Post by: KPack on January 30, 2017, 10:05:58 AM
Reports are that all stalling has ceased and locomotives run buttery smooth through all turnouts.
The screwdriver marks the location of the bridge rectifier:
(http://i279.photobucket.com/albums/kk131/freewake_2008/Railpro/IMG_1670_zpsem9wcx89.jpg)
Note the size of the bridge rectifier and the KA. This in a Bachmann RS3 for reference.
(http://i279.photobucket.com/albums/kk131/freewake_2008/Railpro/IMG_1671_zpsiw3vpiya.jpg)
Schematic:
(http://i279.photobucket.com/albums/kk131/freewake_2008/Railpro/IMG_1672_zps4psaghjb.jpg)
The bridge rectifier used here:
http://www.digikey.com/products/en?keywords=DF04SCT (http://www.digikey.com/products/en?keywords=DF04SCT)
At about 40 cents per rectifier, these seem to be a good option. I'll be ordering at least ten, as well some more tablet speakers for sound installs.
The only question I have, and I'll see if Rob has the answer, is did they just install the bridge rectifier on the red and black track power wires. I assume we just arbitrarily assign one of the wires to be positive and the other negative after the bridge rectifier....black for negative and red for positive for convenience. Am I correct in this assumption?
Post by: SWA737 on January 30, 2017, 10:14:18 AM
Rob
Post by: Alan on January 30, 2017, 10:18:41 AM
eBay is great for buying these. $0.10 each or less if you don't mind a 30 day wait from China.
http://www.ebay.com/itm/10PCS-2W10-2A-Bridge-Diode-Rectifier-NEW-/401167556726?hash=item5d67732076:g:XEcAAOSwU-pXqO95 (http://www.ebay.com/itm/10PCS-2W10-2A-Bridge-Diode-Rectifier-NEW-/401167556726?hash=item5d67732076:g:XEcAAOSwU-pXqO95)
Post by: SWA737 on January 30, 2017, 11:10:26 AM
Post by: SWA737 on January 30, 2017, 11:12:22 AM
Rob
Post by: Alan on January 30, 2017, 11:15:40 AM
Post by: KPack on January 30, 2017, 11:17:54 AM
-Kevin
Post by: Alan on January 30, 2017, 11:24:37 AM
A case could be made that if one loco has keep alive capacitors then all your locomotives should also. It would be interesting to study the MUing function between KA'ed and non-KA'ed. I am sure it will still work but am curious if there is an observable deterioration in function.
Post by: Alan on January 30, 2017, 11:41:04 AM
Quote
As long as the +/- on the KA is the same as the polarity on the exit of the rectifier.
Correct.
The keep alive deal is nothing more than fundamental power supply smoothing. At the heart of virtually every power supply out there. Perhaps you have seen an illustration like this:
The capacitor supplies current to fill in during the AC polarity changeover. This smooths the output voltage. The KA application functions the same except the periods of no supplied current (loss of connection to the track) are unpredictable in frequency or duration so a larger capacitor is needed to make sure there is sufficient current available to cover the longest interruption at maximum load. Electrolytic is a type of capacitor that can store a lot of energy in a small space. They are used in power supplies and KA for this reason. Electrolytic capacitors, with few odd exceptions, are polarized. Hence the + and - markings and why the KA has color coded wire leads. + must always be connected to + with an electrolytic capacitor. Hooked up backwards they explode. Sometimes violently.
Post by: TwinStar on January 30, 2017, 11:43:06 PM
Thanks!
Post by: Dean on February 02, 2017, 02:21:32 PM
Digitrax has a keep alive module that is much cheaper than the TCS module. Will it work using the same circuit?
Thanks
Post by: Alan on February 02, 2017, 04:11:20 PM
Keep alives use supercapacitors, sometimes called ultracapacitors, which are a type of super high density electrolytic capacitor. Large value capacitors, especially electrolytic, are slow changing devices on electrical timescales. They are sluggish. Because of this they do not smooth out high frequencies very well if at all. The DCC square wave is somewhere around 10kHz if I recall correctly. That may very well be above the effective frequency range of the capacitors used in these two commercial keep alive products.
Common in electrical circuits is a combination of a large capacitor and a small capacitor together to filter electrical noise on a supply line. The big slow capacitor supplies large current during the long dips while the little fast capacitor supplies current during the short quick dips. i.e. the big cap takes care of low frequencies, the little cap takes care of high frequencies. Here is a practical example used in my detection boards:
In this circuit the 555 timer chip produces both low and high frequency noise on the supply line every time it changes state. Equate that to losing rail contact (low freq) and DCC signal (high freq). The big electrolytic on the board carries the load for the long slow changing dips. The little ceramic cap next to it reacts fast carrying the load during the short fast dips. The result is noise-free supply line. This big cap little cap combination is extensively used in electronics. If the little cap were not present then the the high frequency noise would still be on the supply line. I suspect the keep alives work the same way.
Post by: TwinStar on February 03, 2017, 02:48:50 AM
Post by: William Brillinger on February 03, 2017, 07:41:04 AM
Post by: William Brillinger on February 03, 2017, 07:41:32 AM
https://sites.google.com/site/markgurries/home/decoders/keep-alive-compatibility
Post by: TwinStar on February 03, 2017, 11:20:36 AM
Post by: TwinStar on February 03, 2017, 01:19:50 PM
Post by: CPRail on February 04, 2017, 06:19:00 PM
If that's the case, I'll be ordering the rectifiers ASAP!!
