[HELP] Generic Building Help Topic

[Jundis]

3 hours ago, Erik Leppen said:

One solution is to control the first and last axle independently, and use a beam to interpolate the intermediate axles. Works like a charm on small models. On large models, probably needs a stronger way to do this.

This really is a simple and great solution! I almost gave up on this model, but now wanna try if it works, thank you very much!

[hawkwind]

Could someone give some explanation on how to efficiently use Bricklink Studio 2.0? I wanted to try designing something in there and then buying the parts and tinkering with it physically, but I am having so much trouble inserting pins into pinholes, getting things at the same elevation, etc. It takes me about five minutes to get a differential in a 5x7 frame.

[Jurss]

that's why I'm mainly using LDD, then in studio do some last corrections, mainly add parts which are not in LDD. Also do rendering of digital model.

[Jundis]

3 hours ago, hawkwind said:

Could someone give some explanation on how to efficiently use Bricklink Studio 2.0? I wanted to try designing something in there and then buying the parts and tinkering with it physically, but I am having so much trouble inserting pins into pinholes, getting things at the same elevation, etc. It takes me about five minutes to get a differential in a 5x7 frame.

Have you activated the "Snap" function (look in the upper panel)? Then it should be easy to do. :-)

[Jurss]

Somehow that function didn't worked for me as I expected. Maybe it s corrected in some last couple months.

[Jonfensu]

Does the new 24t-16t red differential mesh with the 20t double bevel gear with 16t side? I am wondering because the new differential protudes slightly on that side compared to the old version

[anyUser]

1 hour ago, Jonfensu said:

Does the new 24t-16t red differential mesh with the 20t double bevel gear with 16t side?

No

[TechnicBrickPower]

 

On 6/18/2020 at 6:59 AM, Leo12 said:

I like this solution. Thanks for the help!

 

On 6/18/2020 at 2:10 AM, Leo12 said:

Yes, that’s what I’m looking for. Ideally it would have very little change in gear ratio, but definitely not too much gearing down(unlike a worm gear). Both sides should be able to rotate in both directions, and it needs to be fairly torque resistant. If it helps, I’m using it for a three speed differential gearbox. I need to make sure the axle from the differential can’t back drive when the driving ring is disengaged.

 

On 6/18/2020 at 4:22 AM, ukbajadave said:

First attempt, Yellow is input, red is output. Turning yellow axle turns red axle freely. Forcing red axle results in skipping teeth between differential and tan 20z.

 

Here's a smaller way of doing it.

 

Video

https://www.flickr.com/video_download.gne?id=50229833833

 

 

[hawkwind]

Is there a solution to your cv joints rattling like an earthquake when your cars steers?

[ukbajadave]

9 hours ago, TechnicBrickPower said:

Very compact, I must get some of those old diffs 

I still think this seems like witchcraft though...

[thekoRngear]

Can someone tell me what are the jobs of these two parts in a gearbox?

Edited August 16, 2020 by thekoRngear

[pleegwat]

The blue 20z is a clutch gear, similar to the red 16z. A clutch barrel links it with the axle it sits on, otherwise it spins freely.

The yellow piece on the left is a clutch barrel extender, so the clutch wheel can be further from the barrel.

[thekoRngear]

On 8/16/2020 at 5:45 PM, pleegwat said:

The blue 20z is a clutch gear, similar to the red 16z. A clutch barrel links it with the axle it sits on, otherwise it spins freely.

The yellow piece on the left is a clutch barrel extender, so the clutch wheel can be further from the barrel.

Thanks.

In general, is it possible to have the gearbox completely bypassed when in neutral mode?

Edited August 22, 2020 by thekoRngear
No one cared

[lcvisser]

On 8/16/2020 at 12:00 AM, TechnicBrickPower said:

Here's a smaller way of doing it.

 

Video

https://www.flickr.com/video_download.gne?id=50229833833

 

 

This helps. I finally got some time to figure out how this works.

