Author Topic: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)  (Read 4641 times)

0 Members and 1 Guest are viewing this topic.

Online KaosJ

  • Thread Starter
  • Posts: 341
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #50 on: Wed, 27 December 2017, 14:07:33 »
Knowing the manufacturer machine's kerf would help by any chance?

Yes but in this case the machine of the producer will automatically offset the kerf, that's why he also asked which is the "good part" (aka the part i will need) and the part to throw away (e.g squares of the holes in our case). 

I still have to get some accurate measure of the holes of the samples (hopefully will be able to do it before tomorrow evening), i did some fast measurements when i got the samples and snapped some pics, but i have to do something more precise because it's easy to get wrong measurements with a calipher when you do it fast. Basically my goal is to understand if the machine actually calculates the kerf automatically (just like the producer said) or if it doesn't.   And to do that i have to see if there is a -0.1/ -0.15, and believe me if the calipher is not straight straight with the cut line, it is easy to misscalculate/measure wrongly. 
 

Will update more tomorrow on the Kerf.
« Last Edit: Wed, 27 December 2017, 14:09:23 by KaosJ »

Offline duynguyenle

  • Posts: 1311
  • Location: UK - Midlands
  • Personal text? What personal text???
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #51 on: Wed, 27 December 2017, 16:44:22 »
In my experience (using two different sheet metal fabs, UK-based), the standard practice is to give them the EXACT measurement and tell them to take into account the kerf on THEIR end, using their own adjustments. This is for several reasons:

  • Each sheet metal fab may employ multiple laser machines, each with different specs/requirements to suit different jobs based on bed size and different material requirements (pretty standard practice). By giving them the EXACT measurements that you expect the final cut to be, you allow THEM to adjust the measurements based on the particular machine they're going to use on the day
  • Giving them the EXACT measurements at input shift the responsibility to the manufacturer to make sure that the end products come out how YOU expect it. There's absolutely no use in trying to take care of the kerf adjustments yourself, even if you have a sample from them, because they may use a different machine to produce your next order, based on their availble production slots/factory planning/other logistical reasons
  • Don't forget Cherry switches themselves comes with their own tolerance stack. If you follow the official specs, and allow the sheet metal fabs to adjust the kerfs on their own terms, that means that the plate comes out correct. I do not care at all whether or not the switch makes a nice sound when you insert it into the plate, I care about whether or not the plate is made to spec. Sure your 13.85mm cutout will work with some switches (may be slightly tight), but keeping in mind that some batches of switches might be at the upper end of the tolerance band, and using an out-of-spec cutout may cause issues for the odd switches that might be perfectly conforming according to Cherry specs but does not fit into your smaller cutout footprint
« Last Edit: Wed, 27 December 2017, 16:49:04 by duynguyenle »
| QFR            | Leeku 1800    | Raptor K1      | Dolch Pac

Online KaosJ

  • Thread Starter
  • Posts: 341
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #52 on: Wed, 27 December 2017, 19:19:49 »
In my experience (using two different sheet metal fabs, UK-based), the standard practice is to give them the EXACT measurement and tell them to take into account the kerf on THEIR end, using their own adjustments. This is for several reasons:

  • Each sheet metal fab may employ multiple laser machines, each with different specs/requirements to suit different jobs based on bed size and different material requirements (pretty standard practice). By giving them the EXACT measurements that you expect the final cut to be, you allow THEM to adjust the measurements based on the particular machine they're going to use on the day
  • Giving them the EXACT measurements at input shift the responsibility to the manufacturer to make sure that the end products come out how YOU expect it. There's absolutely no use in trying to take care of the kerf adjustments yourself, even if you have a sample from them, because they may use a different machine to produce your next order, based on their availble production slots/factory planning/other logistical reasons
  • Don't forget Cherry switches themselves comes with their own tolerance stack. If you follow the official specs, and allow the sheet metal fabs to adjust the kerfs on their own terms, that means that the plate comes out correct. I do not care at all whether or not the switch makes a nice sound when you insert it into the plate, I care about whether or not the plate is made to spec. Sure your 13.85mm cutout will work with some switches (may be slightly tight), but keeping in mind that some batches of switches might be at the upper end of the tolerance band, and using an out-of-spec cutout may cause issues for the odd switches that might be perfectly conforming according to Cherry specs but does not fit into your smaller cutout footprint

That's what i did and also the producer always said me there is no kerf for them because their software will calculate it.  Aka from both the old producer and the new producer i was said to provide the original file because the Kerf was calculated by the machine.  Now hopefully tomorrow i will measure correctly the sample and will see if the software actually calculated the kerf correctly. 

