Voron TAP - New No probe mod from the Voron Design Team

[Buurman]

3 minutes ago, Wick said:

looks good but I miss the side supports from the magnets, so it could wiggle a bit sideways as theres only the MGN9.... bad for resonances?

Nah, the hiwin rails don’t move at all, it’s amazing. The magnets on the original don’t do much for stability. 

[Penatr8tor]

4 minutes ago, Wick said:

looks good but I miss the side supports from the magnets, so it could wiggle a bit sideways as theres only the MGN9.... bad for resonances?

Actually it's an MGN 7 but still... I don't think there's enough side play to matter. I think the Voron design has magnets not so much to eliminate or reduce side play but more to create a detent or break point so that the toolhead doesn't lift with high extrusion pressure. I mean I could be wrong. Just thinking out loud.

[Wick]

1 minute ago, Buurman said:

the hiwin rails don’t move at all

that wouldn't be good,  from the pic I see a mgn7 instead of mgn9, the magnets should give side stability as it "locks" there in printing position (think I read about this somewhere, hit me dead, can't remember where...)

[Penatr8tor]

1 hour ago, Wick said:

that wouldn't be good,  from the pic I see a mgn7 instead of mgn9, the magnets should give side stability as it "locks" there in printing position (think I read about this somewhere, hit me dead, can't remember where...)

I remember that the Voron team spent quite a good amount of time going over the reason for the magnets and also their orientation during the release live stream. 

[Buurman]

I believe the magnets are more to create a fixed position for the movable part to fall back onto. 

If it were weight only, it could happen that it still differs it not falling fully to its end, or even bouncing up a little of the magnets were not there. But side to side stability, nah, I dont believe that.

It looks like the CNC version has something in the top that prevents "twisting" so that the two part are REALLY close to eachother. I can be wrong on everything... but building the TAP a few times now and mostly taking it apart also.. my engineering brain cant help to wonder on the choices made.. I understand them, just hope improvements are still made in time. I do like the design, dont get me wrong, highest respect for these guys.. 

Its just so heavy.. the rail choice and the magnet/screw combination.. argh.. dont know..

But yeah, my dad always says, the best sailors are always the ones on the shore.. I should shut up  

[Buurman]

On 3/23/2023 at 4:11 PM, Penatr8tor said:

This is interesting... I might have to get one of these. Looks compelling.

CNC Voron Tap – Chaoticlab

uuuhhh...

Pre-ordered  I think very reasonable pricing, and shipping... (my wife wont agree.. but then again.. who cares   if she doenst know... muahahaha!

 

BLACK!!

[Penatr8tor]

@Buurman Agree, Pre-Ordered one as well. I went with the red one. 

I guess we will meet up here once they arrive. 

[mvdveer]

4 hours ago, Buurman said:

Pre-ordered

Oh my, aren't we all so predictable - PRE - ordered - BLACK

[zogzogzog]

As did I - red.

Price-wise, it's very, very reasonable - I would have charged more...

[zogzogzog]

Has anyone seen a spec sheet for this variant?  Even the Omron site dishes up only a Chinese data sheet (that doesn't look clickable)

Evidently the ROG Gladius II gaming mouse is one of the first significant usages.

[Penatr8tor]

22 minutes ago, zogzogzog said:

Has anyone seen a spec sheet for this variant?  Even the Omron site dishes up only a Chinese data sheet (that doesn't look clickable)

Evidently the ROG Gladius II gaming mouse is one of the first significant usages.

I did some searching and came up pretty much empty handed. The only mention of that switch is the ROG Gladius on Reddit that's like 2 years old. So yeah... good point on the switch. That looks like a mystery that has yet to be solved.

[mvdveer]

14 minutes ago, Penatr8tor said:

That looks like a mystery that has yet to be solved.

We got a good team assembled to solve this, by the looks of it

[Penatr8tor]

2 minutes ago, mvdveer said:

We got a good team assembled to solve this, by the looks of it

Agree... We will see and we will determine if this is something worthwhile or not. 

