Brickthus

A few ideas here including standard track, points and flexi-track with 1M sleeper spacing. Mark

Well done Sheepo and Blakbird. This is good to see. It must use a lot of the motor power in the friction of the turntables. Could more motors be added to increase the speed? I have used train motors before. Looks like the blade roots are substantial enough to allow more speed without inducing a blade-off event! Mark

Finally got round to taking some pictures, as far as I got with articulation experiments. There is an axial compromise in the pivot positions, between optimising for gear engagement (no sliding) and optimising for wheel turn to keep non-flanged drivers over the rails. More pics here The parts on top ensure that the two pivots turn by the same amount at the same time. This allows 2-2-4 as shown here, as opposed to 4-2-4 being needed without it in order to have bogies at each end to keep the wheels on the track . I would change some parts to stop any movement from vibrations extending travel on axles and disconnecting the cogs from one another. Since moving on 2 axles would be needed in order for this to happen, there is more resistance than on some variants. It means 14-coupled engines should be possible, as long as they use small enough wheels for BBB wheels to fit the scale! Mark

The snag with using 9V train motors with PF (either the IR Receiver output or the LiPo battery output) is that the 9V train motors don't like PWM - they buzz a lot. Buzzing vibrates the motor core and will heat it up and shorten its life. The PF train motor is a new design that coped better with PWM; it whines but doesn't buzz. I have used a dual 30V 3A bench power supply for driving 9V trains. Plenty of current, enough to stop one train and start another by changing the points at the ends of through-sidings. I usually run 9V trains at 7 Volts for half an hour before changing, with a current limit of 2 Amps. I kept my 9V controller for the yard. What is wrong with your 9V controller? Mark

The minimum radius on a British main line is 250ft. Tracks in yards go tighter; some used to be as tight as 40ft radius, accompanied by a notice saying "only 0-4-0 engines beyond this point" because an 0-6-0 would derail. A book of the railways of Burton-on-Trent showed this in a picture. This was when the town used to have tracks down the main streets for moving grain and coal wagons about for the breweries, before the widespread use of 44 tonne lorries! I build as close as possible to 8mm:1ft scale, 1 stud to the foot. For my large layouts I use 104ft on the main line and 72ft on auxiliary lines. 40ft is only allowed in sidings. My previous layout. For small layouts with a selection of trains I use 72ft on the running lines and 40ft only in sidings. Small layout. A grand curve (of straights) would be nice for the main lines, and close to the 250ft minimum, but most of my trains are able to sustain adequate speed for exhibition running on 104ft curves. I haven't yet splashed out on ME track so I have 2 straights between each curve on the main lines and 1 straight between curves on other lines. This means oscillation is possible, though the longer bogie centre distance means the Pendolino works well with a 104ft curve. I have also modelled flexi-track to 72ft and 104ft curves, with ballast, slope and cant. All these things depend on the compromises of scale and selective compression. The key is to pick a set of them that fits together well - scale size, train length, vehicle dimensions, curve radius, power solution, running speed. A tighter constraint for non-flat LEGO layouts is that the maximum slope on a model railway of any scale (except rack railways) should be 1 in 30. That's 1 plate for every 12 studs long! This is the same constraint used by other model railways. It is also said that the difference between a train set and a model railway is the signalling, but this may be alleviated as more real railways move to ERTMS cab signalling. For scale I have settled on getting better than 'OO' gauge scaling (which is quite poor but accepted in the model railway fraternity) and as good as other model railways for prototypical operation. It is more important to have something running on your layout for the public to see than the number of rivets; we're not trying to make 'P8'! Mark

