • Respect: +3668
    • Designer at Keystone Details
Re: Programming a decoder
« Reply #68 on: March 15, 2018, 09:25:49 AM »
0
Never used one, but I thought I read somewhere that if the input wires get reversed accidentally, you can mess these things up. I think that's why the Econami's have a plug.

Chris333

  • Crew
  • *
  • Posts: 18396
  • Respect: +5668
Re: Programming a decoder
« Reply #69 on: March 15, 2018, 02:17:26 PM »
0
I know I hooked it up backwards in the 2-4-4-2 because it wasn't working the correct way.  I was testing it with just the tender sitting there and someone on Facebook told me I needed to wire the caps to the decoder and the sound board. So maybe the cap was working for the motor, but I didn't have the motor plugged in. But since the sound board plugs right into the decoder I doubt I needed to wire both.

peteski

  • Crew
  • *
  • Posts: 32958
  • Gender: Male
  • Honorary Resident Curmudgeon
  • Respect: +5343
    • Coming (not so) soon...
Re: Programming a decoder
« Reply #70 on: March 15, 2018, 02:26:25 PM »
0
Hooking it backwards will without a doubt damage the caps! Sorry Chris, it looks like you are SOL.   If you are handy soldering, you can replace the caps for a fraction of the new unit's cost. 

This is interesting.  With just four 2.7V caps (which is a standard voltage for Super Caps), the charging circuit will only charge the caps to less than 10.8V.  Then it will only deliver around 10V back to the decoder when needed.  The model will likely exhibit drop in speed when running off of the KA3.

I have a KA2 unit. It has five 2.7V caps, so it can handle around 13V (wich is closer to the typical DCC voltage).
. . . 42 . . .

jdcolombo

  • Crew
  • *
  • Posts: 2265
  • Respect: +973
Re: Programming a decoder
« Reply #71 on: March 15, 2018, 02:42:19 PM »
0

This is interesting.  With just four 2.7V caps (which is a standard voltage for Super Caps), the charging circuit will only charge the caps to less than 10.8V.  Then it will only deliver around 10V back to the decoder when needed.  The model will likely exhibit drop in speed when running off of the KA3.


But you'd only see a drop in speed if the motor was consuming more than 10v, right?  I suspect that most keep-alives come into play only when the engine is moving at slow or moderate speed - for example, to bridge a dead frog for a second on a slow-speed switching move.  In those cases, the motor probably isn't pulling anywhere close to 10v - maybe more like 4-6v.  So in those circumstances, you probably wouldn't notice the lower voltage.

Or is there something I'm missing?

John C.

Chris333

  • Crew
  • *
  • Posts: 18396
  • Respect: +5668
Re: Programming a decoder
« Reply #72 on: March 15, 2018, 02:43:07 PM »
0
I'll try and find new caps.

Pretty sure their site says the KA3 replaces the KA2. As in they don't make the KA2 anymore.

peteski

  • Crew
  • *
  • Posts: 32958
  • Gender: Male
  • Honorary Resident Curmudgeon
  • Respect: +5343
    • Coming (not so) soon...
Re: Programming a decoder
« Reply #73 on: March 15, 2018, 02:54:11 PM »
0
But you'd only see a drop in speed if the motor was consuming more than 10v, right?  I suspect that most keep-alives come into play only when the engine is moving at slow or moderate speed - for example, to bridge a dead frog for a second on a slow-speed switching move.  In those cases, the motor probably isn't pulling anywhere close to 10v - maybe more like 4-6v.  So in those circumstances, you probably wouldn't notice the lower voltage.

Or is there something I'm missing?

John C.

Remember, decoders use a PWM circuit (Not an analog voltage) to control the average voltage to the motor. The pulses are always at the full internal DC voltage in the decoder. So of the full voltage is lower the average voltage at the motor (derived from the PWM pulses)will also be lower.

I didn't know that they discontinued the KA2. Interesting . . .
Chris, what capacitance are those caps?  1F?
. . . 42 . . .

Chris333

  • Crew
  • *
  • Posts: 18396
  • Respect: +5668
Re: Programming a decoder
« Reply #74 on: March 15, 2018, 02:56:31 PM »
0
Yes 1F. Mine are 6.5x10mm and all the one I find online are 8x12mm.