About the switch clipping and making a nice sound when clipping it was just to say, it is tight, it doesn't move at all (sometimes is bad, sometimes is good, i prefer tight).  Again, the default 14mm is in the standard file i provided and will be in the every file i will do (tested or not tested), for a simple reason, 14mm always works, 14mm on the file will result in a usable plate even if there is any error or misscalculated kerf by the machine
 
What I said is that if the ACTUAL hole on the sheet (not the file) is 13.85mm, in my experience it will be good and tight, even considering any error by the switch, never meant that 13.85mm hole must be on the file, where any extreme error or misscalculated kerf might cause it to be for example 13.6mm (definetely not good). 

Hopefully tomorrow i will find out if the kerf was automatically calculated by the software, like i was said, and will provide an update 

« Last Edit: Wed, 27 December 2017, 19:25:35 by KaosJ »

Offline duynguyenle

  • Posts: 1311
  • Location: UK - Midlands
  • Personal text? What personal text???
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #53 on: Thu, 28 December 2017, 07:25:38 »
In my experience (using two different sheet metal fabs, UK-based), the standard practice is to give them the EXACT measurement and tell them to take into account the kerf on THEIR end, using their own adjustments. This is for several reasons:

  • Each sheet metal fab may employ multiple laser machines, each with different specs/requirements to suit different jobs based on bed size and different material requirements (pretty standard practice). By giving them the EXACT measurements that you expect the final cut to be, you allow THEM to adjust the measurements based on the particular machine they're going to use on the day
  • Giving them the EXACT measurements at input shift the responsibility to the manufacturer to make sure that the end products come out how YOU expect it. There's absolutely no use in trying to take care of the kerf adjustments yourself, even if you have a sample from them, because they may use a different machine to produce your next order, based on their availble production slots/factory planning/other logistical reasons
  • Don't forget Cherry switches themselves comes with their own tolerance stack. If you follow the official specs, and allow the sheet metal fabs to adjust the kerfs on their own terms, that means that the plate comes out correct. I do not care at all whether or not the switch makes a nice sound when you insert it into the plate, I care about whether or not the plate is made to spec. Sure your 13.85mm cutout will work with some switches (may be slightly tight), but keeping in mind that some batches of switches might be at the upper end of the tolerance band, and using an out-of-spec cutout may cause issues for the odd switches that might be perfectly conforming according to Cherry specs but does not fit into your smaller cutout footprint

That's what i did and also the producer always said me there is no kerf for them because their software will calculate it.  Aka from both the old producer and the new producer i was said to provide the original file because the Kerf was calculated by the machine.  Now hopefully tomorrow i will measure correctly the sample and will see if the software actually calculated the kerf correctly. 

About the switch clipping and making a nice sound when clipping it was just to say, it is tight, it doesn't move at all (sometimes is bad, sometimes is good, i prefer tight).  Again, the default 14mm is in the standard file i provided and will be in the every file i will do (tested or not tested), for a simple reason, 14mm always works, 14mm on the file will result in a usable plate even if there is any error or misscalculated kerf by the machine
 
What I said is that if the ACTUAL hole on the sheet (not the file) is 13.85mm, in my experience it will be good and tight, even considering any error by the switch, never meant that 13.85mm hole must be on the file, where any extreme error or misscalculated kerf might cause it to be for example 13.6mm (definetely not good). 

Hopefully tomorrow i will find out if the kerf was automatically calculated by the software, like i was said, and will provide an update

I see, thanks for the clarification. Let us know of any updates!
| QFR            | Leeku 1800    | Raptor K1      | Dolch Pac

Online KaosJ

  • Thread Starter
  • Posts: 341
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #54 on: Thu, 28 December 2017, 12:32:28 »
UPDATE

Alright here are some measurements, please note that i own a 15 bucks calipher so nothing extremely precise, and for this i tried to include other types of measurements as well.   

NOTE:  The bottom of the samples is raw brass so it's ugly (in the pics and irl). 

TL;DR: Measures seems to match the file. The kerf seems to be calculated automatically, otherwise the holes of the samples should have been bigger of the holes in the file. e.g. 14mm on file should have been 14.1/14.15mm on plate if the kerf wasn't calculated, but this isn't the case.  On a side note, after measuring all the holes there are some points where the lines are 0.05mm/0.1mm smaller than the file, i guess that's machine error or different kerfs while the laser is moving on different types of lines or any deviation of the brass sheet, nothing that would affect the final product or that you can see, but still that's what i calculated with my poorman calipher. 

