Stay-alive capacitor hacking

Post all your DCC only problems, solutions and discoverys here.
Brianetta
Posts: 190
Joined: Wed Feb 11, 2009 12:49 pm
Location: Jarrow
Contact:

Re: Stay-alive capacitor hacking

Postby Brianetta » Wed Apr 14, 2010 12:23 pm

I'm actually giving Avidemux GTK a go. "Free trial" suggests to me that you're recommending a Windows program, which would be great if I was a Windows user! Thanks, though. The effort is appreciated.

User avatar
Bufferstop
Posts: 12696
Joined: Thu Mar 11, 2010 12:06 pm
Location: Bottom end of N. Warks line

Re: Stay-alive capacitor hacking

Postby Bufferstop » Wed Apr 14, 2010 3:42 pm

Aha! The penguin strikes again! I might just download Avidemux GTK onto the MythTV box that lurks in the top cupboard just behind me. it tells me I've been logged in for just over 7 months. One day when neither the missus or me have to fit in with our customers I'll get round to changing my desktop. But only when there aren't more important modelling tasks to finish :D
Growing old, can't avoid it. Growing up, forget it!
My Layout, My Workbench Blog and My Opinions

User avatar
S_Jay
Posts: 241
Joined: Thu Dec 27, 2007 4:38 pm
Location: Nottingham
Contact:

Re: Stay-alive capacitor hacking

Postby S_Jay » Wed Apr 14, 2010 8:30 pm

Really outstanding Brian! A lesson to all, well done for sticking to this!

Since I last posted I have found a soldering a few extra droppers on My 08 runs very slowly with no stutter over a double slip and or 3-way (mind you they are streamline Peco). I would still recomend swapping your hornby points for Peco Set-track ones, they are much better with a much small plastic frog.

And of course for your next layout you'll know just what to do :)

Well done again!
My Layout, My world, My rules, if that's ok?
My layout with no progress http://www.newrailwaymodellers.co.uk/Forums/viewtopic.php?t=14866
My other layout with progress http://www.newrailwaymodellers.co.uk/Forums/viewtopic.php?f=22&t=18394

Brianetta
Posts: 190
Joined: Wed Feb 11, 2009 12:49 pm
Location: Jarrow
Contact:

Re: Stay-alive capacitor hacking

Postby Brianetta » Wed Apr 14, 2010 11:34 pm

Next layout will have Peco. My current layout does not, because I'm on a budget. I really would have better gear if I could afford it! My Hornby points were on £7 special at Hattons - a price that's still current.

I've taken some more video footage, and I'm practising with Avidemux. I'll have a new video soon. I spent part of this evening putting lighting into a brake van. A completely custom job, with a capacitor so that it doesn't go out completely over insulated points. Poor thing only picks up on two wheels. Re-usable skills rock. Only problem is that it has a nice warm interior light, but no red lights at the back. That'll wait until I get a really nice brake van, and fit a decoder to drive the lights properly.

Brianetta
Posts: 190
Joined: Wed Feb 11, 2009 12:49 pm
Location: Jarrow
Contact:

New video!

Postby Brianetta » Thu Apr 15, 2010 4:03 pm

Here's the video I promised! Image Click to watch.

Apologies for the lame quality at the start. My tools are a still camera that won't let you change zoom whilst recording in video mode, and a few hour's experience with Avidemux.

User avatar
Bigglesof266
Posts: 927
Joined: Tue Feb 10, 2009 9:59 am
Location: Australia

Re: Stay-alive capacitor hacking

Postby Bigglesof266 » Fri Apr 16, 2010 12:20 am

S_Jay wrote:<snip>I would still recomend swapping your hornby points for Peco Set-track ones, they are much better with a much small plastic frog.<snip>

I have a mix of both brand's setrack points S-Jay. All < 12 months old & well maintained. The Hornby are current model "M" sufffixed Hornby points (frog and guide gap have been modified from their predecessor). In this case, the culprit isn't soley attributable either point brand's plastifrog. It's the chassis of that particular loco, which sharing a common chassis also affects Hornby's 'Jinty'.

Setrack points don't cause me grief with any of my other locos (carriages or rolling stock for that matter), including other 0-6-0s from Hornby. eg: Class 08 and Q1. Admittedly, all of my kit is current generation which undoubtedly helps. That said, regardless of the pros and cons of electrofrog versus setrack points, my own experience suggests setrack points receive an exaggerated, undeserved and out of proportion whipping for mixed reason of legacy, elitism, and general disinformation through wannabe alignment perpetuating "I heard/read" mythology. Their geometry notwithstanding, set up properly with adequate DCC feed and contemporary technology locos running a decent decoder, either brand current setrack points work OK IME.