Thanks for the plug Basic train and 360 degree barrel roll from 2007. and . This was before we had the LiPo battery. The most difficult part was to get a full battery box of AAs out of the barrel roll.Glad it has taken a few years for a new, improved 5-wide track solution to be built! What Uefchen has done is simpler and lighter, very nice, and easy to get a long track round the room. It doesn't need such strong supports either, though no humans would want to go on a roller-coaster that flexed like that; already Alton Towers in the UK had an accident on the Smiler when one car ran into another! Just as the more pedantic people on YouTube said of mine "it's just a train", so it is difficult to get a train on these tracks to go any distance without on-board motor power as the friction is too high. This seems to go at a reasonable speed. I liked the larger sprockets better as they run less-noisily on the track. So some challenges for you Uefchen: I wonder if gearing it up a bit and having climb assistance from a fixed chain would make it more like a roller-coaster. This would mean the on-board motor power would be just enough to overcome the friction difference between a track made of rails or tubes and this type of track. Can your train do a barrel roll? I would be interested to know if any of the parts on the rough side of the track, or the edges, show signs of wear. Would it be possible to use rubber-tyred wheels on the rough side of the track, as the equivalent of the wheels that hold a real roller-coaster on the track? I have run wheels inside circles of tracks as a pulley race, as part of the propeller pitch control on a jet engine model. Would it also be advantageous to use some older and some newer tracks in order to customise the loops? Shape 1, Shape 2, Shape 3, Info Mark

This is the same problem faced by the designers of real steam engines. In the end they decided to power only 1 size of wheels, except for a few unusual locos. The key has always been adhesion weight - getting weight over the driving wheels to put traction on the rails. A 4-8-0 scores better than a 4-6-2 because the rearmost wheels are driving wheels. The rear pony truck (-2) should not take much weight because this removes it from the driving wheels. In fact in a LEGO loco it should take almost no weight at all. In my Co-Co diesel locos I have put the weight on the driving motor axles of each bogie and let the trailing pair of wheels float so that the weight of the loco is adhesion weight. Mark

My Class 14 trip-working loco uses a single PF XL motor in the cab. This has enough torque to drive the loco but not the train as well, so I run it with an adjacent brake van having 2x PF train motors, the IR Receiver and LiPo battery. The brake van includes pairs of diodes to drop the voltage applied to the train motors, so that the motors are well-synchronised at steady running speed of 40mph, which is the design speed of the loco. I would use a similar arrangement for a steam engine with any PF motors turning driving wheels on the rails, where an adjacent tender of carriage has PF train motors. The larger scale has an advantage that I hadn't though of as an advantage: Given that most wheels that are compatible with LEGO track, even BBB large and XL ones, are too small to represent the driving wheels of express passenger steam engines (i.e. larger than 5ft diameter), I have suspended other wheels off the rails. This removes the need for the proportion of power to be matched so closely. A bit of "visible wheel slip" is forgivable when operating a train! So it is with my Hogwarts Express, which has a Hall class engine using 40-tooth cogs for driving wheels and 2x train motors under the tender, with another 2x train motors under the first of 4 carriages. For steam engines like this I set the forward motion to suspended wheel matching speed to about 56mph scale speed (8mm:1ft scale), which is a good long-running speed for exhibitions as it's fast enough to enthuse visitors but slow enough not to burn anything out. I have more recently tried cut-down Technic wheel hubs to give a smoother appearance the train driving wheels, though these have not yet found their way into a MOC. A suitable model for the 120-degree 3-cylinder arrangement would be an LNER V2 like Green Arrow, or an A3 Pacific. This would work with a similar matching arrangement at 56mph. It would be difficult (not impossible) to match the speeds of different motors over the full speed range. This would need an electronic circuit to detect the back-EMF of one set of motors and drive the other set of motors in proportion. As I have asked about a slave H-bridge to drive many motors from a single IR Receiver, a speed proportion function could be another feature of it. Mark

