Jump to content
nbTMM

Bruiser (clone) build: Automating the 3-speed gearbox

Recommended Posts

Firstly, lets address the elephant in the room: this isn't a genuine Bruiser, but the chinese HG P407 clone. Tamiya diehards please don't grill me too much, hopefully the amount of other genuine Tamiya gear I own makes up for my sinful purchase :P. The main reason (other than the price) that I went for the P407 is for the almost completely metal drive train and because I plan on bashing it and as the thread title suggests, shifting gears on the fly (there will be an electronic gearbox controller, eventually!). If I rash up the shell or grenade the gearbox, I won't feel so bad as replacement parts are dirt cheap. I was also curious about the build quality of the clone. The aim of the build is to make something which drives as scale as possible, but not neccesarily look as scale as possible. It should accelerate and handle like a 1:1 version. Everyone is probably familiar with the way electric RC cars accelerate - unrealistically rapidly from a standstill and then the motor power falls off a cliff and the car cruises at top speed. I'll be designing an electronic controller to shift gears automatically and alter the motors power band so accelerates more like a real car.

I purposely read as little information as possible about the Bruiser or P407 before buying this as I find one of the most enjoyable parts of the hobby finding mechanical design flaws and working out how to come up with a solution for them. Some of the problems below are probably common knowledge but new to me and maybe I've attacked them from an angle previously not attempted.

So, on with the build!
 
GrJkt7p.jpg

What's this then, it's already built?! Let me explain... This won't be a build in the traditional sense of logging every nut a bolt being put together as per the manual, but instead looking at some of the good and bad points of the kit and re-engineering of the parts that don't work so well. 

The P407 comes as both an RTR which is simply named "P407" and a kit that requires some assembly named "P407A". The RTR adds a transmitter, receiver, ESC, steering and shifting servos and a lipo battery. The RTR doesn't cost significantly more and seems like good value initially, however, I still went with the P407A for a couple of reasons.

Firstly, all the extra gear included in the RTR kit would be surplus to my needs as I already had spare servos and an ESC to use, batteries/radio would be shared with my existing RC cars. I wasn't keen on having all that stuff sit around on my shelf, and probably eventually end up in landfill because I have better parts to use.

The second reason becomes more apparent when you receive the kit. The 'kit' arrives partially assembled - most of the oily bits have already been assembled for you. The front and rear axles/diffs, motor/gearbox and most of the ladder frame chassis come as pre-assembled modules so the build consists of the suspension, installing propshafts, steering, wiring the electronics and attaching a few ancillary bits and bobs. The problem is that the pre-assembled modules aren't assembled with the care and attention that most members of this forum would be happy with. You will want to loctite almost every bolt too if you plan on bashing so tearing down was expected. Thankfully they include a full instruction manual for the pre-built modules in addition to a cut-down manual which just assembles the modules together, although the 2012 Bruiser manual will probably get you out of trouble too since this is almost an exact clone. I encountered bolts with the hex heads chewed out due to being over-torqued, bearings with rust on their casings and dust covers missing and bearings that were seized. These parts were tossed out and thankfully I had replacements on hand already. I figure the less parts pre-assembled the better as all the non-assembled parts were in good order. 

The first modules I tore down were the front and rear axles/diffs. There was a good amount of red grease applied in here, except that not much of it was on the gear teeth haha. I opened up a diff centre and there was ample grease in there too. The internal gears are brass - it was a prick of a job to line up all those gears and get the centre casing back together so I didn't bother disassembling the centre on the other diff - just removed the bolts and loctited them. I noticed that the input shaft of one diff bound up a little bit when I turned it by hand and the problem turned out to be the cast aluminium pinion gear. A bearing fits over the back part of the gear (coloured red below) and since the casting is not perfectly true this caused the pinion and pinion shaft not to run true. The outer part of the gear (coloured blue below) was contacting axle/diff housing. I mounted the gear in my drill press and gently ran a file against both surfaces. I stopped when the bearing started being a loose fit. This helped but it still does not run totally true. Thankfully it no longer binds on the casing and the wobble is mild enough for the universal joint of propshaft to deal with. Ideally you want to mount it in a lathe by the shaft hole and turn it perfectly true.
cLisjym.jpg

