Dapol DRS livery
Two superb looking models, with good lighting hardware. Just need a sound system and appropriate DCC lighting control (I hope).
Thanks to Kevin for providing his photo of 68016
Work has begun on the first loco:
The speaker space is limited (without a lot of milling). Standard 4 ohm bass reflex speakers seem to be unobtainable at present, so in the absence of Bif & Charlie's EM1, I've constructed a custom enclosure that just about fills the available body space, with a centrally mounted ESU 50334 20x40mm drive unit pointing downwards.
Sound decoder is Bif's LokSound V4. (Not aware of other options at this stage, and it has the useful drivelock feature). When he gets back from the Scotland show, I'll request a reblow for the unit I've installed
I want to include yard mode and main/dipped beam on the headlights. To keep it simple I've installed an additional LokPilot FX function decoder.
There is evidence of some last minute changes made to Dapol's main PCB, with a pair of unknown devices that both involve sky-hooked wires to the decoder ground pin. See below for more thoughts on this.
An apparent conflict in the DCC instructions that come with the locomotive:
Amongst the recommended decoders are 21 pin MTC ESU LokSound V4 and Zimo MX644D. The LokSound V4 21 pin decoder is available with either low current logic outputs for AUX 3 & 4 (ESU type 54499) or with full performance Aux 3 & 4 outputs as the slightly more expensive M4 type 64499. However the Zimo MX644D is only available with full performance Aux3 & 4 outputs (Although the still available MX644C does have Aux3 & 4 logic outputs).
My initial assumption was that the Dapol model must be configured assuming full performance Aux3 & 4 outputs, to enable either decoder type to be used. However there is some sky-hooked afterthought circuitry added to the main PCB, which looks very much like a pair of buffer MOSFETs connected to the Aux 3 and 4 outputs. If this is the case, then only the original ESU LokSound V4 type 54499 will work. The buffers act as inverters, which would reverse the sense of a full performance Aux 3 and 4 output, making the Zimo MX 644D (or M4 version of the LokSound) produce different results from those intended by Dapol.
This issue won't cause me any problems as I don't intend to use the Dapol decoder control circuitry. Instead I will design my own minor hardware mods to add additional main or dipped beam headlight functionality. A bit of an own goal for the Dapol documentation!
The locomotives both run smoothly in DC configuration, straight out of the box. So I don't anticipate any running problems when under DCC decoder control.
The Lighting Requirement:
The full size Class 68 seems to be the first in the U.K. to use a new lighting convention: Instead of the previous arrangement using separate day and night headlights, the new lighting convention uses both headlights simultaneously and includes a main and dipped beam option. So in outline, the requirement is as follows:
End 1 directional halo & top markers forward and rear lights in reverse. (Yard mode)
End 2 directional halo & top markers in reverse and rear lights forward. (Yard mode)
Headlights dipped beam / main beam. (Simulated by different intensities.)
Drivers cab light on. (I only illuminate the occupied cab.)
Parking lights (rear lights at both ends).
.......and hopefully Bif's synchronised output to drive a bogie mounted LED to flash with the detonator bangs.
I'll make use of Dapol's arrangements where I can, but rewire where necessary.
The most straightforward approach requires Bif's 21 pin LokSound V4 plus a LokPilot FX V4 function decoder.
The function distribution is as follows: (headlight and rear light share positive feeds on the lighting PCB at each end, so these are powered from a common decoder).
I've mapped the lighting to the controller buttons as follows:
F0 end one lights yard mode (halo plus upper marker)
F18 detonator flash (produced within Bif's sound project, if available)
F19 cab light (stationary only)
F20 end two lights yard mode (halo plus upper marker)
F21 headlights on
F22 headlights main beam (on) dipped beam (off)
F23 Parking lights
Resulting in a function table that looks like this:
Adding stay-alive capacitor for LokSound V4:
I've used my standard arrangement with zener diode protected charging voltage, to ensure that the super-cap is kept within its maximum voltage specification. The circuit now uses a single 15 Volt zener diode as Zimo have recently up-rated their 6800uF super-cap to 16 Volts maximum charge voltage.
I've removed two of the original switches, and rewired the remaining pair to provide individually switched track connections to each decoder.
This enables independent CV adjustments while I'm setting things up.
The custom speaker enclosure, shaped to make as much use as possible of the space available in the upper body shell.
Modified main board, function decoder and stay-alive super-cap.
Noch driver installed in the forward cab
Next I'll see if the lighting arrangements work.........................