Also i did fit different type of switches and all fitted correctly (note the base of the switch without the clip is around 13.85mm), the fit was tight but not too tight. 
Also the stabilizers holes that i did 7mm to be sure, fitted correctly with a super small gap, stabilizers are around 6.7mm and needs around 6.8mm for a supertight fit, while i have a gap here.   

Afterall, i think the cut is on point, Kerf seems to be automatically calculated. Here are some pics.

This is 4.20mm on my file





I wanted to fit the kbdfans blockers inside the plate, the blocker that was 13.9mm at the bottom fitted in every hole, the one that was 14.01/14.05mm didn't fit in every hole. I suspect the holes must be 14.05mm to fit correctly every kbdfans blocker without pushing too much (the base of the blocker seems to have a variance of max 0.15mm on different blockers). 
   




A comparison with KBD75 R5 plate, seems equal, maybe there is a very small difference but i still think kbdfans uses 14.05mm holes (at least that's what i calculated) and in fact the KBD75 plate seems to not be very tight (a bit loose if you ask me)



« Last Edit: Thu, 28 December 2017, 12:47:14 by KaosJ »

Offline Blackhawk

  • Posts: 197
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #55 on: Thu, 28 December 2017, 12:56:45 »
Pro tip: you can use the top part of the calipers to measure internal diameters

Online KaosJ

  • Thread Starter
  • Posts: 341
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #56 on: Thu, 28 December 2017, 13:06:24 »
Pro tip: you can use the top part of the calipers to measure internal diameters

I know ahahah, the back part. Even if you see few pics i spent like 2hours measuring it and the back part is what i used more (what i call back is what you call top but you clearly mean the part to measure the inside of a hole). 

The problem by measuring with the bottom part is that if you insert let's say a 20mm measure in a 19.85mm hole, you can still make it fit easily because that part will scretch the brass slightly (calipher is pointed and sharp SS). Also pretty hard to keep the calipher perfectly aligned when measuring with the bottom part of the calipher.  Even measuring the same exact point might have a 0.3mm variance when measuring with that part, at least for me with that calipher.  That's why i measured with all the parts of the calipher (except with the part to measure the deepth)   

Also the pics above are easier to see with the external part,  found kinda hard to measure with the internal hole part of the calipher and take a clear pic of the dimension. 

If it isn't clear, yeah i used that part mostly ;D
« Last Edit: Thu, 28 December 2017, 13:13:32 by KaosJ »

Offline Blackhawk

  • Posts: 197
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #57 on: Thu, 28 December 2017, 13:13:35 »
Pro tip: you can use the top part of the calipers to measure internal diameters

I know ahahah, the back part. Even if you see few pics i spent like 2hours measuring it and the back part is what i used more (what i call back is what you call top but you clearly mean the part to measure the inside of a hole). 

The problem by measuring with the bottom part is that if you insert let's say a 20mm measure in a 19.85mm hole, you can still make it fit easily because that part will scretch the brass slightly (calipher is pointed and sharp SS). Also pretty hard to keep the calipher perfectly aligned when measuring with the bottom part of the calipher.  Even measuring the same exact point might have a 0.3mm variance when measuring with that part, at least for me with that calipher.  That's why i measured with all the parts of the calipher (except with the part to measure the deepth)   



If it isn't clear, yeah i used that part mostly) ;D
https://i.imgur.com/lcx4Zov.png

Haha, good to know. It always irks me a little when people use calipers incorrectly  :D

Online KaosJ

  • Thread Starter
  • Posts: 341
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #58 on: Thu, 28 December 2017, 13:27:05 »
Pro tip: you can use the top part of the calipers to measure internal diameters

I know ahahah, the back part. Even if you see few pics i spent like 2hours measuring it and the back part is what i used more (what i call back is what you call top but you clearly mean the part to measure the inside of a hole). 

