Hi does anyone know how to fit a stay alive to a DZ123 decoder please

[peteski]

Hi thanks for the info, I did as suggested, measured the voltage on the series caps (including the resistor, diode combination) when powered on track and it read 12.8v (which my track voltage from my NCE SB5 Smart Booster) and then lifted the loco off the track and measured again and it ready 10.2v ?? not really sure what's happening there unless its dropping the 2v across the diode and then doesn't have enough voltage to drive the loco anymore? any suggestions would be great please? I assume if I take the diode out of the circuit the caps will discharge through the 47ohm resistor and still drop voltage across it? cheers Kevin

What you see seems normal.  It is backwards  of what you described. The caps charge at a slower rate through the 47 ohm resistor, and once the track power is removed, the caps supply the power (discharge) through the diode providing power to the decoder and rest of the loco.

When you were taking the measurements did you have your meter's leads connected throughout the entire period, or you the test leads across the caps while they were powered and charging, then disconnected the leads to take the model off the track and then hooked them up again to measure the cap voltage?  Basically I'm surprised that the instant the charging voltage is removed, the voltage across the caps drops from 12.8 to 10.2V. 

If you leave the meter connected, do you see the voltage continuing to drop from that 10.8V?

As for the 1N400X diode, it will drop about 0.7V. I prefer using Shottky diodes because the drop less voltage. Usually 0.5V but often drop even less.  Decoders often utilize Shottky diodes in their rectifiers to minimize voltage drop.

But even if the 10.8V was measured at the output of the keep-alive (0ast the resistor/diode), to me that should be plenty high for the decoder to stay powered up.  If it doesn't keep on running at 10.8V then maybe it is designed to stop the motor when it stops receiving DCC packets.

Did you try the keep-alive on both of the decoders you asked about?

[trainnutt]

Hi thanks for getting back to me, yes I tried in both decoder types, I have just retested as you suggested leaving connections made while lifting off the track.
the initial voltage shown across the 3 x 0.047f 5.5v capacitors is 12.7vdc when I lift off the track the voltage steps down to 11.9 then down in 0.1vdc steps, the motor stops running instantly the loco is lifted from the track so there is no stay alive being provided to the motor. its very odd as other videos I have seen do provide adequate cover for 1 or even 5 secs sometimes. I'm wondering if the capacitors are the reason? or is it something in the decoder cv's I need to change? I have a laiddcc stay alive unit on way so will test the same decoder with that once I get it but its a little annoying as I was hoping to save some time and money making my own.
cheers.

[peteski]

As I mentioned earlier, some decoders are designed to stop the motor as soon as soon as they stop receiving DCC packets fro m the track (regardless if they are still fully powered).  Neither of the decoders you are trying was ever designed for keep-alive capability.

Testing them with a "real" keep-alive will be a good thing to try.

Some decoders which accommodate keep-alives do have a CV which determines how many seconds the decoder will keep the motor powered after loss of DCC packets, but I am unaware of Digitrax having that CV defined.

One thing you might try (if you haven't  already) is to disable the DC operation (bit 2=0) in CV29 then check if the keep-alive will do its job.

[Sumner]

.... the 3 x 0.047f 5.5v capacitors...

First of all I get screwed up in converting from f to uf but using an online calculator it indicates that .047f = 47000 uF.  That doesn't seem right?  The Iowa Scale Engineering 'Power Keeper' is 940uF (2 @ 470uF) and I think the TCN smaller one might be similar.  What is the uF reading for the capacitors you are using?

Also I'm assuming you are running those in series since they are 5.5v so the overall capacitance is only the value of one of them, not all three added together.  That could make a big difference.

I made up a couple using Iowa Scale Engineering's wiring diagram which I think is needed unless you are running caps with a lot higher voltage rating than the track voltage to protect them.  I'm running 12v for the track power.







For the cost difference in buying parts and making it if the Iowa Scale Engineering or TCS fits that is probably a better way to go 

Disclaimer:  I built this and another 'kind of keep alive' but haven't really run them yet so can't report on the effectiveness of them.

Sumner

[peteski]

That 100 ohm 0805 resistor might be a bit under its wattage rating for this circuit.  a 1206 size would be more suitable since the specs recommend a resistor rated at 0.5W and power rating of a 0805 resistor is 0.1W

https://www.westfloridacomponents.com/surface-mount-resistor-sizes.html

[Sumner]

That 100 ohm 0805 resistor might be a bit under its wattage rating for this circuit.....

Thanks, I'll keep that in mind.  It is what Iowa Scale Engineering uses ( HERE ) unless they don't anymore....

.... and the spec sheet for them ( HERE )  says 1/2 watt


Sumner

[trainnutt]

Thanks everyone for your input I am awaiting some laisdcc stay olives similar to the one I know works in a previously mentioned loco so will report back when I have them installed and tested.
cheers.

[peteski]

Thanks, I'll keep that in mind.  It is what Iowa Scale Engineering uses ( HERE ) unless they don't anymore....

.... and the spec sheet for them ( HERE )  says 1/2 watt


Sumner

Power dissipation rating is related to  the component's physical size. With continuous 0.5W it would be running rather hot, but in a keep-alive it will likely not be exposed to continuous operation (only when recharging the caps).

[trainnutt]

Hi all as a further follow up I have now received my laisdcc stay olives and fitted one to the 1st decoder (Hattons decoder) that I asked about, same result no impact at all to the operation of the loco or run on of motor once power removed from the wheels. I also tried on another of the same type of chip and same result, I then tried the stay alive on a decoder i know works with it and it performed well so I'm assuming the Hattons chip has a circuitry difference somewhere in the chip that is preventing the stay alive running, or could it be a cv setting somewhere and if so does anyone have any ideas how i find it?
cheers ,

[peteski]

If the instructions do not mention a specific CV or its setting then I would assume that the Hattons decoder's firmware (internal programming) is configured to stop the motor from running when it is not seeing the DCC packets coming from the track.  As I mentioned earlier, try using the keep alive with the DC bit in CV29 disabled and enabled (try both ways).

For example ESU decoders (designed to work with keep-alives) have CV113 called "Power Fail Bypass"  which specify the time decoder will keep running after the track power is lost.  I doubt that Hattons basic decoder will have that option.