Technic 9398 4x4 Crawler Review: A Double Milestone

The Lego Group has always been reluctant to launch a Technic vehicle with an electric drive ― while various functions powered by electric motors aboard are quite common, the drive was entrusted to the builders' hands rather than motors. This approach is not without a legitimate reason: large powered Technic vehicles (as many MOC's attest) are usually complex to build, tend to be fragile, and often can't produce breathtaking performance. Being the construction perfectionists which they are, TLG obviously wanted to address these problems as much as possible before crossing this line. But now, with 9398 they have crossed it, introducing some important new parts along the way, thus easily qualifying the set as a "Double Milestone", as mentioned in the title.

Upon the ceremonial opening of the box, a familiar sight greets: a couple of bags numbered for easy assembly in two steps ― the chassis and the bodywork, with the latter requiring much less diligence and attention than the former.

BUILDING AND CONSTRUCTION

For a 4x4 crawler with dual live axle suspension, quad portal axles and all wheel steering (actually being a 4x4x4!), building isn't too difficult. Partly to thank for are the large, easily understood instructions with only a few parts added per step in average. There are three instruction booklets, with an additional diptych clarifying the correct orientation and positioning of mechanical parts in several building steps. Also, there are many auxiliary structures, such as beams and frames that hold the model together while under construction and are removed afterwards, and pins whose only purpose is to prevent you from inserting a part wrongly oriented (e.g. a differential).

The front and the rear axles are largely symmetric, each holding its own power motor, while the ball links keep them connected to the main chassis where the servo PF motor is located, which controls the steering through reinforced U-joints. The chassis is actually completely independent of the bodywork and can run without it ― the bodywork is attached only at the last steps. This serves its own purpose: it is intended to be released with two pins accessible behind the openable doors, and lifted to allow easy access to the battery pack. (It could have been placed in the bodywork too, but that would probably keep the Crawler's center of mass far too high for an offroad vehicle.)

Building a 9398, one can notice some tricks, well-known among experienced MOCers, employed to ensure more strength than needed for the conventional models ― e.g. using two 1/2 bushes instead of a single fullsize bush for more friction. The overall result is staggering: this is certainly one of the structurally strongest models in Lego history, surviving rolling falls and quite severe offroading accidents without more than, say, losing a rear-view mirror. And don't think it's light: with batteries, it weighs a bit over a kilogram and a half*.

Overall, it is a satisfying model to build. Only once I've had a feeling I was fiddling with excessive detail, when assembling the front grille which consists of dozens of small parts.

One may feel, and some have published, too, that the remotely controlled 4-wheel drive and steering, dual live-axle suspension and openable doors are a bit short list of functions for a Technic flagship. But I tend to disagree: toying with additional functions might make it more fragile, reduce its offroading performance, and detract from the idea of a good, focused vehicle which does one thing and does it well.

The bodywork was also designed with strength in mind, but still it looks good too, especially since it uses some less typical colours. There are no functions here ― seats and tailgate are fixed, and steering wheel, gear lever and extra lights purely decorational.

All that strength implies something else: this set is a real pain to disassemble, especially because it uses over a hundred friction axle pins, most of which are stuck into other connectors, and you will need a crater-load of patience to pluck them all out.

PARTS

Important new parts, largely responsible for the amount of attention this set has received long before it hit the shelves, are the PF L and Servo motors. As you may already know, the PF L fills the gap between a M and XL motor: noticeably more powerful than the former, but smaller and more easy to mechanically integrate than the latter. This set contains two of those and I'm sure we are going to see these motors a lot in MOC's, too.

Servo, which should put an end to various recentering mechanisms, simply has its axle mimic the movements of the controller (be it small, included in this set, or the large, with 7 levels in both directions), instead of rotating at the speed proportional to it. Of course, this is a perfect part for steering in any vehicle, but will find its usage for many other applications. If you're interested in technical specifications of these new motors, take a look at Philo's website ― he has painstakingly measured them and published the results.

A new IR receiver has been released too, labeled "V2" at the front. It can handle more power, required to run two PF L motors adequately on a single channel, but otherwise it behaves just like the old one. In addition, we've got a new axle with a stop and a perpendicular connector, but they aren't as spectacular introductions as the PF components.

As one would expect, this is an excellent source of general Technic building material ― various beams, panels, pins, connectors, axles and similar parts are all supplied generously. Also, there is a significant amount of valuable, rare parts: beside the still über-rare new PF components, 4 large 94.8x44R tyres, 4 new triangular liftarms, 2 large ball joints, 4 portal axle housings and 2 super-hard 9.5L springs are certainly welcome to any builder's collection.

