Kit: MX645PluX22 with ADAPLU booster board; YLR603010 speaker; SuperCap17000uF
Out of the box, the Hattons/Heljan LNER A3 in O gauge looks excellent - lots of detail on the plastic body, but plenty of
weight, and fine motion gear. No doubt this will be a popular model.
From the factory, the model comes fitted with an ESU LokSound 'L' socket in the tender. There are pickups on the loco and
tender (big improvement over the Mogul), and fair provision for speaker both in the tender and in the loco body. A 6-pin
connector joins loco to tender providing 2 wires for pickups, 2 for motor, and 2 further wires intended for connecting a
speaker in the main body, which you may or may not decide to take advantage of (you could use them to fit lamps or firebox
light instead if you preferred that).
In order to convert this to Zimo DCC Sound, we need to remove the ESU 'L' socket, and replace it with something else. Measuring
the current consumption was a little worrying on DC, which came back on our rolling road as 1.5 to 2A stall, and medium speed
running around 0.75A once settled down. It is a heavy model though and will probably reduce after a running-in period.
What this does mean though, is that a simple OO sound decoder (like the MX645R) will be operating close to its' limits, so to
operate safely we have to consider the next level of power up. That means one of Zimo's booster adapter boards. We decide on
the ADAPLU booster with MX645PluX22, which gives us 1.8A continuous output to the motor and stall current up to around 2 to
2.5A for short periods. Another alternative would have been the MX644D+ADAMTC, but we prefer the solidness of the PluX22
For this part of our example guide, we go for the simplest kit, which will all fit in the tender, and prove a quick and simple
We desolder the model's wires from the 'L' board, noting that Heljan (for a change) have actually used standard DCC colours for
the wires (horrah!). This will make re-connection much easier as we can see what is what.
The model's wires are fairly heavy-duty, and the pickups (RED and BLACK) are paired - joining the pickups between loco and
tender. These are biit too chunky for us to safely solder to the ADAPLU booster board, so we join them to an new length of our
own wire so we can do a neater connection to the ADAPLU board.
Connection is quite simple, with some neat soldering of pickups (RED and BLACK), motor (GREY and ORANGE), and speaker (PURPLE -
note that the speaker wires in the model are BROWN, which is actually the DCC standard colour - Zimo use PURPLE for speaker,
but the rest of the universe use BROWN!).
We also add a BLUE and another GREY wire for our SuperCap17000uF for stay-alive. This connects to the ADAPLU's ELKO-Plus
(capacitor +VE) and MASSE (Ground) pads. The SuperCap is wrapped completely in Kapton tape (as its' body is part of the GROUND
connection). We then protect the underside of the ADAPLU with Kapton too, and use BlackTack to suspend the SuperCap off the
The SuperCap and ADAPLU board fit very neatly down in the well of the tender weight, secured with BlackTack, and we can join in
the wires from the model.
We got the orange and grey wires the wrong way around initially, and the loco ran in reverse to what was expected, so we just
swapped them around on the ADAPLU board.
A YLR603010 speaker is placed above, where the 'L' socket was, secured with BlackTack, facing downwards, so the sound travels
out through the opening underneath.
That's it! We put the tender top back on, connect to the loco and test. The YLR603010 actually works really well in the
tender, with plenty of volume due to the way the sound can escape. Of course, the sound audibly comes from the tender rather
than the loco, but this is a simple approach to get you up and running with Zimo sound quickly.
Things to watch out for...
Worth noting some of the 'gotchas' we encountered with this model, as we had a few issues with our model - nothing to do with
First thing is that the holding pins for the rods were very loose when it arrived, and after a few minutes of running-in, some
of them worked so loose that the rods jammed. It took ages to find the cause, but once tightened, that aspect seemed good.
Next issue is with the metal connector bar that joins the tender to the loco - it is quite small, and the screws that hold it
in are tiny, threading into plastic, but not very deeply. We found that after disconnecting and reconnecting only a handful of
times, the screw into the loco would no longer hold. We used a slightly chunkier and longer screw to fix it, so not a big
deal, but rather disappointed that this wasn't more solidly designed.
Next issue (or rather, set of issues), is not really a fault, but rather a consequence of the model having a plastic body with
a lot of plastic detail parts - bits fall off and break very easily during handling. Easily fixed on the whole, but I guess
this has to be expected because of the materials used. Of course, the up-side is that the model is far FAR cheaper to purchase
than an all-metal equivalent! Watch out in particular for the hand-rail connections onto the boiler, the brake shoes, the
whistle, and the brake rodding on the tender.