The transfer ratio from the left to right is 1 - 2/N, where N is the ratio between the differential ring gear and the driving gear. Here N = 24/8 = 3 and we have a transfer ratio of 1/3. Conversely, the transfer ratio from right to left is the inverse, so 3. If you try carefully enough, you can verify that the mechanism is back-drivable with a transfer ratio of 3:

---> [ 1 - 2/N ] --->
 1                2

N = 3:

w2 = 1/3 w1
T2 = 3 T1

If you try to impose a velocity on axle #2, you feel 3 times the torque that is present on axle #1, including the internal friction of the mechanism. The reason why it is so hard to back-drive is that the gears bind up internally, because they need to overcome 3x the internal friction of the mechanism. If you add more gears, it'll get even more difficult to back-drive.

If you change the 8z for a 16z on the other end of the differential, you have N = 1 and the ratio is -1: direction is changed, but 1:1 speed and torque ratios. In this case the mechanism is back-drivable and feels the same independent of which axle you drive.

I'm curious to see N = 2, which is a singularity. In theory you should be able to rotate axle #1 and nothing would happen on axle #2, while conversely axle #2 should be immovable. For N = 2 you'd need to have an 8z mesh with the 16z ring gear of the differential, or a 12z with the 24z ring gear, without introducing any change of direction with respect to the mechanism shown above. Should be doable with some u-joints or something. 

Edited August 17, 2020 by Ludo Visser
Typos

[hawkwind]

Now I have a new problem: if I load in an LDD model with parts such as the forked suspension arms, they load in, however I cannot find them in the LDD parts inventory. Am I doing something wrong?

[Oliver 79]

On 8/28/2020 at 9:54 PM, hawkwind said:

Now I have a new problem: if I load in an LDD model with parts such as the forked suspension arms, they load in, however I cannot find them in the LDD parts inventory. Am I doing something wrong?

Are you in LDD 'extended'? Some parts don't show up in the normal view. You can also search by part number. You can find them on Bricklink. 

[lmdesigner42]

On 8/17/2020 at 9:05 AM, Ludo Visser said:

I'm curious to see N = 2, which is a singularity. In theory you should be able to rotate axle #1 and nothing would happen on axle #2, while conversely axle #2 should be immovable. For N = 2 you'd need to have an 8z mesh with the 16z ring gear of the differential, or a 12z with the 24z ring gear, without introducing any change of direction with respect to the mechanism shown above. Should be doable with some u-joints or something. 

I tried this out yesterday with the 8z-16z mesh using two 12z gears to connect the 8z gear to axle #1, and that is exactly what happens. It feels really weird having axle #2 seemingly locked in place with no support. With respect to N=3, I think another factor is that axle #2 is rotating in the opposite direction to the differential casing (if input is 1, diff output is 1/3, axle 2 is x, (x+1)/2 = 1/3, x = -1/3). If you wiggle axle #2 back and forth, it's possible to get the casing to rotate.

[lcvisser]

@lmdesigner42 Cool, thanks for trying! I forgot about this little experiment, glad to know theory agrees with reality    I wonder if there's some use case where this singularity can be actually useful, maybe something where you'd put a ratchet normally...?

The rotational direction (or better: direction reversals) definitely plays a role: I guess the weirdness of stiction forces has something to do with that, but I'm not a mechanical engineer. Maybe someone more qualified can enlighten us.

[heyitsdisty]

Hi folks, apologies if a similar question has been asked before:

I usually build tank-steering robots (around 26 by 26 studs width/length) with one PF motor driving the left wheel(s) and another driving the right. Each motors is hooked up to a PF IR receiver slot so I can drive it around.

However, I'd like to free up one of the two IR slots by having both wheels controlled by one motor. I need a mechanism that drives both wheels forward when I push the transmitter stick up, but drives them in opposite directions when I push down.

It's been hard to find a reliable and not-super-bulky mechanism that lets me do this. I've tried Lego Technic Mastery's two-way to one-way rotation converter, but it seems to fail immediately under motor power.

I've also tried a setup of using two 8-tooth gears next to one another and driven via a loose worm gear on an axle (so switching the direction of the axle pushes the the worm gear from one of the 8-tooth gears to the other) but it's just far too slow for the kinds of bots I want to build - even when gearing up the driver axle.

