Stay Alive - 470uF Caps Useless?

[Sir TophamHatt]

Just double checking, the 470uF caps are useless right?

As in, the power they give is so momentary it's carely worth it.

 

If I wanted a non sound loco to travel a couple of inches, I guess I'd need more power?

[tender]

1 hour ago, Sir TophamHatt said:

 

 

If I wanted a non sound loco to travel a couple of inches, I guess I'd need more power?

Well, yes, but why would you want a loco to run uncontrolled for a few inches. If you're having that much trouble you should look at the contacts on the loco, clean the wheels/track etc. Large stay alive's just cover up problems that should be fairly easy to resolve (except for short wheelbase  0-4-0 locos on insulfrog points maybe).

Edited June 12, 2020 by tender

[34theletterbetweenB&D]

6 hours ago, tender said:

... but why would you want a loco to run uncontrolled for a few inches....

On which note, I have often wondered just what problem Lenz were aiming at overcoming with their USP system, which has the capability to maintain control while insulated from the rails. (The demo for those who haven't seen it is to run a USP decoder fitted loco fully onto a sheet of paper, stop it, and reverse off.)

 

I was quite attracted to this when new to DCC, but never found an application for it; admittedly my smallest mechanisms are 0-6-0T, all wheels collecting, live crossings throughout on the layout.

[JimFin]

6 hours ago, tender said:

clean the wheels/track etc. 

Must confess to having one loco fitted with a huge stay alive and use it for pushing the rail cleaner round at the start of a session. 

[Edwin_m]

If I remember my physics rightly, 470uF at 12V stores 5 milli-coulombs, which would power a small (100mA) motor for about 1/20 of a second.  

 

If the rails are really grubby to the extent they are insulating the wheels then I guess the Lenz system would allow control to be maintained on a loco fitted with stay-alive or with a lot of mechanical momentum.  

[Nigelcliffe]

11 minutes ago, Edwin_m said:

If I remember my physics rightly, 470uF at 12V stores 5 milli-coulombs, which would power a small (100mA) motor for about 1/20 of a second.  

 

If the rails are really grubby to the extent they are insulating the wheels then I guess the Lenz system would allow control to be maintained on a loco fitted with stay-alive or with a lot of mechanical momentum.  

 

Lenz' implementation is smart - if it can sense the rail signal is present, then it continues to drive the loco.  If it looses the rail signal, it stops the model.   So, a model can run on very grubby track (the bit of paper demo is the extreme form), but if you take it off the track, or kill the track power, the loco stops.    So, it won't go all over the scenery - no Titfield Thunderbolt rides over fields !   The downside is the Lenz power module is fairly bulky, so might not fit small models where it could have the greatest useful effect. 

 

The advantage of lots of energy store is that the voltage won't deplete significantly when it is called upon, and that situations of high current draw (eg. loco just starting off) can be supported.   But, that risks a loco which can run for several feet over the scenery, jamming the mechanism before crashing to the floor.    The alternative to Lenz' solution is to have a DCC signal time-out in the decoder.  ESU and Zimo (and perhaps some others) have this - a CV setting which says the stay-alive only runs for, say, 0.5 second.  Enough to deal with a pickup problem, not enough to plough a field.  

 

 

 

[Sir TophamHatt]

10 hours ago, tender said:

Well, yes, but why would you want a loco to run uncontrolled for a few inches.

 

Why do you think it would be completely uncontrolled for a few inches? It may only be a few millimetres but by that time, it's stuttered and slowed.

 

Because for me and my current layout, the inch or two will do exactly as you say, cover a multitude of problems, which some are quite difficult to solve.

 

Nearly all my locos are fine without stay alives, even the shortest of wheel bases over insulfrogs (of which I directly power all rails apart from the moving part).

 

But older stock, of which I have a lot of, is hugely benefitted from having some sort of stay alive. No amount of cleaning will change that and I'm not particularly in the market for adding extra pickups, which I see as more faff than adding a stay alive.

Edited June 12, 2020 by Sir TophamHatt

[Sir TophamHatt]

Even more fun is I did make a huge capacity stay alive for the Lima Class 156.

 

It must have been over a year ago now but it worked pretty well. About 20 seconds of movement with no power. But I think I had to go big or not at all as the caps needed to be wired in a certain way. If I had less of them, they wouldn't accept the track voltage and boom!