Kit: MX696KS; YLR603010 speaker (in tender); Modified YLR453520 (in boiler); SuperCap17000uF
The approach in the above example works well, and the volume from the YLR603010 in the tender is good. However, a little more
volume is always useful (the mechanism is quite noisy above 1/3rd speed), and it would be good to shift the apparent location
of the sound forward from the tender a bit. So how to achieve that?
The obvious change is to take the speaker out of the tender and put it under the chimney instead, making use of the 'spare'
pair of wires that the model provides for this purpose.
The loco is designed in a clever modular fashion, so comes apart much more easily than Heljan's Mogul or Prairie models. It is
best to take the cab off first, which is done with 3 screws underneath. This gives us protection of the whistle later when we
take the boiler off, and also gives us access to the wiring, which is useful when re-assembling later too.
In order to remove the boiler, we have to take off the front bogie, then we just have 2 screws - one at the rear of the keeper
plate, and another which was underneath the front bogie. With these removed, the boiler simply lifts upwards. It may require
a little encouragement because the motor's terminal connectors slot into neat electrical connectors - very nice idea, as this
avoids wiring issues.
The motor sits in a heavy metal frame which goes the length of the boiler. It simply slides backwards out of the plastic shell.
Be ultra-careful handling the boiler once it is removed from the chassis though - the plastic moulding has 2 fine stays at the
front which are very easy to catch and break. It would have been better to have those parts integral with the frame, but as
long as we know about them, we should be able to take care.
The metal assembly from inside the boiler is flat toward the front - plenty of space to put more kit. The YLR603010 fits
easily in here, secured with BlackTack. We note that the chimney is blocked-off, so we open this up by drilling, then
carefully filing it out to the maximum diameter - this is the only place where sound will be able to escape, so it important.
On testing this approach, we find that it is only 'adequate' in terms of audio output - the speaker is muffled considerably due
to having very little opportunity to escape. It is much quieter than the same speaker in the tender. There doesn't appear to
be any obvious options to create more openings either. Therefore we look at other types of speakers, including the EM Tang
Band range (the medium sized one fits nicely, but doesn't perform as well as the YLR603010).
The next size up is our YLR453520, but there's no way this is going to fit with its' box. However, the speaker component itself
it quite impressive so we decide to try an experiment... chop the speaker out of the supplied box and create our own enclosure
from plasticard to fit the space!
This gives us a box measuring about 20mm height and a tapered shape toward the front. You can see that we cobbled this
together, wihch is why it isn't very neat, but it is sealed, and the performance is virtually the same as having it in the
original 45x35x20 box.
All assembled, testing it inside the boiler is still quite muffled, with the only opening being the chimney, but it is a little
better than the YLR603010 otpion so we stick with that.
Now that's where we have to stop if we persist with the MX645PluX22+ADAPLU board, since the maximum audio output is 3W. If we
want to use 2 speakers above 3W total then we must upgrade the decoder to a proper large-scale one. Enter the MX696KS...
The MX696KS is almost identical in size to the original ESU 'L' board, so we know that could fit. We could put a small speaker
in the gap in the tender weight, and it would be quite easy to do. However, we liked the YLR603010 so much, that we decide to
go with an approach that alllowed us to use the YLR603010 AND the customised YLR4353520 in the boiler. NOTE: the MX696S is
shown here with the solder-version of its' adapter plate, rather than the screw-terminal version, but they amount to the same
thing - it is more common to use the screw-terminal version called 'MX696KS'.
We choose to put the YLR603010 over the opening in the metal, face-down, to allow the best sound escape, and therefore the
MX696KS must be positioned forward of that. The plastic slope from the coal shute comes down into this area, so we have to cut
that plastic to make space for the decoder. This plastic part comes out fairly easily and we can modify it away from the main
It may seem severe, but it is not difficult, and the changes will be hidden by the coal load anyway - the model is supplied
with a plastic coal moulding which fits nicely, though I suspect that most people will prefer to build their own loads using
real coal or similar.
With the SuperCap17000uF protected and placed into the hole, speaker on top, and decoder to the front, all secured with
BlackTack and protected with Kapton Tape, the tender can be reassembled and tested.
The result is very good - the pair of speakers work together well, giving plenty of volume to drown out the mechanism even at
speed. The sound does still give the impression of coming from the tender end more than the front, but it is certainly better
than only having one speaker in the tender. The larger decoder is also running more within its' ability, so motor control is a
little better too, so overall, very pleased.