I definitely recommend this set. My review is here @tholmol: nice idea to shrink the crane. Need 2 more sets to get the 1/11 actuators! It will need to be strong even to carry the hoses, so I suggest you use more 3.2mm rigid tubes, as the original pneumatic truck 8868 did. @Error404: Consider which pieces use a significant proportion of the value of the set (£170), based on customer services prices: - Pneumatic parts maybe £50 - Wheels £34.56 - Springs £12 - Bucket, turntable, telescopic parts (10) £13 - Panels £25 - PF L-motor and battery box £18 Comes to £152.56. There is a discount for buying a set compared to individual parts. I consider this sort of calculation when I'm buying sets, especially multiples. E.g. if you want the white panels then buying many of Cargo Plane 42025 is worth it - I bought 8! Of course all the interesting parts from 42043 are out of stock, and I expect they will be till next year. The Mercedes badge might never be available as a single piece, or else it will carry a large franchise premium in its price! Mark

For European trucks, the tax and weight bands favour particular configurations. This means a lot of trucks have gone for a single powered axle at the rear, with double powered axles only for trucks that go onto building sites (tippers, cement mixers). The Arocs is one of those. More common for delivery trucks is a tag (lifting) axle behind the rear powered axle or a centre-lift axle just in front of it. Chinese-6 (2 steering axles and a rear powered axle) used to be common in previous decades and is becoming more popular again. For a 4-axle truck, Chinese-6 with an extra tag axle at the rear is becoming more common. I saw a tanker like that today. Where a truck has a fork lift slotted into the rear bumper, or a rear crane, the rearmost axle may be steering instead of lifting. For multi-wheel trailers on an artic with more than 3 axles it is more common in Europe to see steerable trailer wheels. The usual alternative is 2 or 3 fixed axles in the middle of a trailer. Sometimes those trailers have tag axles; I have seen both tag and centre-lift on a 3-axle trailer. We don't have provision for road trains in the same way as the US or Australia, though the UK has examined the idea. One cartoon showed a passing place in the middle of the road train! Mark

I haven't found many 'B' models that I've wanted to build, maybe 8862 was a rare good one (Combine Harvester). The Unimog idea of a different tool on the same chassis was better. I did add the snowplough from the parts of a second kit. I don't do online instructions partly because there's no room in my office for a table to build on, so I've not built a 'B' model since they stopped making paper instructions for them. If so many 'B' models are so badly thought of, the next competition should be to design a better 'C' model for a set where you hate the 'B' model, and to justify why your model is better than the 'B' model Mark

Nice one. I might suggest that, were it not for the compressor, there would be plenty of pneumatic power available from the on-board jet engines Mark

Of course if your models make good use of what is available then you would see no need for change Perfectly logical. Too right that some AFOLs think only of what they want but conversely "you don't get what you don't ask for". The "not going to happen" bit and the "minority market" bit are why I developed another solution. Roll on 3D printing Accuracy depends on the choices you make for your modelling scheme. "Minifig scale" has a wide tolerance. My trains have better scale accuracy than "OO", especially in matching the track gauge to the scale more closely. Hence I aspire to get things accurate and take small compromises on some items, all the while seeking to improve. The standard buffer measurements are OK for 6- or 7-wide but 8-wide is a stretch and it's too far out for 8mm scale. Even so I've used some on smaller wagons, at least before I was able to build more wagons that look the part. Continuous improvement aims to reduce compromises in the scale. I could shave off the buffers and use my own separate ones but there's some risk to expensive parts in doing that, and it would probably look ugly. I suppose if the new couplings are so useless prior to chopping off the buffers then that would be the way to go, but there are other issues with them; it seems there is also more distance between the casing and the magnet as the magnet rotates inside the casing, weakening the train-pulling force and needing 1x2 plates or tiles to supplement the magnets. This compromises the whole function of the piece in connecting a train together reliably - they're not made for pulling heavy trains. In reality the base argument of "it's a toy" comes into play - these coupling pieces are not meant for AFOLs with big or heavy trains; they are OK only for 6-7-wide and modest train length. The 1x2-plate or tile solution is no better than the liftarms and pins I'm using already, so no need to cut the couplings. It comes back to putting useless expensive items (£3 each, £6 per vehicle) in sets so I don't buy the sets - the base argument that the LEGO train sets are toys too. An extra wagon set aimed at the AFOL market is when the commercial side does consider what AFOLs want. Although it would be fine for 6-7-wide trains, the magnet strength vs train length issue arises. Unfortunately the couplings would be £6 in a £20 extra wagon set so extra wagon sets would not be economically viable either (30% waste), which might be why there aren't any extra wagon sets at the moment; the extra wagons would be "toys" by virtue/vice of having the toy couplings. Note that TLG may supply toys but AFOLs make decent models of all sizes; the fun is in making the decent models with the limitations of the parts provided. And of course one could not criticise TLG for making toys! Buffers used to be separate pieces, attached to a 1x2 plate and added to the wagon bases that had integral magnets: Hence the potential solution I would like has already existed! A shame these are not still available as some of the 1x2 jumper plates or boat studs that I have used are a bit big for the scale, though they fit the overall impression of the end of the vehicle quite well. I think TLG panicked and over-reacted. The change from the 12V coupling shoe to the 9V coupling shoe meant the magnet was easier to remove from the shoe, and hence easier to swallow. The magnet could also attach to a 2x2 plate or an arm-end piece used in M-Tron sets. Clearly the swallowing incident required an end to the removability of the magnets from those pieces but a change to close the holes on the train coupling shoe would have been sufficient. I suppose someone thought that a) the magnet could still be removed from the shoe with sufficient force and b) it would cost more to assemble it at the factory anyway so why not make a bigger change. Mark