Next I tore down the gearbox. There were several dud bearings installed in here, which were thankfully all common small sizes (5x10x4 mostly IIRC). The gearbox uses steel gears and anodised aluminium hexagonal drives and collars to shift. The motor spur gear and the planetary reduction gears are plastic, although that is not of concern since these gears are subject to the least amount of torque in the entire drivetrain. I believe the Tamiya gearbox has some material removed from the hexagonal parts so they are more 'star' shaped, although functionally they are the same. There was some light oil applied to the shaft and some grease on some of the gears but not nearly enough for my liking. I added more grease to the gears and reassembled. I ran the gearbox just with the motor attached and shifted it by hand and it didn't want to go into some gears easily so I ended up tearing the whole gearbox down again and taking some 400grit sandpaper to the sharp edges of all the hexagonal drives and collars (highlighted red below), cleaned/degreased all the grinding dust and regreased them. After reassembly the gearbox shifted much better, but still was hesitant to go into gears sometimes. Later I found this isn't really a problem - when the output shafts are under load gears shift much easier.
 

iH0labn.jpg

Once I had the chassis fully assembled and tried shifting via the servo while in motion, I had a lot of problems with mis-shifting. The culprit turned out to be the servo saver. The included ones are not really strong enough to force the hex collar into the next gear when the hexes aren't synchronised initially. When this happens the gearbox can end up stuck in neutral. Sometimes playing with the throttle is enough to align the gears better so that the weak servo spring can get it into gear, other times you have to resort to hunting around for another gear that works. 
Since i'm going to have electronics shift for me, mis-shifts must be avoided at all costs. If you get stuck in neutral you'll have no brakes and since the controller could decide to shift at almost any time, randomly having no brakes and needing to intervene to get it back in gear is a deal breaker. I tossed the servo saver and put a straight plastic horn on. Shifting became 100% reliable after doing this. Of course, this will cause the servo to stall out until it can get it into the next gear so shifting must never be attempted when the car is stationary. Fortunately, this is easy enough to implement in the electronic controller. I also tossed the stock servo saver on the steering and put in a plastic one from a TT02 kit, which required the horn to be shortened slightly.

tQlALrf.jpg

Speaking of steering, many will know that this is the achilles heel of the Bruiser. This is due to the arrangement of the white plastic lever shown below which translates the fore and aft movement of the turnbuckle coming from the servo to side-to-side movement that turns the steering knuckle via a second turnbuckle. The problem is that the plastic part can rock side to side because the pivot hole is slightly larger than it needs to be.  The side to side movement allows the wheels to turn when there is no fore and aft movement of the servo turnbuckle. There is an aluminium bushing inside the plastic part which was slightly longer than it - I shortened it by 0.5mm or so which allowed the steel part to be tightened down fully onto the white plastic lever. I tightened it just enough until I could feel some resistance when I rotated the plastic lever. This tightened the steering immensely however flex in the plastic part still results in a significant amount of steering slop when trying to turn the wheels against something e.g. when rock crawling. Additionally, the stamped steel part flexes. I plan to make an aluminium lever to replace the plastic one and add an aluminium brace from the head of the bolt to the other side of the chassis to stop flex in the stamped steel part.

jDfQwy2.jpg
Next we'll look at the suspension. There is much discussion about removing leaves from the Bruiser suspension to make it less bouncy and have more 'flex' - i'll give my 2cents: no combination of the included leaves works perfectly. With just the main leaf, the rear suspension is a little bit too soft as it sags to about 50% of the travel under the cars own weight. This is probably acceptable for crawling but for the ultimate realism and handling over rough ground at speed, it should sag only about 30-40%, leaving 60-70% of the suspension travel to soak up bumps. At the front, due to the extra weight of the motor and gearbox, just one leaf is way too soft and suspension almost totally compresses under the weight of the car, causing the propshaft and front bash plate to hit the gearbox casing when the car drives over a modest bump.

