Kumbbl Posted January 29, 2013 Posted January 29, 2013 Hello all, subject says quite all: i want to build a switcher between an input- and an output technic axle which allows to change the rotation direction of the output: so in one position the output axle rotates in the same direction as the input one and in the other position the output rotates reverse. Well, this seems to be quite easy with a gear driving ring, just tried it out in 15 minutes, but there is one KO-criteria for me: such a solution has always backlash of about a quarter turn (this is imanent for the driving ring) which is no problem for speed transmissions (e.g. for the reverse gear) but for my purpose this is not acceptable: i need a solution with minimal backlash. Well, an alternative is a solution similar to the unsynchronized gearboxes from the old supercar-days before the 8880 - quite simple to build but at least my try is really not compact but need " a lot" of space... Therefore now my question to all the building-gurus out there: Any idees or hints how to build a as compact as possible rotation-direction switcher with as less as possible backlash? studded would be most welcome... Is my problem-description described in an understandable manner? Many thanks in advance Klaus P.S. Please apologize my sometimes clumsy english writing... Quote
Balrog Posted January 29, 2013 Posted January 29, 2013 Perhaps you could describe what this mechanism is for and why it is important to have as less backlash as possible. That would help understand your problem better and increase your chance for an answer. Quote
Kumbbl Posted January 29, 2013 Author Posted January 29, 2013 well, here we go: this is for a MOD of the 8880 supercar and its 4WS: in real 4WS-implementations the behavior of the rear steering changes depending on the car-speed: with low speed the direction of the rear steering is contrary to the front-steering (behavior of the standard 8880) to achieve a small turning circle and a good manoeuvrability whereas with high speed the direction of the rear steering is the same (but with much smaller steering angle) as the front-steering to get high stability for lane-changing on the highways etc... Now clearer? Quote
MarkusOSX Posted January 29, 2013 Posted January 29, 2013 (edited) Hi,first i was thinking of, was something like this One drive ring must be switched for a direction. If none is switched you have a "neutral". Greetz Markus Edited January 29, 2013 by MarkusOSX Quote
Kumbbl Posted January 29, 2013 Author Posted January 29, 2013 @Markus: Thanks for your reply, but hmm, just tried out and the result is not satisfying: - regardsless of the drivering positions the output has always same direction as input - or i have not understood your approach... - the backlash is as high as described in my initial posting - in my humble opinion a solution with no or minimal backlash can not be achieved with driverings... or i do not understand driverings.. ;-) Best Regards, Klaus Quote
allanp Posted January 29, 2013 Posted January 29, 2013 You could try this: In position 1 the output turns the opposite direction as the input at a 1:1 ratio. In position 2 the output turns the same direction at a ratio of 1:1.5625 reduction. I guess you would want to reverse the direction of the output to work as you want it with 8880. Quote
MarkusOSX Posted January 29, 2013 Posted January 29, 2013 On 1/29/2013 at 11:08 AM, Kumbbl said: @Markus: Thanks for your reply, but hmm, just tried out and the result is not satisfying: - regardsless of the drivering positions the output has always same direction as input - or i have not understood your approach... - the backlash is as high as described in my initial posting - in my humble opinion a solution with no or minimal backlash can not be achieved with driverings... or i do not understand driverings.. ;-) Best Regards, Klaus Sorry, wrong pic ;o) Quote
Alasdair Ryan Posted January 29, 2013 Posted January 29, 2013 On 1/29/2013 at 11:49 AM, MarkusOSX said: Sorry, wrong pic ;o) That's better,yes it does help on having the right picture. Quote
Kumbbl Posted January 29, 2013 Author Posted January 29, 2013 On 1/29/2013 at 10:38 AM, MarkusOSX said: Hi,first i was thinking of, was something like this One drive ring must be switched for a direction. If none is switched you have a "neutral". Hi Markus, sorry, but it seems i'm too stupid for your hints ;-) If i understand your new picture (above) right, then inside each driveing ring is a so called "Technic Axle Joiner Inline", so there is a fixed connection between the two red axle endpoints, right? How you would get the reverse direction? IMHO this new picture does simply nothing... sorry, i'm far from offending you, but i do not understand what do you want to tell me - probably my fault... BR Klaus On 1/29/2013 at 11:33 AM, allanp said: You could try this: In position 1 the output turns the opposite direction as the input at a 1:1 ratio. In position 2 the output turns the same direction at a ratio of 1:1.5625 reduction. I guess you would want to reverse the direction of the output to work as you want it with 8880. er... i believe, i do not understand your solution... probably I'm not experienced enough for it... may i ask you for some explaining words? Quote