For a large mobile construction not requiring high speed I would consider using D-size batteries by getting a 6-cell holder. Those have a lot more capacity - 15.6Ah compared to ~2Ah. The same technology of batteries should give a similar resistance. Duracells work OK for medium power but may have higher resistance than necessary for high power items. The AAA battery box would just end up with half the running time before the batteries go flat; AAAs are not much cheaper than AAs! If the structure were static I would use a LEGO 9V train controller or a dual 30V 3A power supply set to 9V (e.g. for trains powered through the rails, static air compressors or jet engine test rigs). For a low resistance and least waste of power, the PF LiPo battery is a good option - 800mA before the current trip cuts out but it attempts to restart the model if it does that; I tested that with a train. It can give about an hour's running of a high power model (longer for low power) and takes about 4 hours to recharge, so getting one or more than one depends how many hours per day you want to run the model. Alternatively you can get a higher capacity LiPo but be careful as they can overpower LEGO motors; you would need to know what you're doing in radio controlled models to get the right type and get the best from them. Useful tips from Bonox too, especially minimising gearing and optimising the application to the speed of the motor. Mark

I noticed that such a beam is fitted across the front suspension, page 82 stage 48.9. There may be others if the whole chassis is that wide - I've yet to reach the stage where another is used similarly. I've done bags 1&2 so far. Taking my time to enjoy it I had the steering bevels the wrong way round to begin with - too easy to repeat an operation. I also got the pump cylinder upside down but it was easy enough to fix. The steering seems to suffer from too little rake, like most Technic sets. The instructions or box say 28 degrees but that's nothing to shout about. Should be nearer 45 degrees for the front axle. I will see later if taking off the 1M tubes helps, or if replacing the associated pins with 1.5M pins would extend the range a bit. It might be that the rocking beam underneath has reached its full useful travel anyway, and might need a longer lever to extend the steering rake. For larger wheels on the Arocs I might try the 68mm ones from Model Team set 5571 - a bit bigger than the 62mm ones. I had in mind previously to use 68mm ones on a tractor unit and 62mm ones on the trailer(s) - plenty of them from many sets now! Perhaps even Unimog wheels on the tractor unit and steerable 62mm ones for heavy haulage, but it's a long time since I did a big truck MOC. Before starting the build I experimented with the 2/11 cylinders and the end-to-end pieces from crane 8421. The new cylinders don't fit the end-to-end piece, unless the holes for the nozzles are extended. I did succeed in mounting the two cylinders end to end by another method, without them moving sideways, but it's bulky! I hope TLG will issue a new version of the end-to-end piece with 42009's successor in 2 years time! The total extension of 2x 48mm cylinders on 8421 was about 7M against the new cylinders' 6M so it might need two end to end; that would give it more boom lift power for a 3rd telescopic extension, to put right 42009's inadequacy in that department. Mark