The problem by measuring with the bottom part is that if you insert let's say a 20mm measure in a 19.85mm hole, you can still make it fit easily because that part will scretch the brass slightly (calipher is pointed and sharp SS). Also pretty hard to keep the calipher perfectly aligned when measuring with the bottom part of the calipher.  Even measuring the same exact point might have a 0.3mm variance when measuring with that part, at least for me with that calipher.  That's why i measured with all the parts of the calipher (except with the part to measure the deepth)   



If it isn't clear, yeah i used that part mostly) ;D
https://i.imgur.com/lcx4Zov.png

Haha, good to know. It always irks me a little when people use calipers incorrectly  :D

Btw here is an example to make it more clear (done on the kbd75 plate), that's why i think the internal hole part of the calipher isn't accurate enough on plates, basically because a plate is too thin ( around 1.5mm) for that part of the calipher. That part of the calipher imho works good when the hole has a thickness and when the material is strong enough (e.g SS) 

See the 3 pics below. 


I believe the real hole is 14.05mm


...but 13.99mm came out from this type of measurement


... and 14.10mm came out from the same type of measurement, 0.1mm variance from the previous photo, measured in the same point, mostly due scratches of the calipher on plate or/and different angles of the calipher. 
 

TL;DR my goal was to measure if there is a 0.1mm error/kerf on the sample (that is a really small measure) and with that part of the caliper it is easy to take different measurements on the same point of a thin plate. 

 
EDIT: A last pic, this is how i think is better to use the internal part of the calipher when you have to use that part on a thin surface, and that's what i did when measuring with the internal part.   


« Last Edit: Thu, 28 December 2017, 13:40:16 by KaosJ »

Offline duynguyenle

  • Posts: 1311
  • Location: UK - Midlands
  • Personal text? What personal text???
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #59 on: Thu, 28 December 2017, 19:18:55 »
Elastic deformation of the plate when you force it with the jaws of the caliper isn't really gonna mean anything in the grand scheme of things (probably in the handful of microstrains). It is extremely unlikely that you will work the plate to its plastic deformation using any sort of reasonable hand force, and if you do, it's likely that you're stressing the calipers beyond its design limits in the first place. Use the small jaws for internal dimensions like they're intended to, otherwise you just introduce additional parallax errors when you try and line it up like you are doing with the large jaws.

That said, everything seems to be to spec. Good too hear! Would love to get this thing on the road once everything is nailed down.

As for the stabiliser cutout, 7mm is sensible. On my own drawings I tend to err on the generous side and spec 7.15mm wide holes, for personal preference, and they come out just fine on the 1800 plates that I had made in the past.
| QFR            | Leeku 1800    | Raptor K1      | Dolch Pac

Online KaosJ

  • Thread Starter
  • Posts: 341
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #60 on: Fri, 29 December 2017, 03:12:52 »

That said, everything seems to be to spec. Good too hear! Would love to get this thing on the road once everything is nailed down.

Right now i'm just hoping to fix the coat if they can or find a better clear coat service or just remove the coat. 
Yesterday i also started to build mine with one of the sample (standard ansi) and seems to work flowly



As for the stabiliser cutout, 7mm is sensible. On my own drawings I tend to err on the generous side and spec 7.15mm wide holes, for personal preference, and they come out just fine on the 1800 plates that I had made in the past.
 

Yeah here comes to people preference i guess. Not that you can see any real difference between 7 and 7.15mm but whatever, that's why we are doing a custom plate GB.  The important for me is that people will not use the standard swillkb for stabs (aka 6.8mm), it still works but i don't recommend it. 

Elastic deformation of the plate when you force it with the jaws of the caliper isn't really gonna mean anything in the grand scheme of things (probably in the handful of microstrains). It is extremely unlikely that you will work the plate to its plastic deformation using any sort of reasonable hand force, and if you do, it's likely that you're stressing the calipers beyond its design limits in the first place. Use the small jaws for internal dimensions like they're intended to, otherwise you just introduce additional parallax errors when you try and line it up like you are doing with the large jaws.

Well i did measured with the small jaws (it's what i used more as said above), but i measured also with the big part and that was imho the most accurate measurement. 
The goal here was to find out if there was a 0.05/0.1mm difference so the elastic deformation definetely matters in this, since it is easy to take different measures with the small part by just applying a different force or angle (since the small jaws also have a really small surface). On the other side aligning properly with a blocker or a stabilizer housing the big part of the calipher was easy to get better measurements. 
Honestly I'm trying to be super-accurate, as much as i can with my equipment (that is definetely not the best equipment) so for this i also included other types of measurements or comparisons with another plate, i also compared my other plates and custom plates i already had but i only included the kbdfans in the pics because it was black (aka easy to see the contrast on photos). 

Afterall the real test, literally putting different types of switches inside the holes and putting different factory stabilizers, made me satisfied with the dimensions, it worked much better than any other plate i own not too tight, not too lose, firm switches but still not hard to remove, just a nice fit like it should be  :D imho now i should focus on the coat. 
 