[zogzogzog]

I found https://www.bigmarker.com/eetech-media/OMRON-High-Performance-Switches-for-the-Esports-Gaming-Industry

It mentions the D2FP switches in particular.

 

It's a little over a year old.

Edited March 27, 2023 by zogzogzog

[mvdveer]

The D2FP switches are mentioned at the 8min mark and the discussion lasts untill the 15:30min mark.

Does not give much specs but mentions Quality, Longevity, low Friction and low power consumption. The optical sensor (led) is built into the switch itself and response time when triggered is superfast.

[zogzogzog]

Key takeaways (this presentation is not a group read of the data sheet - it's a pitch deck as you'd imagine):

• E-Sports is the target market
• They worked to replicate the "feel" of the mechanical D2FC-series switches in terms of force, actuation range, and tactile feedback - below are from the D2FC series as there was no specific D2FP information in this regard (I'd imagine it being more consistent due to it's mechanical design but there still looks like a curved spring for snap-over operation inside of the D2FP that actually actuates the shutter mechanism)
  • SPST-NO operation
  • 0.44N ~ 0.74N activation force (from the D2FC information)
  • "Movement Differential" (I'm assuming this is the slick way of saying actuating distance) is 0.12mm (max)
  • 0.5mm overtravel
• Lifetime of 70M operations (highest end D2FC switches top out at 60M for the most expensive variants)
• Faster response time than mechanical switches
  • 0.015ms versus 5ms (this is mostly do to the debouncing required for mechanical switches)
  • No signal bounce - due to using an internal optical sensor as opposed in to mechanical contacts
  • This would seem be a win for detecting a threshold for a moving object activating the switch.
• Sealed assembly for consistency
• They claim the "feel" of the D2FP switch is maintained over it's 70M operation lifetime.

Since this is an active component (it has an internal LED as the light source), it requires power - that's not a problem for our application.  It does mean it needs power supplied and then a signal connection.  It has 5 connections from what I could see - pinout was not specified.

No specs on actuation consistency but if the internal mechanisms stay aligned, it would seem that the switches could be quite consistent.  Not about consistency from switch to switch but that's not our concern since once you've dialed in your specific single switch you should be done.  Being that Omron is a Japanese component company, I'd expect the same near fanatical statistical consistency focus that their other components exhibit.  That said, printer sensing isn't the target market but it didn't seem like this wasn't a reasonable application.

 

Bottom line:  It's faster than the mechanical switches due to no contact bounce.  Consistency in actuation distance is not specified but this isn't a data sheet but a sales presentation.  At the time of the presentation, you needed to reach out to Omron for the D2FC and D2FP datasheets (my poking around at components.omron.com didn't run up the D2FP datasheets.

 

Zog, subjecting himself to sale presentations so you don't have to - about 30 minutes I'll never get back...

 

[zogzogzog]

I do find it interesting as to how to maintain a super fast response time if you're polling the switch by modulating the LED power for reduced power consumption - you're polling rate is going to directly impact responsiveness but for us, we're mains powered so we just just let the photons fly!

[zogzogzog]

There are some mechanics between the switch actuation and internal shutter.  The original Voron design provides the shutter as part of the Tap mechanism.  That would seem to be potentially more consistent.

Enough to be important - who knows right now.

[claudermilk]

21 hours ago, Buurman said:

uuuhhh...

Pre-ordered  I think very reasonable pricing, and shipping... (my wife wont agree.. but then again.. who cares   if she doenst know... muahahaha!

BLACK!!

Is there no mod @Buurman can say no to? 

OK, so the switch is like our standard D2F, just with an internal optical setup instead of the mechanical. Interesting. 

[Calvinx]

I printed my own previously shown version using a pair of cross slide bearings in ASA and have never had an issue.

But after seeing the CNC'D version I thought I could do that for much less cost.