FWIW, I have noted that Hornby's R8249 decoders are much more susceptible to stalling at the slightest interruption to and thus loss or weakened (DCC) signal (versus potential voltage drop/complete current loss) caused by that loss of contact traversing the plastifrog than Bachmann's 36-55x decoders are. I don't know why, but have observed simply that it is so. I don't know if a TCS decoder would improve that situ, but I don't think I'd bother installing a TCS decoder in a Railroad 08 even if it would. Cheaper, smarter and easier just to replace the flawed chassis than either all one's points or decoders. I learned a long time ago not to try and turn a CB into an RG or TZ. Or to put it another way, if one needs a race bike, buy a race bike with the performance frame, brakes, suspension, motor etc rather than trying to turn the sows ear into a silk purse. I suppose one might say the same about electrofrog points as well, but in this case, the CB (of points) serves my needs reliably enough.

Brianetta
Posts: 190
Joined: Wed Feb 11, 2009 12:49 pm
Location: Jarrow
Contact:

Re: Stay-alive capacitor hacking

Postby Brianetta » Fri Apr 16, 2010 12:35 am

Well said.

Any future layout of mine would have Peco points, because I'd be using streamline geometry. Electrofrogs might make an appearance, largely for cosmetic reasons, if the wiring isn't too much of a headache. Bear in mind here that my points are entirely mechanical.

I'll be updating everything in time. FIrst "expensive" components will be a better DCC controller and some better decoders. Decoders that support a top speed, for example. Decoders that work on DC...

My layout's being done on a real budget. We've only spent about £300 in total so far, £400 if you include investment in a rotary tool, soldering station and so on. Replacing everything is, quite literally, a last resort.

User avatar
Bigglesof266
Posts: 927
Joined: Tue Feb 10, 2009 9:59 am
Location: Australia

Re: Stay-alive capacitor hacking

Postby Bigglesof266 » Fri Apr 16, 2010 1:38 am

I think Peco setrack points are actually the better design than Hornby's for locos/rolling stock/coaches to actually traverse Brianetta, but they are (i) more fragile and (ii) dearer, their two cons.

I use a mix of predominently Streamline 100 (with traksetta), Hornby and Peco setrack points, Hornby SBG track and a just couple of bits of Peco setrack thrown in here and there, ie: level crossings and short straights in a length Hornby don't manufacture. At the moment, HoG predominates here when it comes to point setting. The two aspects at which HoG is unchallengeable as 'best in class' are reliability and cost. :lol: And it's kind of fun, reminiscent of childhood in the same way as licking an icecream in a conical wafer cone.

That said, I am very much in the try everything & learn curve with my current layout, which one might call perpetually experimental. A mix of SEEP, Hornby & Peco solenoid point motors, side surface mounted, under table extended pin in Peco's mount or countersunk in a cutout connected directly to the point. I love the concept of Tortoise motors, and may well try some in due course, but the budget for implementing them on my current layout (22 points and crossovers) would be significant, so other than one or two experimental when I get around to it, they can wait until layouts Mk III or IV methinks. I am constantly amazed as to why cheap as chips RC mini-servos haven't been deployed in model railroading as a means of manipulating points. Surely some cleverer bugger than me could come up with a PCB mounted IC through which you could signal them like a 10 channel RC receiver does.

A restrictive budget is often where the most fun is to be found. I've had as much fun poking about in a venerable 30 year old comfy and spacious cockpitted Grob Speed Astir as I have X-country in a built for beanpoles 10 year old Rolladen-Schneider LS8, or a so well behaved, ergonomically well thought out and comfortable Boeing could have built it yet solid performer in class Schempp-Hirth Discus CS brand spanker. If I add it up, I think I've spent as much on model railroading in the past 15 months as I spent buying my 3.8M 'tinnie', inc a new motor and new trailer. Shudder. :oops: :roll: It's just so..oo...ooo easy to get carried away spending, and I'm not one for "wanting everything I see" nor an impulse shopper. I've disciplined myself it's time to stop awhile and enjoy using what I already have. Definitely a cut one's coat according to one's cloth man myself, so that said, you've done absolute marvels with £400!!!