The problem with installing the additional leaves is that they do not increase the stiffness until the ends of the leaves 'engage' (touch) the leaf above them. So for the first ~50% of compression the suspension effectively has 1 leaf and is too soft, then the 2nd and 3rd leafs engage suddenly and the suspension becomes way too stiff. This is the cause of the Bruisers standard suspension configuration being comically bouncy. Even with all 3 leaves, the front suspension still compresses too much under the cars weight. The solution is helper coil springs in the front shock absorber assemblies which will increase stiffness through the entire range of suspension travel. I kept the small leaf as I like the 'scale' look of having multiple leaves, and the small leaf doesn't engage until the suspension is almost bottomed out anyway.

Speaking of shocks, the ones included with the P407 are rubbish. Firstly they leaked everywhere. After sorting that out by replacing the o-rings with slightly fatter ones, they had way too much damping even with 350cst oil (thinnest I had), and felt really gritty as the shafts moved so I gave up on them. I purchased a set of Yeah Racing Desert Lizard shocks. These have internal springs and are advertised as 90mm but are actually longer than this (~100mm) because they have a 'droop spring' installed on the other side of the piston which prevents the shock drooping all the way. I'm going to leave this spring out - the leaf springs will limit droop anyway. There is a good selection of springs included so I should be able to find a combination which stiffen the front and rear suspension appropriately. The threaded end of the shaft is too short to fit the Bruisers rubber strut top but this is solved by installing 1 nut and then a brass standoff which the rubber bushing fits over, and then installing the second nut onto the thread of the standoff.

6n7Es2V.jpg

Since the shock body is longer than standard, the standard shock mount on the axle won't work. I quickly cut an aluminium adaptor bracket and bolt spacer/sleeve to see if it would work. Turns out I don't have long enough bolts on hand, so I put some brass stand offs just to mock things up.

k65VLpZ.jpg

5aiNFg6.jpg

KCknnx2.jpg

Seems to be workable! Now I just need to get the correct length bolts, make enough for all 4 wheels and make them look presentable (round over the corners!)

The next problem was that the rear trailing arms made jingling noises as the car drove over bumps. This was fixed by adding a 2nd rubber o-ring to the bottom side where it is attached to the chassis, and adding some adhesive backed closed cell foam around the axle end.
lVkv3Fy.jpg

A final simple and effective mod is to raise the front of the gearbox. I put 4mm of aluminium spacers at each mount. This gives more clearance for the front suspension to compress and tilts the gearbox at an angle, allowing the front axle to be tilted back at the same angle without causing vibrations of the propshaft due to different universal joint angles. Angling the axle backwards introduces caster to the front wheels which will make it more stable at speed. The angle of the rear axle should be adjusted too - there is no handling effect from changing it, it'll just allow the propshaft to run vibration-free again.
ogD2RTJ.jpg

JCxAviu.jpg

Train spotters will notice the red motor. This is a trackstar 13.5t brushless instead of the included 540 brushed motor. There was nothing wrong with the included motor however the brushless is needed for the automatic shifting electronics I intend to build. It probably seems strange that I chose a 13.5t motor when even a 25t is probably overpowered for 'scale' driving. The reason for the motor choice is that I need the 3000kv to reach a realistic top speed without altering the standard gearing and the excessive power of the motor can always be dialled back by electronic control. In fact, having an excess of torque is desirable as torque-limiting can be used to shape the torque curve of the motor in a way that mimics an internal combustion engine and make it accelerate more realistically. We can take away torque at a given rpm but we can't add more, so starting with a powerful motor is more flexible. Also if I get bored with 'scale' driving can just turn torque limiting off and do mad launches :D
Brushless allows motor speed sensing to be implemented with minimal complexity - I went with sensored since it further simplifies the speed-sensing electronics and the motor will run smoother when crawling. Also the spare ESC I had required a sensored motor.

