Sequential shifter travel

[Lesthegringo]

Hi All, as my sim rig is going to be in shipment until at least mid December, I'm taking the opportunity to design some better hardware.

Latest in the line of projects is a better sequential shifter with improved feel and tactile feedback. My current one is good, but acts and feels too much like an electrical switch lever, despite having sprung detents. It works great but I want something that has that rifle-bolt clunking action that feels more like you are activating a mechanical device with a positive over-centre movement. Anyone who drove a 70's Triumph will know what I mean.

I have ideas and some provisional designs but because I am not able to sit in the rig one peice of information that escapes me is the actual travel of the lever at the hand; in other words, from a neutral, stationary position how much fore and aft movement is comfortable for a sequential lever. My initial thoughts are maybe 100mm in each direction, but would like some feedback from you guys. I intend to make the lever some 400mm from the fulcrum point, in order to reduce the angular movement, maybe even longer, but that does not affect the throw.

So based on what you use, what feels comfortable? Subjective I know but it's very difficult to imagine something without any reference points at all here.

The design would be made from simple materials like extruded aluminium, bits and bobs off Ebay and some 3D printed parts, but built to be strong and chunky so that it feels meaty and solid. I will happily put the CAD drawings up on Grabcad once done in case anyone wanted to copy, the intention is to design something with the minimum of machining needed. That means it won't be pretty but it will hopefully function nicely.

Cheers

Les

 

[Elaphe]

I believe that a shifter need a minimum travel to prevent accidental activation. However, a very long travel would make the process of shifting too slow. I suppose it also depends of the design of the shifter, for instance, how the mechanics work to provide the tactile feedback and, of course, the lenght of the lever. A shorther throw could also be lenghtned by installing an extended lever. On the other hand, a compact design with a not very long lever is easier to install on most rigs.

 

[stigs2cousin]

That´s an interesting project that I´ve been dreaming about for some time now.
Since my fabrication skills and tools are limited it´s still a dream.

On topic:
I believe the "satisfieing clunk" could be achieved by shoving an piece of steel around like you would with the shift rod in a real gear box.
My personal car has an IRP quick shifter that is connected to the gearbox with steel parts all the way.
You can feel the syncromesh rings do theyr job ( more or less) and the gears clunking into place.

In the sim version we could use the shift lever to push a steel block past some spring loaded detents.
Maybe even put spring pressure on the steel "slider" itself and let it "climb over the detent" for example a mushroom head bolt.
If a corresponding chamfered whole was pushed over he bolt head it would feel more like a gear clicking into place than just a slick linear movement.

For me this "slick, switchlike" feeling as you discribe above is the only fault of my Heusinkveld sequential shifter.
BTW, with the longe shift rod and knob it has apr. 20mm travel at the top.
More would be too much in my opinion as it would slow down the shift considerably.
Even now I feel hardly up too snuff in the competetive groups I´m racing which of course is only to blame on my ( top level) gear

MFG Carsten

 

[blekenbleu]

minimum travel to prevent accidental activation

Accidents can be mitigated by increasing return spring rate,
but any sequential shift lever forfeits speed
by time to reach from steering rim,
unless one steers single-handedly, forfeiting control...

 

[Elaphe]

Some say this is overengineering, but it is a way to achieve that tactile feedback and mechanical feeling.

 

[stigs2cousin]

I´ve seen that too, but didn´t bite because of the pricetag.
I think it would be possible to build something like that with a hand drill and some files.
( but I only would try if we get another 3month lockdown )

This one looks the part in my opinion, the narrator also mentions the " feedback problem"
Problem is that they ( like most others) just push a dented rod past a detent, there a no mechanical movements when shifting.

MFG Carsten

 

[Lesthegringo]

Thanks guys. That first solution is certainly different, but ultimately it is only another variation on the sprung ball in detent groove theme, albeit a neat one. The Jinx unit is also nice, but the way that is constructed is actually basically what my old homemade unit is, except with beautifully machined parts instead of my crudely machined bits. Both lack exactly what the narrator says, that 'clunk' as the gear engages at the end of the travel which is what I am trying to design in.