Now that I'm getting back into Space, having got a new, fast and versatile monorail system to work well, Blacktron is one of the themes I can have a go at. I missed the original sets as my pocket money couldn't afford more than the Technic I was buying at the time. I prefer the original scheme - lots of black, a small amount of yellow, trans-yellow canopies and a few trans-red dots. Here is the Blacktron train prototype so far. The symbol came out OK in SNOT, not too bricky. The next question is how well black Technic would look as a significant part of the Blacktron zone models in a moonscape layout. Perhaps lime green should be the trim colour of a new Space faction. Mark

"Something special" ought to include saving the trains theme. The best thing TLG could do for LEGO trains is to make the couplings attach to a piece like a 2x3 plate with hole and not to the buffer beam. This would allow standard couplings to be used for trains of any width. As it is they are suitable only for 6-wide trains so they are a waste to builders of narrower, wider or narrow-gauge trains; no point spending £6 per vehicle (£12 per passenger train set or £24 per goods train set) on pairs of couplings you can't use. Hence those builders would not want to buy the train sets and TLG are cutting off half their market. I realise the change was made after a child swallowed some Geomag and got his intestine ripped, but it doesn't need a whole buffer beam to attach the magnet to, to prevent the coupling being swallowed. Until they do that I am reluctant to buy train sets and the LEGO trains theme will not thrive like it should. I use plates 2x3 with hole, a 5M thin liftarm and pegs instead of magnetic couplings for rakes of wagons and coaches. Another item needed is restoration of the 1x2x3 train window glass in trans-clear. Other than that, a new steamer as an improvement on Emerald Night (preferably wheels like BBB large ones) or a new Hobby Train set with versatility and a good set of specialist train parts, letting us get the ordinary bricks from other sets. If it were not for the bad couplings I would buy such a set in bulk. Mark

With the motor off, try switching each pneumatic function and moving it by hand in either direction. Each function should be movable by hand in the direction you switch it to. See if you can hear a hiss at the compressor, where air should be leaking in past its piston as it takes in air when you move the cylinder. If you don't hear a hiss, there may be a blockage. See if the blockage is for all functions or just a few. If it is for all functions, the blockage is probably in a blue tube - the common supply. If it is for some functions only, it would be in the grey or black tube of that function. Mark

Here is how I supported flexi-track on a slope, as well as canting and ballasting the track. The cant and ballast are integral with the track modules; 6 degrees of cant is the maximum on British railways. The system separates into track modules for each piece, composite tiles 2 plates thick to support the track, and scenery providing vertical and lateral support. The whole is a 72M radius curve and my heavy trains are quite happy running on it. A 104M radius curve is also easy enough: The amount of slope is 1 plate per 4 pieces, equivalent to 1 plate per piece of standard track, or 1 in 40. I used jumper plates to provide half-stud-height supports in the middle of each composite tile. A model railway can do up to 1 in 30, which would be 1 plate per 3 pieces. I have used that on straight track on my modular layout; 4 plates in a 48M baseplate with 3 straights. More pics in this folder It is possible to make steeper slopes but, for every change in slope, the train bogie pivot has to flex to accommodate it whilst keeping all the wheels on the track. I was considering whether a roller-coaster could be done. I tried using flexi-tracks back to back with plates joining them. I decided there are easier ways to make a roller-coaster! Mark