As far as the automatic gearbox controller goes I've got an Arduino reading 4 channels from the receiver and outputting 2 channels continuously. Still need to implement reading the motor rpm but that should be straight forward. The controller will sit between the receiver and the ESC / gearbox servo. It'll also tap off the motor sensor cable. By monitoring the motor rpm the controller will decide when it needs to shift gears and can intercept the throttle signal to either blip or cut the throttle to ensure a smooth gear change. Since I'm already intercepting the throttle signal and measuring motor rpm, implementing motor torque limiting/shaping is just a matter of software. Realistically the motor power should be turned down such that the acceleration is the same as the real car - if the 1:1 car does 0-100kmh in 12 seconds, then rc should do 0-11kmh in the same amount of time since the scale is roughly 1:9. With the correct amount of power and a realistic torque curve it should actually accelerate faster to it's top speed by working through the gears instead of just putting it in 3rd gear and planting it from a standstill. Another feature that should probably be implemented is to limit the braking force in 1st and 2nd gear so it matches 3rd gear, due to the braking force being multiplied by the gear ratios. I'll post more when there is something interesting to show!

I think that's it for now :)

  • Like 12

Share this post


Link to post
Share on other sites

Interesting... I'd never have thought of automating the shifting on a 3-speed truck. (Probably because I hate automatic transmissions in 1:1 cars.) Curious to see how it works.

Also, thanks for the info on the HG kit. Hearing that it comes partially assembled takes the last little bit of temptation away from me; if I ever want a Bruiser, it's the real thing, or nothing.

  • Like 1

Share this post


Link to post
Share on other sites
7 hours ago, markbt73 said:

Interesting... I'd never have thought of automating the shifting on a 3-speed truck. (Probably because I hate automatic transmissions in 1:1 cars.) Curious to see how it works.

then again, an electric car eg Tesla doesn't use a gearchange gearbox either, do they?

Share this post


Link to post
Share on other sites

I think it is possible to select 1st gear 2WD and 1st gear 4WD independently with careful EPA setup so the plan is to be able to configure the 3-position switch on the transmitter which is currently used to shift gears, to one of two modes. One mode would select between 1-4WD, 1-2WD and 'D' like a conventional automatic gearbox where you use D for driving around at speed and 1-4WD and 1-2WD for crawling. Another mode would have 1-2WD, 2-2WD and 3-2WD, allowing somewhat 'manual' selection of gears with the benefits of smooth gear changes by having the electronics do rpm-matching on gear changes.

7 hours ago, WillyChang said:

then again, an electric car eg Tesla doesn't use a gearchange gearbox either, do they?

No gear changes, but they produce maximum and constant torque (electronic limited) from 0-50mph or so. Above that torque falls off almost linearly just like an RC car.

Share this post


Link to post
Share on other sites
3 minutes ago, nbTMM said:

No gear changes, but they produce maximum and constant torque (electronic limited) from 0-50mph or so. Above that torque falls off almost linearly just like an RC car.

yeah thought so :) heard somewhere one early Tesla design for either Roadster or S had spec'd a 2speed slushbox between motor & wheels but it didn't prove reliable under torque.

Otoh... back whilst I twiddled thumbs in uni, the Eng Dept was meddling with gas turbines trying to get them to drive a motorcar ^_^ not much torque from these jet powered hairdryers & they idled at 50,000rpm :lol: so getting them hitched mechanically to a drivetrain would've been an interesting task.

 

Share this post


Link to post
Share on other sites
38 minutes ago, WillyChang said:

Otoh... back whilst I twiddled thumbs in uni, the Eng Dept was meddling with gas turbines trying to get them to drive a motorcar ^_^ not much torque from these jet powered hairdryers & they idled at 50,000rpm :lol: so getting them hitched mechanically to a drivetrain would've been an interesting task.

 

 

You've heard of this, though, I assume?

https://en.wikipedia.org/wiki/Chrysler_Turbine_Car

Share this post


Link to post
Share on other sites
3 hours ago, WillyChang said:

Otoh... back whilst I twiddled thumbs in uni, the Eng Dept was meddling with gas turbines trying to get them to drive a motorcar ^_^ not much torque from these jet powered hairdryers & they idled at 50,000rpm :lol: so getting them hitched mechanically to a drivetrain would've been an interesting task.