Post by: William Brillinger on February 04, 2017, 06:25:40 PM
I proposed this idea a long time ago but never acted on it.
http://model-railroad-hobbyist.com/node/22101
I think I'm ready to proceed too. I'm done waiting for the Ring engineering version of this tech.
Post by: Alan on February 04, 2017, 06:32:40 PM
Post by: William Brillinger on February 04, 2017, 06:45:04 PM
...But I am ready to order some KA3's and KA4's for myself. I already have the Diode Bridges.
Post by: TwinStar on February 05, 2017, 04:00:14 PM
I got tired of waiting too. My bridge rectifiers from DigiKey arrived before I left for work and ebay says I have a KA3 and KA4 waiting in my mailbox for me. I'll post some pics in the install forum this week.
Jacob
Post by: G8B4Life on February 06, 2017, 06:30:48 AM
On Marcus' page, about 2/3 way down under "More details of the Stay/Keep Alives." is a pic that shows the TCS KA-1 and KA-2 without the shrink wrap. Further down you can see how he cut a KA-2 in half for a tight installation.
http://www.members.optusnet.com.au/mainnorth/alive.htm (http://www.members.optusnet.com.au/mainnorth/alive.htm)
There is a pdf file on this site, which on page 8 has a basic schematic of a stay/keep alive. Note the diodes as well as the resistor. From memory TCS has 150 ohm resistors.
https://dccconcepts.com/manual/dcc-advice-12-exerpt-b-decoder-standard-connectors-and-stay-alives/ (https://dccconcepts.com/manual/dcc-advice-12-exerpt-b-decoder-standard-connectors-and-stay-alives/)
- Tim
Post by: William Brillinger on February 08, 2017, 08:27:38 AM
Using a KA with the LM-3 and LM-3S
Here is a Wiring Diagram for the LM-3S
And a few quotes from Ring engineering during previous discussions I had with Tim:
"we changed the name of the wire on the 6 pin from 'Common' to 'Negative'. This change is because many DCC people call the blue wire common (although it is a positive wire)"
"The LM-3 and LM-3S have a common pin in place of input 2. Therefore they have one input instead of two. The common pin will allow better compatibility with keep alive and DCC."
"However, since the LM-3 has a common pin you may be able to use one of the other manufacture keep alive's with the LM-3."
And finally the mandatory but covering...
"However, since the LM-3 has a common pin you may be able to use one of the other manufacture keep alive's with the LM-3. Warning: we do not recommend using any other manufactures keep alive's with our LM's because in the worst case it may cause a fire. Connecting other manufacturers product to our LM-3's may damage our LM-3 or the other product. Even if one module worked today with our LM-3, we could not guarantee that other manufactures will not change their product such that it is not compatible with our products tomorrow. This would be: try at your own risk."
Theoretically you should be able to connect a TCS KA to the Common and Negative wires (ON THE LM-3 or LM-3S ONLY) and you wouldn't need the Bridge Rectifier.
Has anyone tried this yet?
DO NOT DO THIS - IT'S A BAD IDEA...
Post by: Alan on February 08, 2017, 09:27:53 AM
Nope, sorry. Until Ring offers specific assurances regarding the LM internal protection circuitry I would stay with the external bridge. This way, should a problem arise, the potential loss is a 25 cent bridge, not a $70 module.
Post by: William Brillinger on February 08, 2017, 09:36:21 AM
Post by: KPack on February 08, 2017, 10:14:04 AM
-Kevin
Post by: Alan on February 08, 2017, 10:42:06 AM
Rail power -> External bridge -> KA -> LM -> Motor =
Well pump -> Check valve -> Tank -> Valve -> Pipe & faucet
What is the flow rate of water coming out of a leak in the pipe? Answer: Only as much as the valve will permit.
Rail power -> LM -> KA -> Motor =
Well pump -> Valve -> Tank -> Pipe & faucet
What is the flow rate of water coming out of a leak in the pipe? Answer: Only as much as the valve will permit plus suddenly every bit of water in the tank.
Supercapaictors are really big tanks!
* the analogy is not completely correct but you get the idea.
Post by: William Brillinger on February 08, 2017, 11:01:32 AM
Quote
I'm glad we have guys who understand electronics on here, because I am amateur at best.
Me too!
Post by: William Brillinger on February 12, 2017, 07:01:50 PM
I received some Keep Alives last week, which I ended up ordering from Litchfield Station (http://litchfieldstation.net/), I wanted to order them from YankeeDabbler, but YD has an error in the price of the KA4 on their website and they never responded to my email or call, which is unusual for them :(
I am very pleased with the service from Litchfield, The order came very fast, the price was the same as YD, and when I asked if they would offer free shipping on my $200 order, they did! :)
I got several TCS KA-3's and TCS KA-4's. they are smaller than I imagined and guess what!?! THEY WORK!!
I did a test install of a KA-3 tonight into a Kato SD40-2
It worked, nothing exploded and it runs fantastically!
It also keeps the sound going for about 30 to 40 seconds after the power is removed if the motor is not moving.