I am relieved to be able to confirm that the updated nozzle shape on all cylinders and valve switches is backwardly compatible with earlier hoses. Whilst they are designed for the new, tougher hoses used from the Unimog 8110 onwards, to allow easier removal, previous hoses don't fall off the new nozzles when the system is subjected to the maximum pressure from a hand pump cylinder. Had this not been the case, I would probably have recommended using the hose joint pieces (13x dark bley ones in the Arocs set) to use the new hoses for the last run to the cylinder and any hose beyond that with the older parts. The T-junctions and hose joiners have not had a new nozzle shape, but those are of significantly lesser value than cylinders or valves if a nozzle broke and TLG had to send a replacement! Noticing that the new pneumatic parts are out of stock when dialling up 42043 on LS@H "I want to buy a piece", this is not unexpected. I would expect it might be 2016 before the parts are in stock. Until then I'm sure TLG would want to ensure we all like 42043 and have no problems with the new parts, hence any stock would be reserved for mishaps. There was an idea that the whole bag of pneumatics might be a preferred supply method, rather than individual pieces, but this seems not to be the case, given that the 7-digit number of the bag has no records on LS@H. No option either for "I want to buy a bag of pneumatics - as many of them as possible!" I have tested the new cylinders and they show good top seals i.e. if you extend the cylinder, put your thumb tight over the top nozzle and compress the cylinder fully, it will bounce back most of the way. This was not consistently the case with the older 48mm cylinders; about 1/3 were OK, 1/3 mediocre and 1/3 poor in their top seals. I used the ones with poor top seals in pairs for point control, connecting the bottom nozzle only. I guess TLG had also found this reliability problem and did two things: They used only the bottom nozzles for the 8421 crane and they didn't use pneumatics in the 42009 crane, i.e. pneumatics was restricted to small pulling loads like the Unimog crane while they developed a solution in the V2 cylinders (no restriction on pushing loads). The cylinder joining piece, no. 53178, from crane 8421, doesn't work with the new 2/11 cylinders because the nozzle is further from the bottom end, clashing with the bracket, but it could do so if the nozzle slots on the brackets were extended in both directions and 4 half-bushes were used in the cylinder fixing (ironically less plastic, hence cheaper bracket if TLG did the mod!). Other ways to connect 2x 2/11s end to end may be easier. For a future crane kit it might not be necessary to use them end to end because the extension is 6M, compared to the 2x 3.75M of the previous 48mm cylinders. I hope we will see a 48mm V2 cylinder eventually, because not all applications need a long throw and pushing the levers of valve cylinders needs a 2-wide cylinder for more than 1 valve at a time. I have 1 48mm cylinder pushing up to 5 valves, though it takes a while to build up the pressure so I'd recommend 4 before using a second cylinder. 2 48mm cylinders driving 6 valves is fine. Mark

A realistic output from the 9V controller is 300mA before the voltage starts to drop under load. Other versions of the LM317 and related devices are available, up to an LM338 with 5 Amps capability (needs a big heatsink). If your trains are much more than a single motor on a set or two then I would recommend something more powerful. I use a dual 30V 3A variable bench power supply, which has been successful for running heavy 9V trains all day at shows. It provides smooth power that 9V train motors like, as opposed to PWM that they hate. Trains folder here In running 9V trains, stick to 200mA per motor for long running and reliability, peak at 300mA per motor. The Power Functions train motor is much more tolerant of PWM and a bit more powerful. The PF LiPo battery delivers 800mA so two engines with a LiPo, IR Receivers on the same channel and 2 motors each would pull all but the largest trains. My Hogwarts Express is in that league, with 4 train motors and gearmotors and lights as well; it has worked well as 9V with the bench power supply but a single LiPo was not enough. It drew 1300mA at the top of the hill. A LiPo battery is potentially more dangerous because of its low source resistance. Fast charging and discharging of LiPos has led to fire issues in laptops, Tesla cars and Boeing aircraft, some of these fixed by spacing out and cooling the cells. I suppose a few small resistors are cheaper than a single potentiometer, hence cheaper for the 9V controller to be stepped. The circuit as it stands would need a 3kohm pot. Most are 10k, which would need a larger balancing resistor and the pair would limit the available current and need a transistor stage, hence the cost escalates. The LM317 data sheet has better circuits for a variable power supply; indeed my bench power supply is based on similar circuits, with a few op-amps and power transistors thrown in. Phoxtane is right; barriers are necessary for trains, to avoid little fingers getting hurt. This is especially true of larger trains with more momentum. I still have a scar from being run over by my 9F steamer as I was repairing the track! Mark