There is some amount of relatively rare white parts which may also be useful. Also there are some orange parts, but far too little for building something spectacularly orange. Majority is still, however, in the familiar red-grey-black scheme.

PERFORMANCE

All that devotion to mechanical strength, suspension and drivetrain pays off: the Crawler is unstoppable by mere mortal obstructions. Although the instructions carefully tell not to drive it outdoors (to keep the parts from being milled by sand and stuck by dust), if you choose to live on the edge and disobey, you will find it drives over rough outdoor terrain very well ― climbs the boardwalks and hills, bridges the gaps and consistently resists tipping over.

This performance largely owes to the portal axles and a very agile suspension (e.g. easily over 10 cm displacement for a single wheel from the plane of other three), and the PF L motors that provide serious power and torque. It is not uncommon for the Crawler, being stopped dead by a wall, to actually spin the rims inside the static, blocked tyres. The large passive torque of the motors produces one more cool effect: when they are turned off during a ride, the car just stops dead in place.

Since the axles are pointing slightly downward when the car is level, steering makes it lean into the curves, similar to the motorcycles. Extents of steering are not too wide, but thanks to the four wheel steering, the turning circle remained small enough for easy maneuvering. Without the rear wheel steering, it would resemble dragging a stubborn dog around.

You will have plenty of fun inventing all kinds of trials and obstacle tracks for the 9398, but make sure to have an assistant with wheelbarrow loaded with spare batteries follow you around. Faithfully following the image of the offroaders, its fuel consumption is not extremely economical.

MODIFICATIONS

With lots of torque available onboard, it shouldn't be a problem for 9398 to carry extra weight, probably added or rebuilt at the rear, behind the cabin. Also, it should be able to pull some load or a trailer, at least not on very steep ascents.

Furthermore, since the bodywork serves no mechanical function and can be simply detached from the chassis that does all the work, it can just as easily receive any other type of custom bodywork, as long as it can attach more or less to it. In fact, a B-model truck follows this approach.

The chassis itself is less prone to modifications ― it is built very tightly and mechanical parts are reinforced like a nuclear shelter. So any attempt would probably mean rebuilding lots of components from scratch.

GENERAL PROS & CONS

+ Good, strong construction, interesting to build
+ Fun to drive (good offroad performance!)
+ Valuable new PF parts and overall building material
+ Very easy to change or rebuild a custom bodywork

- Disassembly horror
- Some components are quite fiddly to build

▪ VERDICT ▪

A Technic flagship which does well what Technic flagship need to do: provide interesting new concepts and useful parts, lots of fun building and using ― and customizing if that's your cup of tea. Cheap it certainly isn't, but the bang-for-buck ratio is still quite good. I'd be crazy not to recommend it.

*1570 g, to be precise.

GALLERY

        

LDR world map with altitudes

One small thingy while collecting ideas and time for the next project: I've created a Lego world map Ldraw/MLCad (.LDR) file with altitudes, for a friend who's preparing a small geography fair and considers building one. For the case you might find some usage for it too, it's also available for download here.
It's a quite large map, 250x125 studs (2x1 meters, approx. 80"x40"), and the model consists of about 130 000 smallest plates stacked where necessary ― it's up to the builder to choose the most convenient reinforcements and substitutions with larger plates, bricks or columns. The altitudes are, of course, drastically out of scale, and there are two colours: white for the land and blue for the seas, as you can see on the rendered picture. It includes the Antarctica and some tiny ocean islands represented by just a single brick, but there should be no problem deleting them in MLCAD or any similar LDR editing application if necessary.

Medieval Watermill

I've been toying with the idea of harvesting hydro power with Lego for some time. While a prototype of a crude yet functional water turbine could be done quite easily with Technic parts, building a small, medieval minifig-scale watermill was too attractive idea to ignore, even if it meant a little detraction from my usual Technic & Classic Space habitat.

As evident from the photos, its design is really simple, yet it took a couple of unsuccessful tries and total rebuilds to make it work smoothly and reliably. A flume, to which a water source is connected at the top end, leads the water over the millhouse and drops it on the water wheel blades, from where it falls into a riverbed, and finally, flows out from the MOC. In the house, the axle is geared down to one third where it rotates the millstone, of course using standard Technic parts. Everything else on this 32x32 diorama is purely decorational.

The first version was designed with an undershot water wheel, but unfortunately the water stream was just too weak to start even the freely-rotating water wheel. I suppose it could be done with a strong stream and the blades that match the riverbed very tightly, but I rather opted for the overshot form, which works much better: it provides sufficient torque, is started easily, and keeps a nice proportion between the water flow and its rotation speed. Actually, it requires very modest flow, as it works reliably even below one liter per minute.