User avatar
ngresley
Posts: 366
Joined: Fri Jan 08, 2010 9:18 am
Location: Canberra, Australia

Re: Stay-alive capacitor hacking

Postby ngresley » Tue Apr 20, 2010 11:05 am

I'm very interested in this thread as I too am starting with the Mixed Goods set, and the first bit of the layout I'm building (branch line terminus) is using Hornby dead frog points. Good luck with your investigations!
"I'd like to die quietly in my sleep like my grandfather, rather than screaming in terror like his passengers"

Brianetta
Posts: 190
Joined: Wed Feb 11, 2009 12:49 pm
Location: Jarrow
Contact:

Re: Stay-alive capacitor hacking

Postby Brianetta » Tue Apr 20, 2010 12:19 pm

ngresley wrote:I'm very interested in this thread as I too am starting with the Mixed Goods set, and the first bit of the layout I'm building (branch line terminus) is using Hornby dead frog points. Good luck with your investigations!

Well, I'm done. The experiment has concluded, with mixed results.

First, the practicalities. Soldering onto the DCC controller is fiddly and requires a good ten minutes with a cup of tea afterwards to recover, possibly also a lie down. Technically, it all works fine, but only the class 08 has enough room inside it for the capacitors I used. The high voltages involved are a problem; they preclude the use of supercapacitors, and limit the choice of regular electrolytics. Chances of successfully modifying the Jinty, which has usable space only in its coal bunker and tanks, are slim.

Performance of the shunter over the dead frogs is wonderful, and I'm really enjoying shunting things about at realistic speeds. Seriously, I can't get enough of it. Perhaps I should get out more. Unfortunately, there's a big drawback in that whenever I program the modified decoder, it goes loopy and lurches off down the line in high speed fits. Not only that, but other locomotives will do the same if they're being programmed at the same time. The programming is always successful, but I can't do it on the layout, and what's more, I can't then use the modified locomotive until the capacitors have discharged. It would be well worth fitting a switch to the loco for programming purposes, that switches out the capacitors completely (making the mod "optional," so to speak). I might do this, if I can find a decent spot for the switch. Perhaps a tiny one, poking up through the chimney hole.

Would I modify any more? No, it's highly unlikely. I'm glad I did this one, though. That was worth it. I'd recommend that anybody with this sort of shunter think about it; just do put that switch in place!

I'm considering putting together an article on this, taking one through from start to finish.

Martin71
Posts: 160
Joined: Mon Apr 16, 2007 8:47 am
Location: Perth WA

Re: Stay-alive capacitor hacking

Postby Martin71 » Tue Apr 20, 2010 1:18 pm

Hi

Once again well done. Now did you find you got better motor control at lower speeds and smoother running? I understand needing the cuppa afterwards, can you imagine the pressure when doing this inside a multi layered loksound decoder?

I am going to be very blunt here, firstly I think you have done an outstanding job with the resources you had and I am referring to the select and the basic Hornby chip.

Its a totally different ball game when you can change the CVs speed curves etc and have a good quality decoder responding to all this.

As to having a switch in it yes I add them too because I found a choke or inductor on the neg leg can be unreliable when programming.

I got these from a robotics shop in the UK. They were about 40p and you can also get right angled ones. Sorry cannot find the link at the moment.

Image

Re visit the project later on, dont give it up and for the future consider maybe upgrading your controller.

Great post.

Martin

Brianetta
Posts: 190
Joined: Wed Feb 11, 2009 12:49 pm
Location: Jarrow
Contact:

Re: Stay-alive capacitor hacking

Postby Brianetta » Tue Apr 20, 2010 1:33 pm

Martin71 wrote:Hi

Once again well done. Now did you find you got better motor control at lower speeds and smoother running? I understand needing the cuppa afterwards, can you imagine the pressure when doing this inside a multi layered loksound decoder?

No chance; I'd get one with a stay-alive built in if I was spending that sort of money. I didn't get better motor control. Its handling characteristics were identical, except that it didn't ever stall. Remember, the decoder is still doing all the driving. I'm just keeping the decoder powered up.

I am going to be very blunt here, firstly I think you have done an outstanding job with the resources you had and I am referring to the select and the basic Hornby chip.

Its a totally different ball game when you can change the CVs speed curves etc and have a good quality decoder responding to all this.

Yes. I probably have the best running Hornby Railroad class 08 in the whole city.
As to having a switch in it yes I add them too because I found a choke or inductor on the neg leg can be unreliable when programming.

I didn't try a choke. Not knowing how it is supposed to improve things, I wouldn't have known what to order, or why I was fitting it.
Re visit the project later on, dont give it up and for the future consider maybe upgrading your controller.

One day I will upgrade my controller, but not for the sake of this shunter. This was all about this shunter, and making it better. I did that. I'm in no position to thrown cash at my locos, decoders, controllers or track, and if I was, of course there'd be fewer problems.

I achieved exactly what I set out to do. I didn't have a choice of loco; that set was a present from my wife. I made it better.