This is what i came up with using 3mm sheet aluminium

Edited March 28, 2023 by Calvinx
spelling

[Buurman]

6 hours ago, Calvinx said:

This is what i came up with using 3mm sheet aluminium

Im all in for innovation!!! I just curious about everything and so I want to test it... 

Still realizing the rails is still way too heavy, so I love what you are thinking of.. would keep it much lighter!

[zogzogzog]

1 hour ago, Calvinx said:

This is what i came up with using 3mm sheet aluminium

 

Nice design - like the preloaded spring idea.

As far as machining that for way less than $50, I've got a couple of milling machines and a couple of lathes and probably the aluminum stock in the shop as well and I personally wouldn't even start if someone gave me $50.  Granted my machines are all old-school manual machines which makes a difference for sure - you tend to do more planning and fixturing but it's not crazy (I think of all the parts that must have been made by my machines before I even got them and I feel like such an amateur).  Could I make them if the end of the world depended on it - sure!  I've made more complicated but I was super motivated and had few alternatives.

This is a job that CNC router like the Omio, or Shapeko or maybe even some of the low end Amazon/eBay/AliExpress jobs could handle - maybe this is an excuse to buy an Omio...

 

$50 does seem really cheap to me for the whole assembly.

[Penatr8tor]

If it's just cut from sheet and there's no surface features then it's easy enough for some vendor like send-cut-send to laser or waterjet cut at a reasonable rate. While the cross-roller slides allow you to utilize the center of the tool head for some other purpose like the spring... they also create an assembly challenge. An MGN style slide the is adjusted at the factory and as long as you don't dis-assemble it, it will run true. With cross-roller slides like the ones pictured. You're relying on the manufacturing tolerances of the aluminum parts and the user to do the alignment. If all 4 rails aren't parallel and have the correct spacing... you could end up with a loose wiggly slide or an overly tight slide. I'm not saying that it can't or won't work, just that extra attention needs to be focused on making sure that the design has some kind of alignment pins or a groove that prevents the slides from moving during assembly.

[aju3169]

On 11/16/2022 at 10:21 PM, Calvinx said:

I can only but support the sentiment by @Buurman

why make life difficult ? me,myself & I would have chosen one of These, the video quite clearly shows they implemented a custom PCB which by its definition of "custom" means that the pcb (and or the sensor) is possibly not going to be 100% readily available, why not follow their supposed own ethos i was advised of earlier of using " off the shelf parts" and use an off the shelf & READILY AVAILABLE part, using a custom pcb and some obscure and difficult to obtain sensor is a MAJOR oversight.

As stated by another earlier in the thread,,,,

I have used this exact sensor in my Voron tap - just need a dremel grinder disc to trim the long side edges a little for fit (though don't cut the copper tracks!).  The two holes on either side on the sensor are a perfect fit for the tap.  Bonus - these are cheap & readily available - plus already come with a connector and cable.  You will need to reverse the logic in klipper to detect when activated and the red LED turns only when activated on probing which is nice.

[aju3169]

On 11/16/2022 at 10:21 PM, Calvinx said:

I can only but support the sentiment by @Buurman

why make life difficult ? me,myself & I would have chosen one of These, the video quite clearly shows they implemented a custom PCB which by its definition of "custom" means that the pcb (and or the sensor) is possibly not going to be 100% readily available, why not follow their supposed own ethos i was advised of earlier of using " off the shelf parts" and use an off the shelf & READILY AVAILABLE part, using a custom pcb and some obscure and difficult to obtain sensor is a MAJOR oversight.

As stated by another earlier in the thread,,,,

I have used this exact sensor in my Voron tap - just need a dremel grinder disc to trim the long side edges a little for fit (though don't cut the copper tracks!).  The two holes on either side on the sensor are a perfect fit for the tap.  Bonus - these are cheap & readily available - plus already come with a connector and cable.  You will need to reverse the logic in klipper to detect when activated and the red LED turns only when activated on probing which is nice.