 

I believe most current main battle tanks (except those built in western Europe) employ gas turbine engines. 5000 lb ft of torque at 1000 rpm gas turbine engines, too. Seems like Eng Dept succeeded in the end. 

Share this post


Link to post
Share on other sites

Those of us old enough to remember Rover made quite a few gas turbine prototypes. Several can be seen at the motor musem at  Gaydon but they never caught on!

Share this post


Link to post
Share on other sites

Got the motor rpm measurement working on the arduino today. It's measuring both rising and falling edges of the hall sensors on all 3 motor phases, so there are 6 counts per revolution. It calculates the rpm based on a rolling average of the counts in the last 0.13seconds, which results in a pretty accurate and responsive rpm measurement (blue trace on oscilloscope). 

Quick demo shifting between 2nd and 3rd gear only (yellow trace on scope showing gearbox servo signal):

Currently it isn't doing anything to the throttle signal, just passes the incoming value straight through to the ESC so the gear changes are jerky. It shifts up when the motor rpms exceeds 58% of the unloaded motor rpm and shifts down when the motor rpm is below 44% of the unloaded motor rpm. The former is determined by the graph below which shows the torque vs rpm curve at the wheels for the bruiser ratios, assuming the torque curve of an electric motor. Under acceleration we want to select the gear that provides maximum torque at a given rpm. Conveniently this occurs around 58% of the unloaded motor rpm at the crossover between 1st/2nd and 2nd/3rd gears for the bruiser. The rpm where a downshift occurs is subjective - it needs to be lower than the upshift point to provide hysteresis otherwise it would constantly hunt back and forth between gears when you're cruising at a speed around the shifting point. I set it to around 75% of the upshift point for now.

COpGbeZ.png

To do:
-Add a brief timeout period after each gear change so it doesn't change gears again immediately before the motor rpms settle in the new gear
-Override ESC throttle signal and apply correct amount of throttle to synchronise the motor rpm to the next gear
-Reduce ESC throttle signal while braking in 1st and 2nd gear to match 3rd gear
-Don't attempt shifts while braking
-add code to select gears manually or enter automatic/'D' mode, according to auxiliary receiver channel(s)
-implement motor torque limiting/shaping, controlled by auxiliary receiver channel(s)
-implement a way of calibrating the servo end-points and detecting unloaded motor rpm with hardware (buttons, leds), instead of entering numbers directly into the arduino code
-Miniaturise the hardware!

  • Like 1

Share this post


Link to post
Share on other sites

Another demo with 3 gears, a timeout period after each gear change and throttle blip/cut between gears. It isn't the best demonstration because with the wheels in the air as soon as it upshifts from 1st to 2nd it wants to upshift again to 3rd because the motor speed hasn't dropped - the wheels just speed up instantly. Same thing when downshifting. It'll work properly when it's actually driving, you get the idea though.

Another to-do that I forgot:
-Alter the shift points when at partial throttle so it cruises with lower motor rpm and downshifts if you floor the throttle.
 

  • Like 2

Share this post


Link to post
Share on other sites

First untethered drive. Needs tweaking to smooth out the shifts, though half of the problem is probably the untamed fury of the 13.5t :lol: 
 

 

  • Like 2

Share this post


Link to post
Share on other sites

More progress on software today. This might be over the heads of non-engineers however it was a necessary step to work out how to implement a torque-limiting algorithm to make the electric motor behave like an internal combustion engine.