One issue I did notice is that I am seeing frequent power warnings. No red warnings, just that the power is on the "low" side of good.
Soldering up the Bridge Rectifier:
I marked the + lead on the Bridge Rectifier with sharpie to keep it oriented correctly, and I used red heat shrink to easily identify it later:
I use a quick plug system so I can easily separate the trucks from the module as seen here:
Here you can see the KA-3 with the LM-2S:
As you can see I have a lot of spare wire in that unit, this is because I have not wired in the lights etc. eventually I hope to have a quick plug set to separate the carbody from the chassis too, then I will shorten everything up. this unit has more work to be done before that happens though.
Cheers!
Post by: Alan on February 12, 2017, 07:22:59 PM
Recall earlier I commented that an argument could be made if a KA is in one loco then it should be in all? You are experiencing why.
Post by: William Brillinger on February 12, 2017, 07:29:52 PM
Quote
Recall earlier I commented that an argument could be made if a KA is in one loco then it should be in all
Yup. And I will be adding them to my entire fleet for sure.
At which point I may look at increasing my track voltage by 2V.
Post by: Alan on February 12, 2017, 07:55:26 PM
You will have to put this on your train room door! ;D ;D ;D
Post by: William Brillinger on February 12, 2017, 09:14:13 PM
Thanks Alan, The voice of reality strikes again.
Post by: William Brillinger on February 12, 2017, 09:43:00 PM
The LM-2S manual says the MAX is 18V.
If I would increase the voltage to ~16V it should be fine and should eliminate the warning condition.
The question is this: Where would I find a 16V DC PSU that is appropriate for the Job.
I wonder if there is an appropriate 1.5 or 2V step-up converter?
On the downside, the KA3 is rated for 12 to 16V input. That might be cutting it close.
**Actually it should be fine, since the KA3 is on the lower voltage side of Diodes.
Post by: Alan on February 12, 2017, 10:05:17 PM
No, forget about a "converter". There are DC-DC converters (called buck converters) but they are pricey for units that can manage the 4 amp output of a power brick. Easily close to $100.
Do I look smug? I feel smug. ;D Bill, you need the same power supply I use - a 15V cheapie switch mode. They have an adjustment potentiometer on the circuit board that has about a 4 volt swing - 2V above and 2V below. A 15V switch mode power supply therefore can be anything from a 13V to a 17V unit. Mine are set at 16.9V to deliver 14.8V to the rails after the 3 diode drop in my occupancy detectors.
$18 bucks and free shipping. How can you go wrong? http://www.ebay.com/itm/RS-75-15-Mean-Well-Power-Supply-15V-5A-/122011245930?hash=item1c686fdd6a:g:9HkAAOSwnFZXWhje (http://www.ebay.com/itm/RS-75-15-Mean-Well-Power-Supply-15V-5A-/122011245930?hash=item1c686fdd6a:g:9HkAAOSwnFZXWhje)
The KA is on the far side of your bridge rectifier (2 diode path) so it will always be 1.4V below track voltage. The 3rd diode is after the KA but before the LM. The LM is the only component seeing the full 2.1V drop.
Post by: William Brillinger on February 12, 2017, 10:32:28 PM
Quote
The KA is on the far side of your bridge rectifier (2 diode path) so it will always be 1.4V below track voltage. The 3rd diode is after the KA but before the LM. The LM is the only component seeing the full 2.1V drop.
That's what I meant, but thanks for the full detail version.
Quote
$18 bucks and free shipping. How can you go wrong?
I'ma do that.
Post by: Alan on February 13, 2017, 09:15:46 AM
The relevance is a 0.7V swing is built into the homemade LM / KA arrangement. Important? I doubt it. Just know that it is there.
Post by: Alan on February 14, 2017, 07:45:14 AM
Bill, check this out. An inexpensive solution for your power supply issue: http://www.ebay.com/itm/150W-DC-DC-Boost-Converter-10-32V-to-12-35V-6A-Step-Up-Power-Supply-Module-Board/132053796647?rt=nc&_soffid=5023882200&_soffType=OrderSubTotalOffer&_trksid=p5731.m3795 (http://www.ebay.com/itm/150W-DC-DC-Boost-Converter-10-32V-to-12-35V-6A-Step-Up-Power-Supply-Module-Board/132053796647?rt=nc&_soffid=5023882200&_soffType=OrderSubTotalOffer&_trksid=p5731.m3795)
Post by: William Brillinger on February 14, 2017, 08:02:29 AM
I wonder if the short circuit protection on the PWR56 side would be affected by using this?
- Bill
Post by: Alan on February 14, 2017, 08:37:08 AM
Post by: William Brillinger on February 15, 2017, 12:49:39 PM
What about using this up converted higher voltage with the AR's and other Ring Engineering components?
Do you see any issues there?
Post by: TwinStar on February 15, 2017, 12:52:06 PM
Post by: William Brillinger on February 15, 2017, 01:21:56 PM
I have my locos dialed back to 75% for top speed, so I have headroom if for some reason there was not enough power at the motor, but I don't see it being an issue.
Post by: TwinStar on February 15, 2017, 01:24:32 PM
Post by: Alan on February 15, 2017, 02:08:55 PM
Quote
What about using this up converted higher voltage with the AR's and other Ring Engineering components?
