Paul-365510

fitted the traction tyres (fiddly) but it totally solves the rocking problem. I think that's the end of this thread. all my current problems are solved. thank you all for the patient advice.

Hi 96RAF, reloading the sound profiles does indeed seem to have sorted out the hiccups with the HM7000. Does this mean that Hornby found the issue and fixed it? I would be relieved to know that this issue was nothing to do with my inexperience with decoders. Also delighted to have the same low speed control that I was seeing with the LaisDCC but with sound! Perhaps I didn't need to fork out so much money on the Loksound decoder for my Mallard, which was DoA in any case, and I'm awaiting a replacement from the vendor. But it will be good to have experience with 3 different decoder brands. Thanks everyone for the help with this. Cheers.

oh wow. thanks. I was given the train by a friend whose son wasn't interested, it never had the traction tyres (dunno why) but I can see where they would go. You've saved me from mutilating a train (or a 15+ points). I've ordered pack. thanks again.

thanks Going Spare but there are no traction tyres on the tender. the wheels on the tender have a small but noticeably deeper flange than the wheels on the loco. The second and last pair of wheels float but the first and third pair are the fixed driving wheels and when viewed from the side it's quite easy to see the wheel hitting the insulated part of the frog lifting very slightly and rocking the tender as it does so. but it's an old FS (15+ years) running on new points, so maybe there have been some subtle changes? It's only the wheels hitting the frog that lift, the channel in the outer guides is the full depth of the rail and causes no problems.

I will try that, thanks 96RAF.

I checked the back-to-back distance of the wheels on the tender (with calipers) and they are the same as the loco, carriages. Getting out my glasses and pushing the tender slowly over a set of points on a bench at eye level I can see that the issue is that the flange of the front and rear wheels on the tender is deeper than on the loco and when it hits the frog it is rising up slightly because the groove in the frog isn't quite deep enough for the flange. So I could (a) get out my dremel and deepen the slots on every point..., I think not, or (b) grind the flanges down slightly on the driving wheels of the tender which feels a little brutal and there's no way back from there. Or © do nothing and just let it wobble as it goes through the points. Advice welcome. This can't be the only time that this issue has been observed.

This is cut and paste from the Hornby guys response: "The decoders need to be covered in electrical insulation tape, however they will need some space to breath so do not cover them fully." I had just assumed that since the (8 pin) LaisDcc and Loksound decoders come in a plastic sleeve, there could only be some benefit in likewise covering the HM7000s. I shall check the FS tender wheels back-to-back. Once I've looked up what that is :) Thanks.

further update. Got a response from Hornby support on the decoder going AWOL on BT. I worked through some suggestions but nothing helped. Then some decoder harnesses arrived so I followed previous advice, put one into the FS that was the subject of the original issue, and then swapped the naughty HM7000 to that and added a stay-alive. It's all working great now - Bluetooth, sound and all. Doesn't explain why it was unhappy in the Mallard (but I've ordered a Loksound for that one now) but it does demonstrate again that the HM7000's can be badly behaved without a stay-alive. Also I asked in my support request about whether the HM7000s needed wrapped in insulating tape and the answer was "yes, but leave some uncovered so it can breathe". A somewhat ambiguous answer, but what I'm doing now is wrapping them, installing them with a blob of hot glue, and then cutting away the tape from most of the exposed top. I have another issue, should I open a seperate thread for it? I've noticed that the FS tender rocks as it goes over points - the wheels don't seem entirely happy with the frogs. Seems odd given it's a Hornby train on a Hornby point, but I can't be the only one who has noticed it. It's only the tender, not the loco or the carriages.

I've read chunks of the manual, and in particular the details around auto-calibration, and it was that process I was attempting to apply. But I've not gone through all 130 pages :) The trains are all good runners and had been working perfectly on DC for over a year, so we can definitely rule out any problems with dirty track etc. (and I've numerous soldered power connections). The FS runs beautifully without sound with an LaisDCC decoder, without stay-alive, right down to creeping at 5mph, but does not run at all with the HM7000. The 0-6-0 and 0-4-0 tanks are working well with the HM7000 as long as I install a stay-alive, without the stay-alive they just twitch and reset after a few mm of movement, but they work fine with the LaisDCC decoders and no stay-alive. But I fear I'm taking up everyone's time here, and you may be concluding that the problem is me. I don't think it is but I'm tempted to conclude that putting an HM7000 into a pre-DCC loco is a hit and miss affair, and that even in a DCC ready Hornby loco, it's not certain to work without tweaking. I think maybe I should try an old Hornby TTS, or a modern Bachman, Dapol, Loksound or Zimo. Would anyone like to to recommend the decoder they think is most likely to work, even for me?

Fishmanoz ...er no. I didn't see that option, just the general support email address. I'll wait for a response, hopefully they'll figure out the best person to deal with it. 96RAF - it moves with the LaisDCC decoder inserted (but in fits and starts, maybe needs a stay-alive) so I don't think the motor itself is jammed, but maybe you meant something else. But I'll wait for Hornby before thrashing around any more.

ok repeated unlink/deletes did eventually delete it. Thanks for the suggestion. I could then find it as a resettable device. it took several goes before the reset made it linkable again (and you have to kill the app after every reset or it always reappears as resettable rather than linkable). But even after linking to it again, which you might assume meant it was using Bluetooth, it is still not showing up as connecting on Bluetooth and controllable from the app, only from DCC. And on DCC when I throttle up it just makes a worrying buzzing noise and doesn't move. I've opened a ticket with Hornby...

unfortunately you can't delete it, if you can't connect to it over BT. or at least I can't find any way to do that - unlink and delete fails because it's not connected.

further update. Got my hands on a second-hand (but mint) DCC ready Hornby Mallard. Stuck in an HM7000, did the updates, and loaded the A4 profile. No stay-alive at this point. It didn't move much - just twitched when throttled up. a-hah! I thought. Time to try the auto-calibration previously recommended. Following the manual's recommendations I did a reset first. The app then wanted to update Bluetooth (again) so I let it. The app sat for a few minutes on zero progress on the status bar (which was odd), and then displayed the "remove from track to power cycle" screen. Which I did. Well it seems the BT upgrade failed and now there is no BT at all so I can't use the app with this decoder. I tried using DCC to do a reset, which seemed to work for the CV values but doesn't help with Bluetooth. I tried the auto-cal i.e. putting zero into 149, clearing the F0 function etc. Put on the track, hit F0 and it moved a little bit a few times and then stopped. A stay-alive may help get it moving (as it did with the J83) but I don't know how to get Bluetooth support back. The manual doesn't cover this, but does anyone know how to do a full factory reset on the HM7000 decoder to restore the old Bluetooth firmware? or have I just blown another £60 :( I can try contacting Hornby support if that's the last resort.

great info 96RAF. I'll try that when I get another HM7000 and harnesses (all ordered). Yes I was aware that the LaisDCC were cheap knock-offs. I have one because it was in a second hand engine I bought. But cheap and cheerful though it may be, it does work smoothly (in this instance anyway), and it's small and comes pre-wrapped. I don't understand why others come naked and you have to get out the insulating tape...

I'm loathe to unsolder the LiasDCC - the train is running so smoothly at incredibly low speeds, but maybe I'll install an 8 pin harness to make experiments a little simpler, although there's not that much room in the tender - I should maybe put the harness into the engine, it's presumably empty space in there. I'll try it at some point and report back. thanks.