Below left we have the typical torque vs. rpm curves of a DC motor for various throttle positions. Below right we have the throttle signal vs. rpm without any torque limiting active - if we apply 50% throttle at the transmitter, the ESC receives 50% throttle regardless of motor rpm. Pretty straight forward stuff.
9BxMhqE.pngMUNFPqI.png
Below left we have the torque vs. rpm curves that I want the motor to have to mimic an internal combustion engine. I constructed them from linear segments as it simplifies the algorithms/code to implement them. Maximum torque occurs around the middle rpms and there is a drop off in torque at low and high rpms. Also, at low throttle positions the motor has a powerband that extends up to much higher rpms than normal for a DC motor, making it behave more like an internal combustion engine which does not experience the effects of back-emf. Note that all the curves just have to fit within the triangle made by the 100% throttle curve in the previous graph. This is because we can always apply up to 100% throttle signal to the ESC to produce the desired torque, even if only a very small % of throttle is being applied at the transmitter. The reason for choosing a powerful motor now becomes apparent as the bigger the triangle the more flexibility we have in drawing curves which have more torque at higher rpm versus lower rpm. On the right we have the throttle curves that are needed to implement the torque curves - note how when a constant 50% throttle is applied at the transmitter the throttle signal going to the ESC varies from ~10% to ~75% depending on the rpm of the motor to achieve the desired torque.

E3YbocW.pngSL61Xro.png
The result of all this is that I now have a set of equations which take the motor rpm and transmitter throttle % and calculate the throttle % that should be sent to the ESC. Simple!

  • Like 1
  • Thanks 1

Share this post


Link to post
Share on other sites

Front shock mounts are complete (until I decide to build something better/cleverer, which likely won't ever happen :lol:). The white plastic parts flex a bit due to the torsional load, it'll be fine... I got rid of the small leaf spring because it was just too stiff when the suspension was more than 50% compressed. The shocks have the softest springs installed.
yUd5V0F.jpg

Bash plate installed:
XNyX8QQ.jpg

Flexin':
MHiwWU2.jpg

 

 

 

  • Like 1

Share this post


Link to post
Share on other sites

A little driveway run to test out the electronics - it's getting there! This is still a bit too fast to be realistic. The torque limiting doesn't work 100% as expected - I think the TBLE02S has it's own throttle curve or does some kind of rudimentary torque limiting at low rpms because it seemed to have too little torque down low and too much up high. I turned up the torque at low rpms to make it driveable for now. More investigation required.
 

 

  • Like 4

Share this post


Link to post
Share on other sites

I haven’t commented in here yet, but cannot any longer. For me, this might genuinely be the most impressive build i’ve ever seen on here. The concept is amazing and the execution is first class. I wouldn’t even know how to start a set up like this. 

Share this post


Link to post
Share on other sites
On 5/25/2019 at 6:48 AM, nbTMM said:

A little driveway run to test out the electronics - it's getting there! This is still a bit too fast to be realistic. The torque limiting doesn't work 100% as expected - I think the TBLE02S has it's own throttle curve or does some kind of rudimentary torque limiting at low rpms because it seemed to have too little torque down low and too much up high. I turned up the torque at low rpms to make it driveable for now. More investigation required.
 

 

Woah! it definitely sounds like it's going the right way, I can totally hear it going up through the gears like a manual shift.

Share this post


Link to post
Share on other sites

Taking a break from electronics to work on mechanical stuff again. The plan was always to do some de-monsterfication and step 1 to accomplish that was to get some more realistic wheels and tyres. The standard tyres were too big and wide anyway - the new shocks wouldn't fit between the chassis and rear tyre if I mounted them the same way as the front. The front tyres could also rub the body on full lock full compression (more about the body in a later post ;))

Enter RC4WD 1.9" steel 'wagon' wheels and 4.19" Goodyear Wrangler MT/R tyres: I've wanted to do a build with these wheels ever since seeing some on a CC-01. These are called 'sunraysia' wheels in Australia and are hugely popular aftermarket off road wheels. The 4.19" tyres are small enough to fix the fitment issues and monster truck look, but still big enough that they don't look weird with the amount of lift that the bruiser has.