The only potential issue with a DC up converter, or any buck converter for that matter, is ripple current resulting from the internal inductor design. Here is an article explaining: https://www.digikey.com/en/articles/techzone/2015/apr/the-advantages-and-drawbacks-of-dc-to-dc-voltage-converters-with-integrated-inductors (https://www.digikey.com/en/articles/techzone/2015/apr/the-advantages-and-drawbacks-of-dc-to-dc-voltage-converters-with-integrated-inductors)
An LM has awesome ripple rejection as evidenced by its ability to run on a DCC layout. A DCC signal is the very definition of a highly rippled supply line. The LM functioning proves Ring incorporates sufficient supply rail smoothing in the LM circuit. No reason to think they wouldn't do the same in the other components. I don't think you have anything to worry about with the up converter other than it will waste a little more electricity as heat.
$18 bucks for a fully conditioned power supply is still the better way to go however.
Post by: Alan on February 15, 2017, 02:31:18 PM
Quote
What is the net effect of lower voltage at the motor? Poor performance or just a lower top speed?
Lower top speed for sure but that is rarely a problem. Models run too fast anyway unless you have a gigantic layout. I'm with Bill, I have my locos dialed way back so their top speed is more correct for the curve radii of my layout. They don't move fast but what's your hurry? The end of the layout is less than 100 track feet away! :P
The performance issue is something Kevin may be able to shed light upon with his battery deal. The LM uses PWM. So, even at speed step 0.1% there is still full track voltage delivered to the motor however briefly. The beauty of PWM is the ability to turn a motor very slowly because the full voltage burst overcomes internal friction and loading on the motor. Something low voltage can't do. With reduced voltage the ability to nudge a motor is also reduced. This begs the question: Is there a slow speed operational difference of a loco between 14.8V and say 12V? The math says there must be. Curious what the eye says.
I spent way too much time and money making my power system work like tits-on-ritz so I will not be able to definitively answer this question. I can say LM equipped locos creep along very well at 14.8V.
Post by: William Brillinger on February 15, 2017, 02:40:30 PM
Quote
$18 bucks for a fully conditioned power supply is still the better way to go however.
Same question: Are there going to be any issues using AR's etc with these?
Post by: Alan on February 15, 2017, 02:51:41 PM
RP is selling the same thing just in different packaging.
Post by: TwinStar on February 15, 2017, 02:55:30 PM
QuoteWhat is the net effect of lower voltage at the motor? Poor performance or just a lower top speed?
Lower top speed for sure but that is rarely a problem. Models run too fast anyway unless you have a gigantic layout. I'm with Bill, I have my locos dialed way back so their top speed is more correct for the curve radii of my layout. They don't move fast but what's your hurry? The end of the layout is less than 100 track feet away! :P
I model the Twin Star Rocket. She would hit the top of the siding at 100mph and still make the platform for her station stop! I like to be prototypical. ;D ;D
Post by: KPack on February 15, 2017, 03:06:53 PM
Quote
The performance issue is something Kevin may be able to shed light upon with his battery deal.
Here's what I've noticed so far with a battery:
1.) The top speed is a bit lower, but that is no big deal for me....like others I lower the top speed on all my locomotives to somewhere between 60-70%. That leaves me plenty of room to play with it, but since this locomotive and it's sister unit will be running locals and switching industries, I really don't want them to go fast.
2.) Start voltage is affected. I had to up the start voltage in order to get the locomotive moving.
3.) Pulse width modulation is affected. I couldn't put my finger on it while testing my rig, but Alan identified it. I noticed the locomotive wasn't starting as smoothly as it did before and would require a bit more "umph" to get it rolling (you can see this in the video). Once moving, it will glide along as slowly as I want. But getting everything moving initially takes a bit more power. It has to be that the motor is only given 12V instead of the normal 14.8V, and this doesn't quite overcome the friction of a static locomotive as easily as 14.8V does.
I run on a friend's DCC layout periodically and for whatever reason he has his Digitrax set up to give something like 16-17V to the rails. I'm at the upper end of Railpro's safety net when I run there (constantly getting a overvoltage warning). The locomotives take off immediately upon minimal throttle, whereas on my own rails they crawl along smoothly.
-Kevin
Post by: TwinStar on February 15, 2017, 03:10:15 PM
Quote
I run on a friend's DCC layout periodically and for whatever reason he has his Digitrax set up to give something like 16-17V to the rails. I'm at the upper end of Railpro's safety net when I run there (constantly getting a overvoltage warning). The locomotives take off immediately upon minimal throttle, whereas on my own rails they crawl along smoothly.
-Kevin
That's too high. Does he have it set to O scale? I know an N scale setting will give under voltage alerts.
Post by: Alan on February 15, 2017, 03:14:20 PM
Quote
I model the Twin Star Rocket. She would hit the top of the siding at 100mph and still make the platform for her station stop! I like to be prototypical.
Too funny
Post by: KPack on February 15, 2017, 03:28:42 PM
-Kevin
Post by: Alan on February 15, 2017, 03:59:25 PM
Quote
Never had an issue with it, which speaks to Ring's engineering on the LM.
Agreed. It also speaks to the general robustness of today's electronic components. Manufacturing marvels they are.