 

« Last Edit: Fri, 29 December 2017, 18:30:10 by KaosJ »

Online dr_derivative

  • Posts: 102
  • Location: Cork, Ireland
  • Not a real doctor
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #61 on: Fri, 29 December 2017, 17:54:39 »


Would a 2u switch/stab cut out like this be OK in a custom plate? I'm worried the 1.4mm section could be too thin and fragile to manufacture.

I'm hoping to use a cut out like this for SMK switches since they have clips at the sides rather than at the front and back like Cherry.

Online KaosJ

  • Thread Starter
  • Posts: 341
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #62 on: Fri, 29 December 2017, 18:29:09 »
Show Image


Would a 2u switch/stab cut out like this be OK in a custom plate? I'm worried the 1.4mm section could be too thin and fragile to manufacture.

I'm hoping to use a cut out like this for SMK switches since they have clips at the sides rather than at the front and back like Cherry.

Legit question, this is one of the things that i can't say for sure now but will tell more during the GB, this is something that i have to ask to the producer (along with many other questions).     

AFAIK from previous experiences with laser cut services, it is recommended to not let things smaller than the thickness of the material on the file, because the laser might burn that part, but in your case we are really close to the material thickness (1.4mm vs 1.5mm) so i think it will work without problems. For sure don't do a 0.3mm feature, but this seems ok to me (still I will tell you something more precise when I will have a better answer) 
« Last Edit: Fri, 29 December 2017, 18:32:10 by KaosJ »

Online dr_derivative

  • Posts: 102
  • Location: Cork, Ireland
  • Not a real doctor
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #63 on: Fri, 29 December 2017, 18:41:43 »
Show Image


Would a 2u switch/stab cut out like this be OK in a custom plate? I'm worried the 1.4mm section could be too thin and fragile to manufacture.

I'm hoping to use a cut out like this for SMK switches since they have clips at the sides rather than at the front and back like Cherry.

Legit question, this is one of the things that i can't say for sure now but will tell more during the GB, this is something that i have to ask to the producer (along with many other questions).     

AFAIK from previous experiences with laser cut services, it is recommended to not let things smaller than the thickness of the material on the file, because the laser might burn that part, but in your case we are really close to the material thickness (1.4mm vs 1.5mm) so i think it will work without problems. For sure don't do a 0.3mm feature, but this seems ok to me (still I will tell you something more precise when I will have a better answer)

Thank you  :thumb:

Offline kiettv12

  • Posts: 29
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #64 on: Sat, 30 December 2017, 15:41:31 »
will you ship to asia ? south east asia ?

Online KaosJ

  • Thread Starter
  • Posts: 341
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #65 on: Sat, 30 December 2017, 16:52:35 »
will you ship to asia ? south east asia ?

Yes, and shipping should not be exagerated, tracked ensured should be around 12 euro with the bulk shipping contract that i got. 

Eventually can set up a proxy but i don't think it will be worth toward Asia, maybe a US proxy would be cool if anyone wants to jump in! 
« Last Edit: Sat, 30 December 2017, 16:54:09 by KaosJ »

Online KaosJ

  • Thread Starter
  • Posts: 341
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #66 on: Mon, 01 January 2018, 09:24:55 »
Here is the built one with the standard ansi sample. Happy new year everyone  :D 




Online TalkingTree

  • Posts: 2020
  • Location: Italy (142)
    • My projects
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #67 on: Mon, 01 January 2018, 11:38:18 »
Neat. I'm really looking forward to this GB.

Online TalkingTree

  • Posts: 2020
  • Location: Italy (142)
    • My projects
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #68 on: Tue, 23 January 2018, 07:12:06 »
Other than asking for an update, I'd like to know if it's also possible to have some engravings.

Online KaosJ

  • Thread Starter
  • Posts: 341
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #69 on: Fri, 26 January 2018, 09:37:05 »
Other than asking for an update, I'd like to know if it's also possible to have some engravings.

Will have an update on monday I guess, including the engraving question, not totally sure if it is possible but i ask. 

Online TalkingTree

  • Posts: 2020
  • Location: Italy (142)
    • My projects
Re: [IC] Brass plates: Custom and 60% ANSI/ISO (Soon)
« Reply #70 on: Fri, 26 January 2018, 09:42:00 »
not totally sure if it is possible but i ask.
Much obliged.