allanp Posted January 29, 2013 Posted January 29, 2013 (edited) It is a crash gearbox, similar to the ones used before 8880. Imagine that in both positions the input shaft is turning clockwise. In position 1 the input shaft is in it's forward position. The 16t gear that is on the input shaft is meshed with the 16t gear of the output shaft. So when the input turns clockwise, the output shaft turns counter-clockwise at the same speed. In this position the other gears turn but are not used. They are only used in position 2. In position 2 the input shaft is in it's back position. It is now no longer meshed with the 16t gear on the output shaft. It is meshed instead with the 20t tan gear. So when the input shaft turns clockwise the 20t tan gear turns counter clockwise at a slightly slower speed. This then turns the next 16t gear which is meshed with the other 20t tan gear on the output shaft. This will then turn the output shaft clockwise at a slower speed and same direction as the input shaft. Edited January 29, 2013 by allanp Quote
Alasdair Ryan Posted January 29, 2013 Posted January 29, 2013 (edited) Will this work for you ok? If you require the same speed you could gear it up or down to your requirements. Edited January 29, 2013 by Alasdair Ryan Quote
allanp Posted January 29, 2013 Posted January 29, 2013 On 1/29/2013 at 12:45 PM, Alasdair Ryan said: Will this work for you ok? If you require the same speed you could gear it up or down to your requirements. Nope! Quote
Alasdair Ryan Posted January 29, 2013 Posted January 29, 2013 On 1/29/2013 at 12:52 PM, allanp said: Nope! Go away...... Quote
Kumbbl Posted January 29, 2013 Author Posted January 29, 2013 On 1/29/2013 at 12:45 PM, Alasdair Ryan said: Will this work for you ok? If you require the same speed you could gear it up or down to your requirements. to be honest: nope - reasons: - if i suppose the geay axle of the driving ring soild be the input, then there are two outputs, each with other direction but i need one output which changes direction - the main disadvantage is the high backlash which can not be avoided IMHO with driving-ring based approaches Thanks to all offering me driving-based solutions: But i have already built a driving-ring-based solution - all of them are basically the same - but all of them have this high backlash which makes them unuseable for me - reason see above... therefore my question for other innovative ingenius ;-) solutions with as low as possible backlash... best regards Klaus Quote
MarkusOSX Posted January 29, 2013 Posted January 29, 2013 (edited) to explain.... The green colored shows you the working drivetrain depended on the position of one of the red drive ring If the drive rings in another position, the direction is changed The important thing is, that only one drive ring got another position, the other stays in the middle Here you can see the parts that are used How this will make you understanding the function. Greetz Markus Edited January 29, 2013 by MarkusOSX Quote
Hopey Posted January 29, 2013 Posted January 29, 2013 Would I be correct in saying that with any solution involving selecting one of two gear trains that the whole thing's going to get out of sync pretty quickly? i.e. if the wheels aren't all pointing exactly ahead when you switch over, then they aren't going to be aligned. I'm thinking of something involving linkages instead. I don't have a full solution yet, or the tools to do a proper model, however if you could attach the steering arms to some sort of offset cam, then make it so that you can somehow switch (slide?) them to the other side, then the same input should have the opposite steering effect. Not sure what the mechanism would be. Might need a crudely drawn image: If the grey things represent some linkage to the steering arm, then turning the blue axle will steer. If you could somehow switch the attachment point from A to B, you could reverse the steering direction, with no backlash and without losing alignment. Not sure how though. Quote
Kumbbl Posted January 29, 2013 Author Posted January 29, 2013 On 1/29/2013 at 1:19 PM, MarkusOSX said: to explain.... The green colored shows you the working drivetrain depended on the position of one of the red drive ring If the drive rings in another position, the direction is changed The important thing is, that only one drive ring got another position, the other stays in the middle Here you can see the parts that are used How this will make you understanding the function. to be honest again: nope - Markus, have you built your solution in real live and tested? Take a look at your first picture, which the second driving-ring-element in blue: with your working green drive train you will have (supposing the left axle is the input and the direction is clockwise): - the green driving ring-axle is connected so therefore power goes to the middle left 16t gear which has direction counter-clockwise - therefore the green output axle has again clockwise direction (= same as input) Now take a look at the second picture where all three axles are green coloured (supposing again the left axle is the input and the direction is clockwise): - then the left middle clutch gear is jst an idler (because the middle driving ring is not positioned) (rotating counter-clockwise) to the second driving ring which is now connected so power goes to the right clutch gear (rotating clockwise) which - then drives the middle right 16t gear (counter-clockwise) which - then drive the output axle clockwise (same as the input) To make a long story short: regardless of the position of the drivings rings in your solution the output has always same direction as input... What i have i overseen, where is my error in reasoning? I have built your solution exactly as your pictures describe and it proves the story above... On 1/29/2013 at 1:40 PM, efferman said: you want do something like this? exactly - but not studdless but studded - but same prinziple.... Quote