When I was designing/testing my benchmade power district circuit breaker design I put an LM through an electrical torture test. I bought the module just for that purpose fully expecting it to fail or be damaged from the abuse. The poor thing was exposed to probably hundreds of short circuits, fully stalled motor tests, and at a wide range of breaker trip currents. Once, a component choice mistake of mine caused the breaker to go into oscillation. The LM was being turned on and off at a pretty high frequency for several seconds before I realized what had happened. It came through all this unscathed! It is in an operational loco today doing just fine. RP LMs are bulletproof.
Couldn't help but think of all the forum posts where someone says they fried a DCC decoder. I now know the only way to fry a LM is with a torch!
http://www.lkorailroad.com/circuit-breaker-and-block-detector-final-units/ (http://www.lkorailroad.com/circuit-breaker-and-block-detector-final-units/)
Post by: TwinStar on February 15, 2017, 04:27:28 PM
No, I checked and it's set to HO. I flipped it to N a few times while I was over there running and that put it right on target. Problem is when I do that it messes with all his settings on his reverse loops. So I just run on the high voltage and keep an eye on the module temps. Never had an issue with it, which speaks to Ring's engineering on the LM.
-Kevin
That's interesting. All we run on top of with Free-mo is Digitrax. I haven't seen voltages that high that I can recall.
Post by: TwinStar on February 17, 2017, 06:42:42 AM
Post by: nodcc4me on February 17, 2017, 08:51:35 AM
While we are discussing KA, BLI Blue Line locomotives have KA capacitors installed alongside the circuit board, which also have 8-pin connectors. Has anyone installed an LM in a Blue Line loco? If so, does the KA do the job?
Post by: TwinStar on February 18, 2017, 11:06:30 PM
I'm trying but apparently I can't load photos here for some reason and it's been so long since I've had to use image hosting that I can't remember who I used to use.
I'm not ignoring the request but trying to remember how to do this.
Jacob
Post by: Alan on February 18, 2017, 11:11:53 PM
Post by: TwinStar on February 18, 2017, 11:15:20 PM
Your attachment couldn't be saved. This might happen because it took too long to upload or the file is bigger than the server will allow.
Please consult your server administrator for more information.
Back
Post by: Alan on February 18, 2017, 11:19:50 PM
Resize your image down to around 600-800 pixels wide and try again.
Post by: TwinStar on February 18, 2017, 11:24:26 PM
Post by: nodcc4me on February 19, 2017, 07:54:23 AM
Post by: TwinStar on February 19, 2017, 01:34:07 PM
I'm a Mac guy and don't use MS or Paint. However, I used to use Seashore back when you had to resize everything before posting. However, my security settings wouldn't open a program that wasn't blessed by Apple. So, I had to change those. !@#$%, this Mac is reminding me of DCC. Anyways, here they are. Finally.
[ Jacob, I inserted the images into your post so others could see them at full size without clicking on them. - Bill (admin) ]
Post by: nodcc4me on February 19, 2017, 02:10:49 PM
Post by: Alan on February 19, 2017, 02:24:13 PM
https://www.kaptontape.com/1_Mil_Kapton_Tapes.php (https://www.kaptontape.com/1_Mil_Kapton_Tapes.php)
Post by: Alan on February 19, 2017, 02:48:06 PM
Quote
Alan, no, I don't think I'm going to be making a product here.
You (we?) should reconsider. Super caps in the value needed are 15 cents each in quantities of 1000 (7 cents 10,000 and up) on Alibaba. Drilled 100mm x 100mm boards (qnty 1000) are less than $1.00 ea from US. Overseas would be a fraction of $1.00. If we are willing to solder them ourselves (12 solders) I bet the total cost of a KA could be kept below $2.00 without having to invest in a huge inventory.
Don't these things sell for like $25.00 each? Not a bad return on a $2.00 investment.
Post by: William Brillinger on February 19, 2017, 02:52:44 PM
Quote
You (we?) should reconsider.
If you make 'em, I'll market 'em. 8)
Post by: TwinStar on February 19, 2017, 03:44:32 PM
Thanks for the pictures Jacob. Sorry you had to go through so much trouble. Nice install. The KA-4 fits perfectly back there. Your Proto E7 is a bit different than mine. Looks like you removed some weights to make room for the module. This is one of my first RP locomotives so I have an LM-1 installed on top of the factory board, which is mounted behind the motor above the gear tower. No room for a KA-4 with or without the board without removing weight, which I hesitate to do. Thanks again.
It's no biggie. I needed to learn how to resize again anyways.
I didn't remove any weight and only removed a few small mounting posts that the board was attached to. I have an older E8 and the frames/drives/shells are night and day different. I did remove the crappy P2K motor and replaced it with a Kato and NWSL drivelines.
Once these five E units are complete I'm moving to a six pack of GP-7's and will probably replace the motors and ad KA's as well. These KA4's might fit in the cab of the short hood but I haven't really looked yet. I don't imagine I'll ever be able to use a Ring solution if it comes about due to the bridge rectifiers we're using and the need to keep things the same voltage at the motor.