047Z6Qf.png

aDo2tj4.png

The wheels come with hubs that accept a standard 5mm shaft and pin (left). To mount these wheels on the bruiser, I'd need the hex hub for the wheels (right) and an RC4WD bruiser hex conversion kit. For whatever reason, the hex conversion kit came with bolts that were 2mm longer which looked a little weird as too much thread stuck out of the wheel nuts. The shorter bolts that came with the wheels had heads which were just a hair too big for the hex hub, causing them to bind on the hex adaptor, so I ground them down to suit.

gpWvsG8.png

This is the standard 3-lug bruiser hub, which obviously won't work with the RC4WD wheels.
iEnYW6q.jpg

Up front, the hex conversion kit (left) consists of a wheel axle which is basically identical to a standard TG10 wheel axle, except that it has a black finish, and a 12mm hex adaptor which has a skirt to accept the large wheel bearing. Both bearings are reused from the standard hub setup (right).
4RpeZY2.png

The TG10 style axle simply inserts into the steering knuckle from the rear, then the large bearing and drive pin are installed onto the outside of the knuckle. The hex adaptor rides on the axle/pin and outside of the bearing. The inner race does not touch the hex adapter. If you're familiar with 1:10 TC cars then the outer bearing is the opposite to normal - the inner race is stationary while the outer race rotates. 

AbA1GPJ.png

Upon installing the wheel and tightening down the wheel nut, all was not well. The bearings bound up a little so the wheel was hard to turn. Usually you design for a loose fit and add shims to tighten it up until it is just right. Here it was already too tight with no shims, so the only easy solution was to install a thinner bearing. I first tried replacing the inner 10x5x4mm inner bearing with a 10x5x3mm, however the engagement with the dog bone end on the driveshaft was marginal so I decided not to go that route. Instead I changed the 16x8x5mm outer bearing for a 16x8x4mm. This solved the binding issue. Three 0.2mm shims between the hub and bearing had it running just right. The other side required the same treatment however with only two shims.

cATbOkL.jpg

  • Like 2

Share this post


Link to post
Share on other sites

The rear axle required a bit more work. RC4WD supplies replacement rear axles for their hex conversion kit, which have 4mm threaded ends and a drive pin hole. They are otherwise exactly the same as standard - or they would be if you had a real Bruiser. The real bruiser has splined ends where the axle goes into the diff. On the HG P407, the axles have flats instead as this is probably cheaper to make.
ZzsPaLQ.png

Fortunately the outer diameter of the splined section is the same as the rest of the shaft, so I could file some flats in the splined section to make them compatible with the P407. The presence of the splines instead of a smooth shaft seemingly has no impact on functionality.

eeDvamV.png

With that solved, the rear axle can be reassembled with the new shafts, drive pins, rc4wd hex adaptors and wheels, much the same as for the front axle. The fit was too tight and the bearings bound up exactly the same as the front, but I'll fix that later as the primary concern right now is to correct this:
r4QPcO3.jpg

2ykXa4S.png

As seen above, the track of the rear axle is about 15-16mm wider than the front. This isn't too noticeable with the bruisers monster truck wheels but looks weird with skinnier wheels and tyres. I've seen other people narrow the rear axle in various ways usually involving removing the large outer wheel bearing and cutting the cast axle casing with a hacksaw. I didn't want to do that. First, it's a non reversible mod. Secondly, removing one of the wheel bearings isn't a good idea, especially on a car as heavy as the bruiser as it transfers the wheel load via the axle to the bearing near the diff, and the axle will inevitably bend a little when this happens which causes friction and things wearing out faster than they should. I saw the opportunity for a less destructive narrowing mod in the standard bearing arrangement. The standard setup uses a 10x5x4mm inner wheel bearing and a machined aluminium sleeve which the 16x8x5 outer wheel bearing is located by. If you remove the aluminium sleeve, a second 10x5x4 bearing will fit inside the axle casing. It's smaller than the standard bearing but is much better than no bearing.
d2PemeU.png

5r9tWXL.png

This allows the hex adaptor to sit about 5-6mm further inward as the width of the 16x8x5 outer bearing has been eliminated. It also converts the outer bearing back to a conventional configuration where the inner race rotates with the axle and the outer race is stationary, which in turn removes the requirement to use the special hex adaptor. A thinner 12mm hex adapter can be used. As can be seen above, the pin hole in the axle needs to be moved in about 5.5mm to suit the new setup. Much cursing at blunt drill bits later was have this:
12VWXV6.png

bHOLug9.png

The only thing left to do is shim the axles so they don't have end play nor bind the bearings when the wheel nut is tightened. This can be done in 4 places (below) between the e-clips and bearings to either shift the axle inwards (tighter) or outwards (looser). There was about 2mm of space to take up at the two outer locations since I removed the standard machined aluminium sleeve parts and the bearings need to be pressed outwards as the e-clip riding against them prevents the axle being pulled out, while the e-clip riding on the bearings near the diff prevent the axles being pushed in. After taking this pic I actually replaced each of the 10x5x4mm bearings in question with two 10x5x3mm bearings, therefore only a couple of shims were required instead of like 20 shims.
vuRRSsV.png