DLuders Posted January 29, 2013 Posted January 29, 2013 @ Kumbbi: Perhaps you could adapt Sariel's Direction Sensitive Gearbox ( ) -- it is a "Compact 2-speed gearbox, where gears are changed when the direction of motor’s rotation changes."This gearbox utilizes a working principle of a ratchet with gears. Only one end of the ratchet is attached, the other one is loose. There are gears on both ends of the ratchet, and it causes it to tilt accordingly to the direction of rotation of the axle it is attached to. To create a pressing force, that helps the ‘loose’ ratchet gear to engage receiver gears, an axle pin with friction is used. That friction consumes some of the motor’s torque, but creates an effect of the ratchet being strongly pressed to whatever obstacle it comes in contact with – in this case, the receiver gears. "The design has mupliple advantages: is extremely compact, pretty robust and needs no external gear changing mechanism. There is one, quite crucial disadvantage: with default gears setup, it can only drive iutput in one direction. So, it is possible to have e.g. vehicle that runs at two different speeds, but only forwards. Obviously, the output can have its direction of rotation changed by a separate mechanism, but that is very troublesome. On the other hand, it is possible to use even numbers of gears for both outputs of the receiver gears, thus obtaining forwards and reverse gear with different gear ratios. Both these options have a limited range of application, however the default setup may be useful for compressors, where the direction of rotation doesn’t matter, but the speed does. "As of the end of 2008, this gearbox design was used in one of my constructions only, in a SandRail, a small buggy able to drive only forwards, where its ability to handle high torque was tested. The performance of the gearbox was satisfactory, and the main conclusion was that it works better when braced in bricks that when braced in liftarms. An ldr (MLCad format) model of the gearbox is available here." Quote
efferman Posted January 29, 2013 Posted January 29, 2013 so you should use sliding axles. while i have experimented with this chassis i have tried many possibilitys and this was the only which had worked properly Quote
Alasdair Ryan Posted January 29, 2013 Posted January 29, 2013 (edited) Will you not get any backlash from using sliding axles? I tested my idea and I did not find any backlash. Edited January 29, 2013 by Alasdair Ryan Quote
efferman Posted January 29, 2013 Posted January 29, 2013 no, with my final solution i did not have a backlash. Quote
Kumbbl Posted January 29, 2013 Author Posted January 29, 2013 On 1/29/2013 at 2:07 PM, Alasdair Ryan said: Will you not get any backlash from using sliding axles? I tested my idea and I did not find any backlash. hmm, can not imagine ... tested several driving-ring based solutions and all of them have the backlash the driving ring has within the connected clutch-gear: You can simply test it: connect the driving ring to the clutch-gear so there is the power-connection. Then fix the clutch gear with a finger and try to rotate the axle - you will always notice that you can rotate the axle fpr about a quarter turn - must be because a driving ring has only 4 "tooth" which can grab the same 4 inner tooth of the clutch gear - and exactly the distance from one of these tooth to the next one will be the backlash... and this backlash can IMHO not be avoided when using driving-ring based solutions - regardless if simple or complex ones... or do you have another opinion? so therefore i can not imagine that you can test your solution without any backlash... ;-) anyway: in the meanwhile i agree with efferman: cause of the backlash problem only a sliding axle solution will work or something like the gearbox of Sariel - i have already tried this yesterdays, very cool and ingenious solution (as many of sariel) and can quite easily adapted to switch directions by by using his ratched as a somekind lever but then fixing the lever in one of the position comes up as new problem... I have a working solution fitting in the 8880 - based on a sliding axle - but for my taste it has some drawbacks: - not synchronized (sariels gerabox above is per se synchronized) - needs quite a lot of space so i must place the seats of the 8880 one stud more outside which could be acceptable but coudl also not ;-) Therefore my search for a really compact solution without backlash.... maybe i should think a little bit more about allanp's idea or sariels gearbox... nevertheless already now many thanks to all of you for your valueable hints and ideas! Klaus On 1/29/2013 at 2:20 PM, efferman said: no, with my final solution i did not have a backlash. efferman, this sounds very promising! may i ask you for some construction details or zooms pictures of the inner details? Quote
MarkusOSX Posted January 29, 2013 Posted January 29, 2013 On 1/29/2013 at 1:56 PM, Kumbbl said: to be honest again: nope - Markus, have you built your solution in real live and tested? Damn... sry, you´re right. No i did it only virtual Quote
Kumbbl Posted January 29, 2013 Author Posted January 29, 2013 On 1/29/2013 at 1:19 PM, Hopey said: Would I be correct in saying that with any solution involving selecting one of two gear trains that the whole thing's going to get out of sync pretty quickly? yes, indeed, this could happen - therefore my idea was changing direction only when car is driving straight ahead, i.e. both steerings are in center position - then they will remain in sync... but of course you are right, this could happen... Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.