 

But like I say, all this was ages ago and I couldn't remember whether it was these tiny ones that are often supplied, which are really not that great.

[Sir TophamHatt]

2 hours ago, Nigelcliffe said:

The alternative to Lenz' solution is to have a DCC signal time-out in the decoder.  ESU and Zimo (and perhaps some others) have this - a CV setting which says the stay-alive only runs for, say, 0.5 second.

 

I've read a bit about this.

Do you know what CV to change?

 

I have a Zimo MX632, which comes with the 470uF stay alive. I have a Zimo SC68, but it says it's suitable for MX633 - but I'd have thought surely the MX632 would work with it?

Not sure whether I should try it. I think the last time I tried the SC68, it didn't work. Whether this is a faulty item, incorrect CV or simply incompatibility with the chip, I don't know.

[nswgr1855]

On 13/06/2020 at 05:59, Sir TophamHatt said:

 

I've read a bit about this.

Do you know what CV to change?

 

I have a Zimo MX632, which comes with the 470uF stay alive. I have a Zimo SC68, but it says it's suitable for MX633 - but I'd have thought surely the MX632 would work with it?

Not sure whether I should try it. I think the last time I tried the SC68, it didn't work. Whether this is a faulty item, incorrect CV or simply incompatibility with the chip, I don't know.

 

There is a section in the Zimo manual that talks about storage capacitors. It shows a circuit that is used to limit start up current and protects the capacitors from over voltage. I typically use a bank of 6  2.7V 1F capacitors on my fleet of old brass H0 locomotives. I have ignored the maximum capacitor value in the instructions and have had no problems. Of course you can limit the time the model runs, again the CV is in the Zimo instructions, I have never used it.  Much easier to fit capicators in most cases then to fit extra pickups, which can easily cause short circuits and require replacing after about 20 years of running and regular maintenance to remove lint. Also pickups on non driven axles cause drag that limits maximum train lengths considerably. For those worried about trains running uncontrolled leaving the rails onto scenery, fix up your track work and safe working procedures.

[jamespetts]

On 12/06/2020 at 08:52, Sir TophamHatt said:

Just double checking, the 470uF caps are useless right?

As in, the power they give is so momentary it's carely worth it.

 

If I wanted a non sound loco to travel a couple of inches, I guess I'd need more power?

 

I find that, for motorised models, 1,000µF makes a difference in N gauge, and 2,000µF is ideal. Most of my N gauge models are fitted with 4x 470µF tantalum capacitors, which is 1,880µF, which helps a great deal.

 

For 00 gauge, aim for 3,000µF. You ideally want 1-2 seconds of running for maximum reliability. More than that is unnecessary and risks runaway problems in fault conditions (although bear in mind that running time loaded will be less than running time unloaded).

 

For carriage lighting, much lower amounts are fine: 220µF is ideal for most carriage lighting situations.

 

Edit Incidentally, you can always wire multiple capacitors in parallel, so the supplied 470µF capacitor does not need to go to waste. Just make sure that you use one of the correct voltage: 25v is ideal, but you can use 16v if you have a charging circuit that limits voltage (check the manual to be sure).

Edited June 21, 2020 by jamespetts

[Grovenor]

Time out is CV11 and is highly recommended so most decoders should have it.

See https://www.nmra.org/sites/default/files/standards/sandrp/pdf/s-9.2.2_decoder_cvs_2012.07.pdf

[tractor_37260]

On 12/06/2020 at 08:52, Sir TophamHatt said:

Just double checking, the 470uF caps are useless right?

As in, the power they give is so momentary it's carely worth it.

 

If I wanted a non sound loco to travel a couple of inches, I guess I'd need more power?

 

Space permitting, I find a 2,200uF cap works well with my Zimo 645R decoders, 470uF not being of much use.

 

Edited June 22, 2020 by tractor_37260
correct text

[jonnyuk]

i have 3 loco's fitted with stay alive, Hornby m7, Hornby terrier and Heljan 1370, all fitted with 3 470's, they run brilliantly and as two of them are for shunting it really does make them a pleasure to run (i use insulfrog points and double slips). in my opinion if you can fit them into your loco, do it.