Being able to switch the LokPilot decoder off quickly via the switch accessible through the roof, with the exhaust cover temporarily removed, works really well..... as this needs to be "off" during programming adjustments but "on" during lighting tests.
Good news...... The lighting works as intended and its looking about right..... except that the top marker and the halo lights seem a little on the bright side.... I'll try winding the intensity down via the ESU dimming function. Halos and top markers are fed by a common function output at each end. Dropping the halo & top marker drive intensity from 31 (max) to 12 works well!
NB: The photos shown below were deliberately taken with under-exposed settings to stop the camera sensor saturating and show the halo lights in action.... the lights all appear much brighter in reality!
Using the LokProgrammer to drive the loco, there was slight hesitation at the minimum speed step. This was eliminated using the following settings: K slow: 24; I slow: 2; K: 48; I: 90. (40kHz freq).
Bif's sound project:
Bif's sound project is now on board via the Lokprogrammer. Well up to Bif's usual high standards! I think my custom 20x40mm speaker is working at least as well as the intended slimmer bass reflex speakers (with a similar sized drive unit). However, the bass is a bit lacking compared to that possible with an EM2 speaker. Some substantial milling and consequent adhesion weight reduction would be needed to make the necessary room and unfortunately, I don't have a friendly milling facility available. However, I do plan to try a Zimo Jumbo twin sugar cube style speaker on the next 68 in case that gives better tonal results.
Bif kindly included his detonator flash LED drive signal on Aux 3 for me (absent from the standard sound project as it would conflict with the Bif 68 lighting arrangements). With my extra function decoder, I have been able to keep Aux3 on the LokSound free for this purpose. I'm fitting a white chip LED close to each front wheel on the driver end bogie. The two LEDs will be connected in series and then connected to the decoder via a 1k common series resistor.
Detonator flash LEDs fitted and working..... That confirms that the skyhooked Dapol circuit is indeed a buffer amp, almost certainly an N-Channel MOSFET.
Time to start work on 68005! Then I'll think about a video.
A few changes for 68005:
68005 will be my "number two" Class 68. This will be the second loco when double heading and the tail loco when top and tailing. The most obvious requirement is for a tension lock coupler at both ends.
Dapol have failed to respond to or acknowledge my enquiry via their website about purchasing a second coupler compatible slotted valance. Annoying! particularly as the accessories instructions indicate that an extra valance is provided..... which is not correct, as although a separate slotted valance comes in the accessory bag, there is no valance pre-fitted on the rear of the loco as it arrives.
Nil desperandum! I've carefully attacked the unslotted front valance (dam) with a few sharp instruments and successfully converted it into a slotted one. Just to add insult to injury, the spare coupler, which WAS provided, was sticking in the raised hook position, so not incredibly useful.... fortunately I have a few spares... sorted.
Next: repeat the circuit mods used on 68002. This 2nd loco will not be fitted with Detonator flash LEDs, so Bif's standard sound project will be fine. I'm looking forward to comparing the different speaker solutions described above. Listening to You Tube clips of real 68s, the lack of extreme bass because of the miniature speakers' limitations really does make the model sound rather unrealistic compared to the real locos. I hope the sugar cube technology at least slightly narrows the gap. Only other solution I can think of would be a bigger speaker with a good bass response, under the baseboard, driven by another synchronised Bif decoder!....... might try that!...... Hmm works rather well! More on this later!
Simplified Function Map:
Without the lighting operation decoder detail, here is the function list:
Class 68 Bass Boost:
A characteristic of the Class 68 is a deep bass background engine sound. This is not reproducible via traditional OO gauge speaker types (although an EM2 might get close, but requires milling of the chassis to fit.) So I decided to try plan B:
I had a single rather old LokSound V4 in my spares box. (Liberated from an earlier loco when replaced by a Zimo decoder, before the days of Drive Lock.) I have now used this to produce a stationary sound source fitted with Bif's 68 sound project, programmed in an identical manner to 68002. I've mounted the decoder on a plasticard base, with terminal blocks for the external interfaces and a couple of flashing LEDs driven by the decoder as a reminder that it is active.
The decoder drives two 4 ohm Hi-Fi speakers in series, which I have located centrally within the loft, to produce the background bass sound. The two speakers are just small Panasonic bookcase size speakers borrowed from a music centre, so nothing special.... but they do make quite a difference!
Initially I've set the address to be the same as 68002 so that the two decoders work together, however I may in future assign a different address so that I can use the PC to connect the two decoders in consist instead, enabling it to be used on either 68002 or 68005 without having to reprogram the address CV.
See what you think on the You Tube video via the link below.
To see the the 68s in action on You Tube, please click here