Still, the amount of torque provided this way is really minimal: if you will want to build one, forget about doing any "serious" work such as lifting weight unless you are ready to accept it will be drastically geared down and thus intolerably slow, or even worse, attaching it to a motor axle (using it as a generator) to provide some tiny amount of electric power. A seriously large MOC with complex bucketed wheel(s) that could handle flows of at some 20L/min or more could perhaps do that, but building this I didn't want to venture into megalomania.

Watertightness is ensured by a bit of cheating: both the flume and the riverbed have double walls and floors that snugly fit one into other, with a flexible plastic kitchen foil spread between them. It remains watertight even when deformed by the studs connected through it, and if cut precisely at the outer edges with a razor blade, it can be completely hidden from view. The flume is at a slight angle to help the water flow towards the wheel, and of course, the riverbed has a dent at the far end to let the water drain exactly there, instead of flooding the entire diorama.

This MOC has confirmed the basic idea of using hydro power, and the next step might be an advanced Technic water turbine that could handle considerable flow and provide more efficiency and power.

GALLERY

VIDEO

Bricks at an Exhibition

A Lego exhibition roaming Europe took a stop in Zagreb, and I went to take a look about a week ago. It was nicely organized and displayed, and I took my camera along to snap a few photos. Rather than dump the entire gallery below, I've picked a few that were most interesting. (The captions are embedded in the photos.)

Technic Gear Ratio Finder

A while ago, trying to build some kind of an orrery, I've had a problem of finding a combination of Technic gears that yields a very specific and precise gearing ratio. To help myself out, I've written a simple program, and hope it may help other builders facing a similar problem too.

So, basically, what it does is ask for the desired total gearing ratio, maximum number of gears involved, and the information which gears (not) to use, then display the gear combination that satisfies the input parameters, or at least comes closest possible under given limits. I have to admit it is a bit crude on the surface, but it works, is simple to use, and does not mutilate your computer with registry bloat. The full instructions and notes are packed in a Readme.txt file in the ZIP available for download, but let's just run quickly through them here.

There is no particular installation; just unpack the downloaded LegoGearCalc.zip in a folder of your choice, and start LegoGearCalc.exe. A text window should open, asking you for a first parameter: the desired gear ratio, that is, the number of turns you want the output axle to do while the input axle does exactly one full turn. This can be an integer, a decimal number, or even an expression using basic operators. E.g. entries such as 14, 0.18, 6/13, (5+2)/3*9 etc. will work fine.

Then, enter the number of gear pairs you would like to involve at most. Default is four, but you can search up to five pairs. Finally, specify any of the gears you would like to exclude from the search by entering their numbers of teeth separated by commas. Worm gear is 1. Just hit Enter (leave the entry empty) to use all Technic gears.

After a few moments at most, the program will reply with a list of gear pairs that come closest to the desired ratio, or an exact solution if there is one. Occasionally there are several ways to reach the same result, or it can be reached using less gear pairs than set as a maximum. Of course, in your system they need not be in the order from the list. Two screenshots of typical usage examples are below, with the entries typed by the user in cyan.

Many gear combinations are not easy to mesh together since they require specific spacing, but there are various methods of creating a non-integer stud distances between axle holes one can find (or better yet, invent).

(Or click here to view or download just its instruction text file.)

EXAMPLE SCREENSHOTS

 

Moved to legoism.info

One small step for a DNS registry, one giant leap for this blog: it's moved to a new address - www.legoism.info, so update your bookmarks.
(Though, you don't really need to; old URL's should redirect to the new ones...)

Red Raccoon

At last, the project that has been residing on my desk since March is finished, and here is its result ― the largest and heaviest Technic vehicle I have built so far. As expected, developing an off-road car of this size proved just as challenging as a racing car. An off-roader is more generous in terms of volume and component layout, but on the other hand, its chassis has more complex mechanical requirements to satisfy.