And I've tried using a differential with one side going off to a gear-and-ratchet mechanism - so pushing up drives both wheels but pushing down locks one of the wheels in place while the other keeps moving - but it seems the differential always slips and/or the ratchet locks the gear in both directions!

Does anyone have any ideas?

[chekitch]

16 hours ago, heyitsdisty said:

Hi folks, apologies if a similar question has been asked before:

I usually build tank-steering robots (around 26 by 26 studs width/length) with one PF motor driving the left wheel(s) and another driving the right. Each motors is hooked up to a PF IR receiver slot so I can drive it around.

However, I'd like to free up one of the two IR slots by having both wheels controlled by one motor. I need a mechanism that drives both wheels forward when I push the transmitter stick up, but drives them in opposite directions when I push down.

It's been hard to find a reliable and not-super-bulky mechanism that lets me do this. I've tried Lego Technic Mastery's two-way to one-way rotation converter, but it seems to fail immediately under motor power.

I've also tried a setup of using two 8-tooth gears next to one another and driven via a loose worm gear on an axle (so switching the direction of the axle pushes the the worm gear from one of the 8-tooth gears to the other) but it's just far too slow for the kinds of bots I want to build - even when gearing up the driver axle.

And I've tried using a differential with one side going off to a gear-and-ratchet mechanism - so pushing up drives both wheels but pushing down locks one of the wheels in place while the other keeps moving - but it seems the differential always slips and/or the ratchet locks the gear in both directions!

Does anyone have any ideas?

I have no purely mechanical ideas, but if you have the switch (8869).. One motor directly on output and that motor controls one side but also controls the switch.. The other motor connected via the switch controls the other side... That should work i think.. 

[heyitsdisty]

7 hours ago, chekitch said:

I have no purely mechanical ideas, but if you have the switch (8869).. One motor directly on output and that motor controls one side but also controls the switch.. The other motor connected via the switch controls the other side... That should work i think.. 

This is a pretty neat solution, thanks. Sadly I don't have the switch, but I may well end up picking one up.

One issue; presumably I'll need a 24-tooth clutch gear to turn the switch without stalling the drive motor. Is there an alternative way to let the switch slip that's less friction-ey? (Even if I add a gear with a friction pin on the switch-fed motor to balance out the friction on the switch-turning motor, I'd rather not be intentionally handicapping the speed/power of the drive, if possible.)

[chekitch]

7 hours ago, heyitsdisty said:

This is a pretty neat solution, thanks. Sadly I don't have the switch, but I may well end up picking one up.

One issue; presumably I'll need a 24-tooth clutch gear to turn the switch without stalling the drive motor. Is there an alternative way to let the switch slip that's less friction-ey? (Even if I add a gear with a friction pin on the switch-fed motor to balance out the friction on the switch-turning motor, I'd rather not be intentionally handicapping the speed/power of the drive, if possible.)

You are right, but I remember making something to flip a switch via rotation without a clutch a few years ago.. But I can't remember it now.. 

I'll try later, if it works, I'll write..

[chekitch]

@heyitsdisty

Sorry, I checked.. I ended up just using a lever  (flexibly connected with 45590) that was "slapping" the switch lever. It was just a weird contraption so it was ok for me, but is not of use to you...

But I remembered something I saw... I didn't try it but if it is just what you are looking for if it works robust enough  Lego Two-Way to One-Way Rotation Converter Mechanism - Lego Technic Mastery

[Jeroen Ottens]

How about this? Rotating the grey axle will push the green lever with the red clutch gear either left or right. You then have at least two different outputs that you can use to control one track going always forward independent of the direction of the motor, while the other track goes backwards and forwards with the motor.

[iLego]

6 hours ago, Jeroen Ottens said:

How about this? Rotating the grey axle will push the green lever with the red clutch gear either left or right. You then have at least two different outputs that you can use to control one track going always forward independent of the direction of the motor, while the other track goes backwards and forwards with the motor.

This will work fine for light loads. I think the request was for a mechanism which can take more load.

Edited October 6, 2020 by iLego