Bigmet
Posts: 8038
Joined: Tue Aug 14, 2007 2:19 pm

Re: Stay-alive capacitor hacking

Postby Bigmet » Tue Apr 20, 2010 2:52 pm

Definitely write the project up, and why not try offering it to the mags, nothing like encouraging others to have a go.
Brianetta wrote:... Any future layout of mine would have Peco points, because I'd be using streamline geometry. Electrofrogs might make an appearance, largely for cosmetic reasons, if the wiring isn't too much of a headache. Bear in mind here that my points are entirely mechanical.
...

After what you have mastered, Peco electrofrog points will be a snap. And they really contribute to good running too, both the electrical continuity, and closer tolerances combined with larger radius giving reliable running. If you use the medium or large radius points and lay them carefully you can literally eliminate derailments on points.

A stiff wire pushrod to operate the point from the board edge, using a double pole slide switch to change polarity, makes for a very simple mechanism. You want the type of slide switch which has a cube of plastic as the 'handle': drill a hole in it at right angles to its' movement, bend the end of the push rod 90 degrees, and it locates in the hole. The back and forward motion of the switch moves the point tie bar (you may need a spring in the push rod to take up excess motion). Very little wiring too if using DCC. The point frog has to be connected to one or other of the DCC bus wires. If you run the DCC bus wires near the edge of the board for easy access, then it is a short link to the two input poles of the slide switch. The output from the switch can be connected by a short flying lead to the pushrod, and the power is then connected by another short flying lead from the pushrod to the point frog. I love making one component do two jobs, and there is never any doubt about which wire feeds which point: it is always fed from the pushrod attached to the tiebar.

Martin71
Posts: 160
Joined: Mon Apr 16, 2007 8:47 am
Location: Perth WA

Re: Stay-alive capacitor hacking

Postby Martin71 » Tue Apr 20, 2010 3:21 pm

Hi

No chance; I'd get one with a stay-alive built in if I was spending that sort of money. I didn't get better motor control. Its handling characteristics were identical, except that it didn't ever stall. Remember, the decoder is still doing all the driving. I'm just keeping the decoder powered up.


I must say I am surprised but you know the running characteristic's of your loco best. The DC ripple is cleaned up by the caps "generally" removing the jerks at low speeds and giving finer control.

I achieved exactly what I set out to do. I didn't have a choice of loco; that set was a present from my wife. I made it better.


Yes you did and same here my wife got me the DCC set about 3 years ago as a surprise and thats how I got back into it. Watched your video excellent and I also drink my tea with more milk than tea.

Couple of great suggestions from Bigmet!

m

Brianetta
Posts: 190
Joined: Wed Feb 11, 2009 12:49 pm
Location: Jarrow
Contact:

Re: Stay-alive capacitor hacking

Postby Brianetta » Tue Apr 20, 2010 4:26 pm

Martin71 wrote:Hi

I must say I am surprised but you know the running characteristic's of your loco best. The DC ripple is cleaned up by the caps "generally" removing the jerks at low speeds and giving finer control.

OK, I think you're confused as to exactly where these capacitors are, and what they do. The rectifier is the power source to the decoder. It hardly cares what voltage it receives, as long as it's above about three volts. That's enough to power it up, and get it sampling the track for a DCC signal. To drive the motor it needs a surplus above that; if the track sampler shows a DC voltage, then the motors are driven with the available surplus. If it's DCC, it waits for a command addressed to it, and uses the entire surplus (capped at 12V) as modulated pulses to drive the motors. Bear in mind that this surplus is not from the sampler, it's from the rectifier. Pulse modulated DC is going to thoroughly mask any ripple, as is the fact that it's gone through a few ICs in the mean time. That's in addition to the fact that there's barely any ripple present on a rectified square wave.

What I was taking advantage of is that the decoder doesn't switch into DC mode if the sampler reads no voltage at all. By keeping it powered up with a capacitor, it just continues obeying the last commands received, and will continue to do so until it gets new ones or until the power supply fails.

Since my capacitors are just maintaining that power supply, and are not filtering any output to the motors, there's no change in the running characteristics.

In fact, there is a small ceramic capacitor sat across the motor terminals that hisses with the pulse modulated power. I've been very tempted to dike that out. It's noisy (it hisses at the pulse modulation frequency) and it's very likely to be reducing the voltage to the motor by smoothing the pulses out. Good sharp pulses are excellent for high torque at low speeds. It's one area that DCC absolutely shines.

Yes you did and same here my wife got me the DCC set about 3 years ago as a surprise and thats how I got back into it. Watched your video excellent and I also drink my tea with more milk than tea.

I'm sure it was darker in real life...


Return to “DCC Forum”

Who is online

Users browsing this forum: No registered users and 4 guests