If you want blue sprocket wheels to go with the original blue livery then there are 4 in set 70731 Ninjago Jay Walker One. Mark

For 9V trains with 9V lights it's easy enough to power the 9V lights from PF or use a PF light brick instead. For trains without lights it is unlikely there would already be a wire from the 9V motor to inside the loco body in 9V mode. This is not too difficult, just rebuild the bit between the bogies to ensure the 1x2 holes in the carriage base are not obstructed. The main issue is the volume of PF items and leads inside the loco. Battery box and IR Receiver for starters, but everything else adds more leads. I converted a larger MOC diesel loco from 9V to PF for experiments and as a template. The loco already had a wire connecting 2x 9V motors for better load sharing, so no difficulty there. In the experiment I didn't change the motors so it can run in either 9V mode or PF mode. Unfortunately 9V train motors hate the PWM drive of PF. The battery box, receiver and 2x light bricks (with leads) took up significant space but there was still room for a small breadboard for an automatic light direction circuit. So a 2-bogie diesel MOC is not too difficult to convert - enough space inside. A large steam engine might be a custom rebuild to convert from 9V to PF. The tender is where most electrical items are housed. The motor tends to be structural in a steam engine (2 here); different for PF. The two diesel shunters I've built with PF are a real squeeze. On one the XL motor fills the cab and there is barely room for a wire between the SNOT panels, even with most electrical items in the brake van! The other is crammed with BB, IR Rx, 2 motors and a gearbox. Neither would have suited the 9V era (though I did try the first one with 9V gearmotors) but it goes to show how a lot of trains (particularly those with large wheels) are now as much about Technic building as train building in the PF era. This makes a decent challenge for AFOL trainheads but may put off younger fans; this is where the Emerald Night was good. If your layout has tunnels then an IR Repeater circuit may be useful, to enable the train to receive signals in the tunnel. For exhibitions an SBrick would be better as it gets over the problem of interference from anyone with an IR handset in their pocket. Mark

Jim, thanks for a great review. I will probably buy 2 sets, 1 to build and 1 to begin experiments with the new parts. I have uses for the extra panels and I identified uses for the grill support pieces when I bought some Exo-Suits. I will have to test the differences between new and old nozzles with recent (Unimog onwards) and older tubing and also 3rd party tubing, to check for hold-on and pressure characteristics. I have a new air compressor so plenty of capacity for pressure testing At over 3 bar it can blow the silicone grease out of the valve switches - definitely too high! Stick to 2-2.5 bar. The inner diameter of the nozzles will be interesting to see if it's the same as the old parts. Also to see if the diameter is the same all the way into the cylinder (top nozzle visible). On older parts there was often a burr that could be drilled out with care (mind the top seal). From what you say about slowness and jerkiness of crane operation it seems the stiction has not been reduced in the new parts. TLG probably stuck with the proven technology of the pistons. I guess the pneumatic parts would not be available separately till January 2016. I look forward to seeing whether there will be a new 48mm cylinder in the coming years. Since I do logic systems with pneumatics, the longer cylinders are most useful for actuation, rather than moving other valve switches. This means I would buy some but not too many (compared to 10x 8455 with 10 cylinders and 7 switches in each set). Mark

Not impossible but: - Conventional parts would give it very little thrust. - Using an Airjitzu spinner would be possible. Maximum thrust of a spinner on a PF LiPo battery is equal to the spinner's weight. - A LEGO propeller with significant thrust would be much larger. I have made propellers with as much thrust as a desk fan, but for ~40W electrical input a large static power supply is needed! - Other solutions using non-LEGO motors, propellers an power supplies would be possible. 11.1V LiPo with model aircraft motor perhaps? Mark