Length: 58 studs
Width: 27 studs, 29 including tyres
Height: Variable, bodywork only: 21 studs
Weight: 2.79 kg
Suspension: Full independent
Power: Power Functions and NXT; 8 motors, 2 pairs of lights
Drive: 2x PF XL motor, 4WD
Steering: Independent front and rear steering, 1x NXT motor with Ackermann geometry each
Features:
  • independently adjustable front and rear ride height
  • lockable central and half-axle differentials
  • 5-speed gearbox
  • drive, steering, differentials, ride height and gearbox remotely controlled
  • headlights and rear lights
  • openable roof (to change battery packs) and bonnet

The first design stage required a development of a good half-axle component that, besides drive and steering, allows ride height adjustment and has a lockable differential via control axle. When I have settled on a working and compact design (which can be seen on the photos below), I've used it symmetrically for both front and rear axles, with some minor changes required at the front to allow the bonnet to close. The central section of the chassis, which connects these two half-axle components, houses a master differential which is linked to the output axle of the gearbox. Two NXT motors, positioned at the sides of this central chassis section, control the steering with a standard rack & pinion mechanism. I've tried to keep them as low as possible to lower the center of gravity, important for off-road stability.

Red Raccoon is powered by two parallel PF XL motors linked into a main power axle with knob wheels, which leads to a 5-speed gearbox at the rear of the car. It is actually a variant of Sariel's linear gearbox, modified so that it doesn't require a 16L axle (whose order and delivery, I have to confess, I haven't had enough patience to wait for), for a price of being one stud longer. As all linear gearboxes do, this one needs slower speeds to shift gears reliably, but in practice it works quite well. Its important advantage is its ability to change gears using one simple motion ― moving the middle axle ― which is done with a liftarm connected to an NXT motor. That was also one of the main reasons for introuducing NXT to the car: an NXT motor can perform very precise movements, in this case in steps of 60°-30°-30°-60°, required to change gears accurately. The gearbox output axle leads, as mentioned earlier, to the master differential.

All three differentials (one master and two on the half-axles) can be locked with a single remotely controlled XL motor, which is attached to a system of clutches and arms that engage and disengage dog-rings on each of them. However, since the master differential intentionally locks more slowly than the half-axle differentials, in both ways, with good timing (matter of a few seconds) one can also lock only the half-axle or only the master differential, rather than having just the option to lock or unlock all of them together.

Remaining two motors have the task of adjusting the front and rear ride height. Each of their output axles is connected, through a clutch, to two cascaded worm gears that turn small 2L liftarms on which the entire suspension (with springs) rests. With their turning range of about 135°, both ends of the chassis can be raised or lowered for about two studs, while the springs themselves allow for about two studs on top of that.

The lights are just standard two PF pairs, and the difficulty was not in positioning them, but rather in trying to avoid their miles of wires getting tangled into gears sticking from everywhere. The roof can be opened to replace the PF battery pack or an NXT smart brick easily. The bonnet opens too.

As one can notice on the photos, cramming all these batteries, motors and mechanics within the bodywork constraints meant there would be no room for interior decoration, even in a rather large model like this one. However, I've tried to keep the top 3-4 studs below the roof empty, to give at least some impression of this car having windows. Mentioning the bodywork, once again I went in the direction of "conventional" Technic design comprising mostly of angled beams, panels and an occasional soft axle, but leaving enough openings to reveal the mechanics inside. The bodywork is studless, as is the entire car actually, with a single exception of a rear PF infrared receiver support.

One more effect of cramming lots of equipment was a deeply asymmetric internal component layout, resulting in the car having a slight off-balance ― it leaned slightly (about one half a stud) to the right, which was solved by small readjustments of the suspension.

The current wheels based on Technic Medium rims with the tyres from 8880 can be replaced by the old Technic 24x43 wheels, visible on one of the photos. This change slightly increases traction on very rough terrain, but reduces performance on flat, fine surfaces. One can easily mount these old rims to the hubs with two Technic pulleys and a 3L axle with a limiter, per wheel. Being of roughly the same size and similar steering pivot point, this replacement doesn't require steering or wheelspace geometry changes.

Since the car is controlled both through PF and NXT, I've built a controller that has a shallow cradle for my mobile phone (used for NXT control via Bluetooth) between the PF IR remotes. That altogether is quite large for a typical hand so perhaps not the most elegant way, but it works.

The overall performance of the car is satisfactory, and it handles sharply if both steering axles are used. Also, all the components and features of the car work reliably, and the chassis is strong enough to allow rough driving and occasional drops without any damage. However, since it follows the form of an SUV more than of a standard all-terrain car, its ground clearance may not be sufficient for really hard-core off-roading.

One more point that should be improved by the successor is the steering. It could be a bit more stiff ― weight of the car now produces grip for which the control components need significant force to overcome, and due to many parts involved that incur backlash, slack and bending, it's not always easy to handle. This would, though, require a significant redesign of half-axle components, in turn demanding a redesign of the entire chassis. And at least a small fake engine and a bit of interior would be nice, but would require increasing the overall size by about 15-20%. Why not?

GALLERY

 

BASIC VIEWS

 

GENESIS

VIDEO