Post by: TwinStar on February 19, 2017, 03:45:36 PM
Jacob, is that black electrical tape? If so, it tends to leave a gooey mess if you remove it after a period of time. Try Kapton tape next time like shown in Al's installation. Kapton is heat resistant and leaves behind no adhesive.
https://www.kaptontape.com/1_Mil_Kapton_Tapes.php (https://www.kaptontape.com/1_Mil_Kapton_Tapes.php)
It is and next week will be 'was'. Thanks for the link, I wasn't aware of that product. I found the same thing on ebay much cheaper with free shipping if anyone else is interested.
Post by: William Brillinger on February 19, 2017, 04:05:11 PM
Quote
I don't imagine I'll ever be able to use a Ring solution if it comes about due to the bridge rectifiers we're using and the need to keep things the same voltage at the motor.
Jacob, can you please explain what you are referring to here?
Why will the bridge rectifiers affect using a Ring solution?
What need to keep things the same voltage at the motor?
Post by: TwinStar on February 19, 2017, 04:09:08 PM
Based upon this I don't imagine the two would play so well together. Perhaps I'm wrong.
Post by: TwinStar on February 19, 2017, 04:10:50 PM
Sorry Rob. I was posting while you were. By sheer accident almost answered your question. Sure, RP will run fine on 12.5V. Top speed may suffer.
A case could be made that if one loco has keep alive capacitors then all your locomotives should also. It would be interesting to study the MUing function between KA'ed and non-KA'ed. I am sure it will still work but am curious if there is an observable deterioration in function.
Here is what I was referencing in my above post.
Post by: William Brillinger on February 19, 2017, 04:29:11 PM
Post by: Alan on February 19, 2017, 04:35:35 PM
Any KA, regardless of manufacturer, if connected ahead of a module will require a bridge rectifier.
DCC AC or unknown DC polarity voltage is derived from the rails and fed to the module (decoder). Inside the module the power moves thusly: --> Bridge rectifier --*2--> Filter caps --> Regulator --> Stabilization caps --*1--> PWM controller --> Output to motor. The LM3 pin arrangement allows us to tap into where I marked as *1. DCC decoders and what we need in an LM is to be able to tap into where I marked *2.
It is literally nothing more than an additional 'wire' inside the module and a pin in the connector. Simple, right? Not so. It means redesigning the LM PCB, something that would be expensive for Ring.
We are not waiting on a Ring KA per se. We are waiting on Ring to make available connection to the output side of the internal LM bridge. With that, we could use any KA we wish. Should that occur then, Jacob, your KAs will still work fine. You will simply discard the bridge rectifier and attach you KA to the new LM wires.
Post by: Alan on February 19, 2017, 05:15:22 PM
Post by: Morebassman on February 19, 2017, 07:13:29 PM
Post by: TwinStar on February 20, 2017, 10:14:12 AM
Post by: nodcc4me on February 20, 2017, 10:30:13 AM
Post by: Alan on February 20, 2017, 10:52:06 AM
Another consideration is the forward voltage drop. Typically, it is 1.4V for a common bridge rectifier. Units with lower drop are readily available. The rectifier in my link above only has a 750mV drop. For use with an LM, the lower the forward voltage the better. Lower forward voltages will reduce the, what Kevin calls additional oomph to get started, effect.
You can filter by both forward voltage and case style on most electronics suppliers web sites. Here is the Mouser filter:
Post by: KPack on February 20, 2017, 11:09:49 AM
Quote
...what Kevin calls additional oomph to get started...
That's a scientific term right there. Perhaps I should trademark that?
Thanks for the link to the SMD rectifier. I had wondered if there was something like that out there. I'm all about SMD components....ever since I made the switch to SMD resistors and I realized how much space it free'd up. If I was really organized I could etch a small board and solder on all the resistors I need and the bridge recitifier, then just solder the wires to the board. Much cleaner. But etching is nasty stuff and takes a good amount of work from what I understand.
-Kevin
Post by: Alan on February 20, 2017, 11:52:08 AM
Quote
But etching is nasty stuff and takes a good amount of work from what I understand.
Not true. Very easy to do. Check out this post: http://www.lkorailroad.com/control-panels-part-vii/ (http://www.lkorailroad.com/control-panels-part-vii/) You see I use an old lamp, kitty litter pans, and a picture frame. Not exactly hi-tech but works great.
You may be thinking about the old negative resist process. Yeah, it was a PIA. Required an expensive UV bulb and was very finicky. In my younger days I made more junk boards than I did good boards using the negative resist process. Then positive resist came on the market. What a godsend. Made circuit board creation Fisher-Price simple. MG Chemicals is the dominant US supplier of positive resist materials: http://www.mgchemicals.com/products/prototyping-and-circuit-repair/presensitized-boards/positive-presensitized-600-series (http://www.mgchemicals.com/products/prototyping-and-circuit-repair/presensitized-boards/positive-presensitized-600-series)
The most challenging part of the process is not the etching of the boards. Etching is easy. The possibly challenging part is creating the circuit artwork. But for the relatively simple circuits needed for loco installation that shouldn't be a problem. The key is to draw with vector art, not raster art. I use Adobe Illustrator however it is an expensive program if you don't already own it. Inkscape is a free vector art program that would work great for circuit art. You could even use the Draw objects in MS Office PowerPoint if you wanted to. They are vector.
