HM7000 Hornby's implementation of "stay alive".

[ColinB]

Now I read a lot from all the posts about the way Hornby has implemented "stay alive" on this decoder. From what I can gather reading a ton of posts, Hornby are driving the charging circuit via a port on the microprocessor as opposed to the normal method. So for all you techs out there, is it true?

[LTSR_NSE]

It is categorically true that it is not charged by normal pickups method. Charging is controlled by firmware therefore IC.

I’ll leave further specifics to actual electronics experts (who’ve gone beyond KS3!) 😉

[Fishmanoz]

Is the HM7000 microprocessor involved in PB charging (slight reframe of Colin’s question)”. Most certainly yes.

How do I know? RTFM page 120, Smart Charging, plus V2 Sound Profile feature to allow for charging when stationary.

[ColinB]

Great thanks for the info, I will read the manual, I had a long think and worked out how they do it. So it means you don't have all those programming issues you get with a normal circuit. The reason I ask is on my older ringfield based locos I want to add lots of "stay alive" having seen what can be achieved by Accurascale, which will compensate for the lack of pickups and mean they run smoother.

[ColinB]

Well I read the entry, surprisingly very little but you have told me enough. It does mean I can make my own out of a load of tantalum capacitors. I don't why they limit it to only when the loco is running, if they wanted to stop it charging during during a programming session just use a timer.

[Fishmanoz]

Which is what V2 sound profiles do.

[Rallymatt]

I am absolutely not an expert in this area, but sure I read somewhere that you can’t just use a standard ‘stay alive’ set up on the HM DCC decoder, there is more to it than just the capacitors?

I only mention it for sake of someone not damaging a decoder if they overlook something.

[ColinB]

I know Fishmanoz, but the wonderful document actually says that. They obviously didn't get round to changing it.

From what I can see Rallymatt it is just a bank of capacitors, there may well be a series resistor but with only a two wire input it can't be too complicated. I assume this must the technique Zimo use as I have never had an issue reading CVs with their "stay alive". You probably could make you own by using one of the function outputs of the decoder, the only issue being that you would have to manually turn it on with a function command.

[SteveM6]

This thread just seems to be rehashing debates that have already been documented on this forum.

The HM7K decoders have circuitry within the decoder that is undoubtedly related to the charging protocols for the Powerbank (I don’t know or care how they do it). However the PB can be used on its own when soldered directly to a TTS decoder (I have done it without it going bang)which suggests the circuit in the PB is nothing special. It may react differently with other brands though.

It may be possible to build your own stayalive to plug into a HM7K decoder but I don’t see why anyone would try given their size and price.

[ColinB]

The reason I asked was because when ever I add "stay alive" I cannot read back CV values. Now I know you don't have issues but I do. So I was looking at HM7000 for that reason. The Powerbank last time I looked was about £16.00 and it is huge, the last pack of tantalums I bought was just under £2.00 and I fitted them to a strip of veroboard. I used tantalums as the space in a P2 loco is so limited a standard electrolytic won't fit. I don't intend attacking my new loco with a Dremel as some do. Anyway because of my previous profession, capacitors are not something I am short of. You guys have told me the information that I wanted, which I am grateful for.

 

[SteveM6]

This may, or may not, help.

I was surprised with my P2’s lack of space in the tender. Having gone to he trouble of designing the interior to accommodate the speaker and decoder, I would have thought it relatively easy to provide a void for the power bank.

I have fed that back to the project team and suggested that the sloping bit in front of the speaker would have been a good place to start.

In the end I worked out a route for the standard wires through the chassis and up through the coal hole in the tender - no cutting or drilling needed and the wires disappear at normal viewing distances.

The coal load still fits over the PB although I did add a coating of real coal before refitting it.

[SteveM6]

Colin, I do have a solution but I’m unable to post it while I’m at work. I will add a description and photo later.

[LTSR_NSE]

@Colin are you intending to make diy stay-alives for 3rd party decoders or HM7k ones?

If HM7k - it might be worth purchasing one, removing the shrink-wrap & identifying/proving the circuitry it contains. I believe Hornby have stated that using a non-HPB stay-alive would invalidate HM7k warranty - which suggests that being certain is vital.

[SteveM6]

Colin, as promised.

I found the same issue with my Thane of Fife. I have suggested to Hornby that they design in a void to accommodate the power bank - the sloping bit in front of the speaker housing would seem a viable space.

I found a route through existing holes in the chassis to feed the PB wires under the chassis and back into the coal space through the shovel hole. The plastic coal load still fits and covers t

[SteveM6]

I’m having real problems trying to post a photo (apologies to Mods if you get many versions filtering through the ether!)

Colin, my solution to the problem of no space for a PB in the new P2:

There is a route for PB wires through to the underneath of the tender chassis that allows you to route them up between loco and tender and into the coal space through the moulded ‘coal-hole’ into the void under the fake coal load.

The PB fits there with enough space for the coal to clip back in and hide it. The only modification is to cut the PB wires (the plug won’t fit through the existing holes) and resolver and add some heatshrink.

Using this method no cutting or drilling of the loco is needed and whilst the wires are in the open they are virtually invisible at normal viewing distance.

[Fishmanoz]

I know Fishmanoz, but the wonderful document actually says that. They obviously didn't get round to changing it.

 

 

which is why I referred you to the manual and to the relevant change made in the V2 profiles. Especially as these together confirm microprocessor control (that doesn’t mean direct connection Daedalus).

[ColinB]

Daedalus I know that, the polarity and voltages are all wrong anyway. It depends on the micro they use but most outputs are open collector so in my crude circuit you connect the output port in the micro to the base of a PNP transistor via a resistor that does the actual charging. I did draw the circuit out just to figure out whether it would work. I am sure there are other ways to do it. That is why I said you could make your own was using a function output (in microprocessor terms an output port), so say the second function with the flying lead on an 8 pin device. Either way I was just interested because I see the advantages of decent "stay alive" in my Accurascale class 37 but a lot of the systems I have used previously have given issues when reprogramming.

[96RAF]

Colin is essentially correct.

As previously stated in other discussions about the Hornby PB, it contains components that are triggered by the decoder. The charge comes from the decoder 12 volt rail and the ‘charge now’ trigger comes from the MCU.

Initially this trigger was set to occur only once the loco was in motion, but following reports of stuttering, this has now been set to occur after a 1.5 second delay after the decoder is powered.

Why the delay - it is to cater for the service track programming case where the short programming burst time will never exceed the delay time, hence there is not the programming problem that has been seen before on some decoders with a stay alive installed.

[SteveM6]

Well, somehow a picture got through and is shown above.

Colin now has a viable solution to fitting a PB to a P2 without cutting up the model and without the need to know how Hornby have facilitated control of the charging process.

I know I have said this before but having now fitted seventeen power banks to locos from three different manufacturers and ranging from tender locos to small tank locos, I have yet to find one that won't accommodate the standard PB - but you do have to be creative to find some space on occasion.

[SteveM6]

As I understand it, in programming mode only a small current is applied to the rails so that when a loco with a stay alive is placed on the track it effectively draws all the current for charging if there is no delay built in.

The delay therefore allows the decoder access to the reduced power to avoid the programming issues sometimes seen.

I believe that the programming issues can be overcome by programming on main which applies a higher voltage or by unplugging the capitors - my TTS locos wiTh stayalives are all connected via a micro plug for that very purpose.