As he mentions, you have a number of tactile clues when moving a shifter. The lever is just linked to a disengaged mechanism in the gearbox that only moves the rings and synchros during part of the lever travel. Breaking it down to how I look at it, the tactile forces are the following

The self centering force that keeps the lever in the neutral position
the weak 'over centre' feel of moving the lever away from that neutral position
a short amount of movement with a constant spring force trying to return it to the centre position
a sudden increase in force as the mechanism actually starts moving the gear parts with a mechanical feel of the lever coming into contact with the mechanism
the 'over centre' feeling as the gear is actually engaged (and on an H pattern locks in place with a detent) - the 'clunk'
the hard stop as the change mechanism reaches the end of its travel
the constant rate smooth movement from the limit of travel back to near the centre position
the weak 'over centre' feel of the lever moving back to the rest position

One last very subtle feel is that the lever should not vibrate back and forth around the centre position like a knife sticking out of a table; gearboxes are ful of syrupy oil that damps movements like that out very effectively. As a result both my homemade H pattern and sequential shifters have dampers built in to deaden the feel in that respect

I do have a solution to to that sequence of movements laid out above that I hope will result in the correct feeling. I will be making it and trying it, and the Jinx unit actualy helped me in terms of understanding how to fix it to the P1-X rig that is still unopened in the container with all my other personal effects.

As for the original intent of the post, that also has been addressed by the Jinx video; the 100mm I initially imagined each way is far too much, the Jinx unit looks like it uses less than half that. That's what I will work with then, max 50mm throw each way. and by simply making an adjustable fulcrum and some matching actuator holes on the lever, I can fine tune that at least for my initial purposes.

Wish me luck

Les

 

[stigs2cousin]

As he mentions, you have a number of tactile clues when moving a shifter. The lever is just linked to a disengaged mechanism in the gearbox that only moves the rings and synchros during part of the lever travel. Breaking it down to how I look at it, the tactile forces are the following

The self centering force that keeps the lever in the neutral position
the weak 'over centre' feel of moving the lever away from that neutral position
a short amount of movement with a constant spring force trying to return it to the centre position
a sudden increase in force as the mechanism actually starts moving the gear parts with a mechanical feel of the lever coming into contact with the mechanism
the 'over centre' feeling as the gear is actually engaged (and on an H pattern locks in place with a detent) - the 'clunk'
the hard stop as the change mechanism reaches the end of its travel
the constant rate smooth movement from the limit of travel back to near the centre position
the weak 'over centre' feel of the lever moving back to the rest position

Don´t you think you take this a little serious?

Good luck with your solution, hope it pans out like planned ( and I´m interested how it works out and maybe to copy it if I may)

About hte shifters travel I compared the video with my Heusinkveld shifter and the travel in the video is not much larger than the HSV which has measured apr. 20mms.

Even my ( H-Pattern) quickshifter in my car has apr. 30mms of travel with a stick longer than the Jinx´s.
So, I suggest you design at least one position with a travel that short just to see how it feels and to keep it at least not too timeconsuming to shift.

Good Luck Carsten.

 

[Lesthegringo]

Serious...? Me....?!!

Thanks for the feedback on the travel. Thankfully the throw is easily adjusted by making the actuation mechanism sit further up the lever, but the downside is that means a larger overall envelope plus there is a danger that you lose rigidity. I don't think it will be an issue for me though, looking at the P1-X rigs online it looks like I have a reasonable amount of space to play with.

I will probably try and make an all metal unit in order to maximise the strength, but what I will aslo do is design it with standard meterials and parts and some 3D printed parts in mind to make it so anyone who wants to do a garage version without recourse to any complicated machining can do so with hand tools and some limited 3D printing

Feel free to copy and improve anything I make!

Les

 

[Lesthegringo]

I have some progress on this, but have to concede that at least at the moment, a good solid unit is going to be difficult without some machined metal parts.

The issue is not making it work how I believe it should, but the forces on some of the parts. I suspect even good 3D printed parts will have a limited lifespan unless you turn down the forces. I reckon it will still feel better than what I have, but will still be a weaker feel than a nice solid unit, undermining the overall sensation.

From what I'm looking at it still will not be super engineering, but garage machinists will have a bit more work than just some drilling. Anyone with access to a desktop mill will not have a problem. I need to build a couple of test units to see where the weak points are to see which bits really can be 3D printed and which do need metal. It may be that 3D printed parts with some small diameter bolts will be enough.

Interesting project, hopefully with a successful outcome

Les

 

[Lesthegringo]

Guys, I need to know the thickness of the sandwich plates for the P1-X rig - I believe that they are 10mm, can you confirm? I need the dimension for the mounting offset needed for the lever fulcrum

Cheers

Les

 

[EsxPaul]

I think 10mm is correct if this PDF is accurate.

 

[Lesthegringo]

That's great, thanks - I have the P1-X version and it doesn't have the same amount of info, so this is very helpful

Cheers

Les