I just installed a KA1 keep-alive module in a 1st-release Proto 2000 0-8-0 switcher. Does anyone know how long it takes for the caps to power up in the module?
This module has enough capacitance(?) for a 2-5 sec. interuption in power. With the power shut off and the front and rear LED headlights left on, it takes ~20 sec. for the LEDs to go out. The TCS KA1 literature that came with the module doesn't have any capacity ratings listed at all.
Thanks for the input...
Tom
https://tstage9.wixsite.com/nyc-modeling
Time...It marches on...without ever turning around to see if anyone is even keeping in step.
It should only take a few seconds to charge up the module once power is on.
How long it can power the decoder, you'd have to know the value of the capacitors used, and the current draw of the loco when in motion with the sound going. Knowing the capacitor value and voltage, you can calculate how much energy is stored in it, and knowing the power draw of the loco, you know how much power it needs. Sort of like a battery - a 1 amp-hour battery can supply 1 amp for 1 hour, or 1/2 amp for 2 hours, or .1 amps for 10 hours...
Using the current display of the PowerCanb you might be able to see how long it takes to charge the capacitor. Take the running loco off the track so it discharges the capacitor, then set the display to show the power, and put the loco on the track. The current should be a bit higher than usual and then gradually drop down to whatever it used to be for the idle loco. That is, if the PowerCab current display is sensitive enough to see the difference. It may charge up too fast for you to see much of anything.
You can guestimate based ont he headlights, if it keeps the headlight lit for 20 seconds, and we assume a typical white LED and 1K resistor, that's 9ma for 20 seconds, but there is the unknown draw of the decoder board that needs to be added. A high guess would be 100ma - so if it can supply 100ma for 20 seconds it can come close to providing 1 amp for 2 seconds - which sounds about right if it can keep the loco moving for 2-5 seconds.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
tstageThe TCS KA1 literature that came with the module doesn't have any capacity ratings listed at all.
The following site shows the capacity (and dimensions) of many available capacitors. I think it was updated in 2016 IIRC. Note 2 suggests that the "joules" column is the most consistent way to compare them. I keep the KA1 and KA2 and Current Keeper on hand and so I can choose based on room available, etc.
EDIT: Here's the link I forgot to include:
https://sites.google.com/site/markgurries/home/decoders/keep-alive-compatibility
I also recently installed a KA1 in a P2K 0-8-0. Overall, the KA1 (with about 20% storage of the KA2) solved my problem of minor sound hiccup interruptions on normally clean track. I let it charge for a couple minutes and with the headlight off it ran about 1-1/2 seconds when the power was shut off. Not great but sufficient. When similarly charged and the headlight on but not moving, the headlight continued for about 1-1/2 seconds. My headlight is a 3mm LED with a 1K resistor, if that matters. Not sure why our results are so different.
Paul
Modeling HO with a transition era UP bent
Paul,
The built-in resistors on the LL8-LED decoder are probably 1K and the replacement LEDs were each 3mm "warm". When I ran the 0-8-0 without lights and the power shut off, it continued for ~5 sec. before stopping.
Does it have anything to do with yours being a sound decoder and mine a motor-only decoder? Does a sound decoder draw more current than a motor-only decoder - even when the sound is turned off?
tstage Paul, The built-in resistors on the LL8-LED decoder are probably 1K and the replacement LEDs were each 3mm "warm". When I ran the 0-8-0 without lights and the power shut off, it continued for ~5 sec. before stopping. Does it have anything to do with yours being a sound decoder and mine a motor-only decoder? Does a sound decoder draw more current than a motor-only decoder - even when the sound is turned off? Tom
Mine has a LokSound Select micro (3/4 amp) 6Aux version (not board replacement) hard wired in so indeed we're talking different decoders as well as mine is sound. I wasn't certain I experimented with the sound off so repeated the test with the sound off. With the headlight on (not moving) and with the loco at 1/2 (adjusted) top speed I verified that I do get about 1-1/2 seconds for each of those tests.
I will tend to put the KA2s or Current Keepers into future installs where space is not an issue. I don't recall them costing alot more than the KA1. So if I'm bothering with this on select locos (probably 10%) I will usually go for the higher storage capability.
Sound decoders will draw more, even with the sound turned down, the processor is usually significantly more powerful than the ones used in motor only decoders, although it shouldn't be a HUGE difference - a few 10s of milliamps maybe. I suppose it could be worse - something else for me to test out with one of my better meters. But it doesn't take a huge difference in power draw to make a 1-2 second difference is the time a capacitor can power the system.
Joules is a good unit to start with. 1 amp-hour-volt is 3600 joules, so if the storage capacity of the capacitor is 54000 joules, at 15 volts it will supply 1 amp-hour. (Joules/3600)/volts = amp-hours. So say 1000 joules at 15V, is .0185 amp-hours, or 18.5maH - it could provide 18.5ma for 1 hour. The KA1 is listed int he table as 4.13 joules. At 15V we get .00008 amp hours. Or .288 amp-seconds, so 1 second of life at a 288ma load. For 5 seconds the load would have to be under 60ma. That seems low based on the rating. Maybe my math is wrong. The KA2 has 4.4x the joules, but also 1.5V lower voltage, so the total energy is less than 4x the KA1 - which based on the time estimates is exactly right. 18.23 joules at 13.5V is .00038 amp hours or 1.368 amp seconds, so it could go 1.368 amps for 1 second, or .274ma for 5 seconds, less than 60ma, it could run for 20 seconds.
All depend son if I di dmy math right, AND if the joule rating in Marcus's table is correct. It appears to be: http://www.electronics2000.co.uk/calc/capacitor-charge-calculator.php
On a lighter note, I am reminded of physics teacher Mr. Gralla (a definite character) in my high school (in the prior millenium). I believe it was he (vs. Penn State profs) that insisted that our joules are pronounced as "jowls" (vs. "jewels").
I don't care enough to google it but rather prefer to enjoy the thoughts. I did have to look up the plural of jowl (and ask my supervisor how to spell jewels due to a gaseous brain event).
peahrens due to a gaseous brain event...
due to a gaseous brain event...
I JUST had one of those episodes over on another thread in this forum. Welcome to to the club, Paul!
I was taking a video of the motor-control of the TCS LL8 and tested the KA1 keep-alive module again. I was incorrect about the LED headlights lasting 20 sec. after the power was shut off...
....it's actually 36 sec...and that's with both front headlight on & rear headlight dimmed (Rule 17). At speed step 10 current output for the motor is still in the 5-6 sec. range before it comes to a dead halt. I can live with those numbers.
I measured the current draw of one of my Stewart Baldwins, with the Canon motor, at like 25ma free running on DC before I put a decoder in it. The LED was probably half of it. Those locos will coast a considerable distance when the power is shut off - AND the headlight stays on due to the BEMF from the motor. I can only imagine if a true supercap keep alive like the KA1 would fit in there, how long it would run after the power dropped.
Proof positive that the rule of thumb of .5 amp per HO loco is WAY conservative unless you are running vintage equipment.
Interesting info on the KA/current keeper series in the link. I have a bunch of KAs and current keepers installed. The layout takes about 45 seconds to die after cutting power…. The econcamis seem to have about 45 seconds of idle time left after the power is cut.
see stuff at: the Willoughby Line Site