The next obstacle was that some of the unthreaded 5mm shaft was now poking through the hex, and the wheel hubs only have 4mm holes.
2J35AQS.png

I drilled the wheel hubs to 5mm. Now the wheels fit all the way on, but some unthreaded shaft still pokes out the front of the wheel, so the wheel nut won't go all the way on. I threw some 5mm washers on as a dodgy fix just so the wheel nut would tighten down. I'll probably grind that part down and cut the threads further on to the shaft with a die. The axles need to be cut shorter as all that thread sticking out looks silly.

mPQoNT0.png

The blue hexes can barely be seen once the wheels are mounted 

seuVVKc.png

And most importantly the rear track has been reduced almost to that of the front axle. There's still a couple mm difference although I don't think it'll be noticeable. I could always space out the front wheels slightly (although that increases scrub radius which is unrealistic), or find/machine thinner hexes for the rear.

3Uo9uIh.jpg

That's all for now!

  • Like 2
  • Thanks 2

Share this post


Link to post
Share on other sites

Axle threads extended and excess cut off with the magic of a dremel, drill press, die and hacksaw. Not pretty, but functional. It's times like these I wish I had a lathe!
Dx5rwhj.png

NT5DIh3.png


Rear wheel hubs don't look out of place with the fronts now :) 

hxfbEjT.png

 

  • Like 2

Share this post


Link to post
Share on other sites

The rear shocks were mounted exactly the same way as the fronts were. I had to cut away parts of the rear swing arms for the aluminium spacer to fit - this isn't an issue as the swing arms are just for show and don't bear any suspension loads. I could have reversed the mounts so the shocks were to the rear of the axle instead of the front as there are additional holes in the chassis to do this, however I think it looks better this way as my shock mount is tucked away and better hidden. I'm glad I didn't attempt to narrow the rear axle any further - the shocks are close to the tyres. If one rear wheel is fully drooped and the other is fully compressed the shock body does gently rub the tyre but not enough to be a problem.
IK10mEK.jpg


For springs I used the same configuration as the front with just the main leaf and the softest spring inside the shocks. This was a little too stiff since there is less weight over the rear axle than the front. The roll stiffness is higher at the rear because the springs and shocks are mounted further outwards from the centre of the car than on the front axle. This made a rear wheel lift off the ground first when flexing over uneven ground. Making the rear springs slightly softer than the front would match the roll stiffness between front and rear and give the most flex. Removing the internal shock spring made it too soft, so I kept the soft shock spring but moved the mounting position of the leaf shackle on the chassis. By moving the mounting point forward the shackle lays down slightly which results in a softer spring rate from the leaf. I also moved the shackle mount down slightly to regain some ride height. When the suspension is bottomed out the shackle is almost horizontal so this is as far forward as the shackle can be mounted without it trying to over-centre. The red circles show the original mounting holes on the chassis.

8NKlorF.png

I made an aluminium steering lever to replace the plastic one which flexed a lot. Also fashioned up an aluminium brace which ties the bottom of the lever pivot to the other side of the chassis. The steering is far stronger now. The weak point now actually seems to be the plastic rails that the servos are mounted on - it flexes under steering loads.

fVx0xZ4.png


The front bash plate also got a trim so the front suspension can now compress fully without the edge of the plate hitting the gearbox

MTFmiVW.png

  • Like 3

Share this post


Link to post
Share on other sites

i am surprised you aint done a upfront steering converstion i did it on mine and it is one of my fav upgreades so far

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...