Print the art at 1:1 on an inkjet transparency and you are ready to go. 25 minutes later you will have a ready-for-solder circuit board.
Here are some examples of homemade circuit boards used on my railroad - power district circuit breakers and occupancy detectors.
Post by: Alan on February 20, 2017, 07:17:43 PM
The keep-alives you guys are building is the way to go with RP. Or, clean your track. :P
Post by: melarson on February 20, 2017, 09:46:13 PM
It is confirmed you should NOT connect a keep-alive to the ground pin and the blue wire of a LM3.
I have to say, this makes me wonder why the new "negative" pin on the LM-3 is there? If not for keep-alives (and I assume also batteries), then what? It is astonishing to me that Tim would make such a connection available that is apparently so dangerous if miss-used. Did he say what the purpose of the negative lead is supposed to be?
Michael
Post by: G8B4Life on February 21, 2017, 04:36:06 AM
I have to say, this makes me wonder why the new "negative" pin on the LM-3 is there?
I have to say I wonder as well. Given Rings statement on the LM-3
Quote
A 'Negative' terminal is better for battery powered connection and will be more flexible to add a keep alive product.why would you only go half way with the redesign of the LM board needed to bring these connections out to the user? Maybe Ring has some wizard circuitry up his sleeve that'll allow his KA's to be attached to the blue and negative wires.
Speaking of circuitry, Alan I'm constantly amazed with the homemade circuitry work you post on your blog and now here. Inspirational, helpful and I must say better "fire under the backside" tutorials than others I've seen out there.
- Tim
Post by: Alan on February 21, 2017, 09:53:32 AM
When you think about it the answer could not have been any different. The ground pin is immaterial. The root problem is huge capacitance on the Vcc rail at the load. Absolutely the worst place to put it in any electrical circuit.
One good use I can envision for the ground pin is to add a small non-lethal amount of capacitance. Just enough to keep the LEDs lit and the sound system from resetting when momentary loss of rail contact occurs but not so much as to fry the output transistors from a short circuit. Much like what other DCC sound decoders do with their external capacitor. This wouldn't do anything for the motor function but, as I understand it, momentary sound resetting is a common annoyance for some. It may fix that problem.
Tim isn't going to reveal granular design details. Nor should he. He has a business to protect. We can only speculate and deduce from basic electronics facts. We suspected, and Tim confirmed, the blue wire / ground pin is not a safe place to connect a motor KA. All this exercise did was confirm you guys are doing it the right way with your homemade KA circuits. At least on LM1, LM2, and LM3 units. Who knows what is in store for LM4s.
Post by: Alan on February 21, 2017, 01:14:22 PM
Quote
Maybe Ring has some wizard circuitry up his sleeve that'll allow his KA's to be attached to the blue and negative wires.
KA with built in current regulation - done.
Mark Gurries lists regulated/not-regulated on the web page Bill linked to earlier however, regulated in the DCC KA world means voltage regulation as noted in Mark's answer to question #4. We need current regulation. If I were a betting man my chips are on a RP KA product that has built in current regulation. Then blue wire/ground pin makes total sense. The addition of the ground pin may be laying the groundwork for such a device.
It is worthy to note our entire discussion revolves around protecting your LM. If you are confident you can install in such a manner that a short circuit will never occur inside the locomotive then putting a store-bought KA on the blue wire/ground pin will work. Essentially, a $70 gamble on your part. Tim Ring on the other hand knows people will make mistakes and the last thing he wants is more warranty costs or bad public image by not honoring warranty. When he receives a returned dead module he has no way of knowing that a KA killed the module. Hence, he is doing the logical thing by saying don't attach a KA.
Post by: William Brillinger on February 21, 2017, 01:29:11 PM
I just want to take a moment and publicly thank you Alan, for your invaluable contributions to this forum.
I hope I speak for everyone when I say I really appreciate your time and input here.
There are so many great people here and I am thankful for every one of them, but your efforts stand out as above and beyond.
Thank you.
Post by: Alan on February 21, 2017, 02:01:45 PM
Quote
I hope I speak for everyone when I say I really appreciate your time and input here.
Thanks Bill. It is fun playing RP detective trying to piece together the clues! They are just clues however. Being thrown into the conversation for little more than conversation's sake. If the thread helps someone build a better RP KA for themselves then so much the better. Glad to be a part of it.
Odd isn't it, for someone with no need whatsoever for KAs to get so engrossed in the subject? Clean your track dagnabbit!
Post by: nodcc4me on March 21, 2017, 04:10:28 PM
Post by: KPack on September 20, 2017, 10:40:37 AM
I remember Bill saying somewhere (might have been in an email to beta testers) that he has changed to adding KA's to LM-3S's without a bridge rectifier so that there isn't a drop in voltage to the module. Apparently the drop in voltage was causing the loco to not play nice with others. Bill can you confirm?
I have a KA4 and a bridge rectifier installed in two different locomotives and did some testing over the past few days to see if I had the same issues. I did not. The KA'd locos MU'd just fine with any others and there was no perceptible difference in behavior. The KA'd loco was not slow on starting and always ran perfectly with any other loco (this includes a lashup of 5 very different locomotives). It did fine as either a leader or a follower.
The only thing I can think of that would cause such a difference between my experience and Bill's experience is that on both of my locomotives were "fresh" installs. Meaning, both modules had all data erased (previously used), new programs and sounds loaded, and the Full Load Motor Current was redone. Bill, did you redo the Full Motor Load Current when you installed a bridge rectifier and KA? I think that may have made the difference for me. I assume the FMLC is different for a locomotive that is seeing less voltage (12.9 on the readout).
-Kevin
Post by: William Brillinger on September 20, 2017, 11:52:11 AM
I may have been hasty with my conclusion that the Bridge Rectifier was causing issues between locomotives.
I will do some testing this evening to confirm how my units are behaving.
Yes, I reset the FLMC values after making any changes to a locomotive.
One thing I am wondering about: I recall reading that you have adjusted the startup speeds. I have not done this, and it may be a factor.
Post by: KPack on September 20, 2017, 01:29:06 PM
Basically I try to get my locomotive to run as smoothly as possible and not take off like a rocket. Railpro is hands down far smoother than DCC ever was, and far better at slow speed control. I can have a full MU set crawl along at a snail's pace perfectly in unison. Never could get that will DCC.
-Kevin
Post by: Alan on September 20, 2017, 01:40:59 PM
I adjust the start voltage depending on the locomotive. I try to get it moving at about 5% throttle or so. Most of them are set between 7-11% start voltage. I also drop my top speed on almost all of them to 70% or lower, again depending on the locomotive's gearing and how fast they like to move. Adjusting the top speed will affect how fast the locomotive moves at low speeds.
-Kevin
I've found when working with new locos it is sometimes necessary to revisit these two settings after the unit gets well broken in. I have a 22"R loop for break-in but still like to tune after the powertrain has some miles on it.
Post by: William Brillinger on September 23, 2017, 08:46:39 AM
Quote
I may have been hasty with my conclusion that the Bridge Rectifier was causing issues between locomotives.
I will do some testing this evening to confirm how my units are behaving.
It appears my issues with my units that still have LM-2S's are more related to a surging issue. This is an issue with the locomotive, not the RailPro module. It's one of my Athearn Genesis GP38-2W's that hit the floor a while back during an op session. It seemed like it was ok at the time, but there might be something broken in a gear tower.
Post by: William Brillinger on October 27, 2017, 01:49:22 PM
Here is how I am installing KA's in my LM3 series units: https://rpug.pdc.ca/index.php/topic,383.msg2872.html#msg2872
Yes there is a possible issue with short protection, but this is what Tim Ring designed these inputs for.
If you are concerned about this, then use the Bridge Rectifier and put the KA's on the outside of the LM power inputs.
WARNINGS:
- This only applies to RailPro LM3 modules. Older Modules need a Bridge Rectifier to install a Keep Alive
- This will probably void your warranty from Ring Engineering.
Wire the Keep Alive as follows:
Module 9 Pin Blue (Common Wire) to KA3/4 Blue +12v Wire
Module 6 Pin Yellow (Negative Wire) to KA3/4 Black/White Ground Wire
(https://rpug.pdc.ca/index.php?action=dlattach;topic=383.0;attach=573;image)
Post by: nodcc4me on October 27, 2017, 02:13:45 PM
Post by: TwinStar on November 09, 2017, 02:12:08 PM
Thanks.
Post by: William Brillinger on November 09, 2017, 02:38:23 PM
Here are my images for wiring an LM-2 with bridge rectifier:
Post by: TwinStar on November 09, 2017, 02:54:50 PM
Post by: TwinStar on December 08, 2017, 04:33:16 PM
Thanks.
Post by: Alan on December 08, 2017, 04:44:51 PM
Where did the KA install with bridge rectifier diagram go? I can't find it after the rearrangement.
Thanks.
Bill's Nov 9 reply two replies above yours in this thread.
Post by: William Brillinger on December 08, 2017, 04:47:40 PM
Or, I have the instructions also posted on my website at the bottom of this page:
http://pdc.ca/rr/catalog/product/wiring-harnesses-keep-alives-and-other-accessories/63
Post by: KPack on December 08, 2017, 05:14:24 PM
-Kevin
Post by: TwinStar on December 08, 2017, 05:29:25 PM
Where did the KA install with bridge rectifier diagram go? I can't find it after the rearrangement.
Thanks.
Bill's Nov 9 reply two replies above yours in this thread.
I apologize. I was looking for a drawn diagram and completely glossed over that. :-[
Post by: William Brillinger on December 08, 2017, 05:48:40 PM
And I agree with Kevin, the KA's work as advertised and the Bridge Rectifier does not cause any issues when MUing with units that do not use them. All of my Locos are KA equipped.
Post by: EsselW on February 14, 2018, 11:24:06 PM
Post by: KPack on February 14, 2018, 11:59:53 PM
-Kevin
Post by: Alan on August 12, 2019, 08:11:17 PM
https://phys.org/news/2019-08-supercapacitors-turbocharged-laxatives.html (https://phys.org/news/2019-08-supercapacitors-turbocharged-laxatives.html)
Keep you eye peeled for ionic electrolyte super capacitors coming onto the market. Maybe your next keep-alive will run for hours!
Post by: Dean on September 29, 2019, 10:44:44 PM
Post by: Alan on September 30, 2019, 05:59:13 AM