The Razer

Design Concepts
1

Hi Everyone,

This is my first post so a big hello goes out to the MIDIBox community! :slight_smile: I have been following the MIDIBox community for a couple of years now because I am a big SID music fan and I love seeing the MIDIBox SID projects that people have put together, but now the time has come for me to get involved.

For a long time now I’ve dreamed about the perfect MIDI controller. I’ve always found traditional keyboards to be very limited because of their percussive nature, I’ve always yearned for something with more expressive control and more stage presence. After recently seeing the “Stribe” project and discovering Softpots I think I may finally now be able to realise my dream…

Introducing the Razer.

Obviously it is designed to be held like a guitar.

The most important feature of the Razer is the two touch strips on the “neck” (or should I call it the “barrel” he hee :wink:). Each of these strips is a 2-axis analogue controller responding to both position along it’s length and pressure.

A bullet list of features goes like this:

* 2x 500mm touch strip with pressure sensitivity

* 24 key velocity sensitive keyboard

* 2 axis expression bar

* 9 button pads

* 6 buttons for transport and LCD nav control

* 2 rotary encoders

* 2x16 LED Display

* USB MIDI interface and power

* MIDI In and Out jacks

The Razer will be built using the new STM32 Core.

Note that this is an early design sketch and that I haven’t really thought through the exact positions and number of the buttons and knobs, I just placed a bunch around the keyboard because I knew I would need some.

At the very least it will have the following play modes:

  1. Pitch and aftertouch on touch strip. Note trigger, velocity and modulation on keyboard

  2. Pitch, note trigger and velocity on keyboard. Pitch-bend, modulation, aftertouch and CC on touch strips

  3. Play a different sound on each touch strip (hammer on style) and a 3rd sound on the keyboard

I’m sure I will add more play modes as time goes on.

So this is a pretty amibtious project, especially for a MIDIBox newbie. I realise I will have to program my own MIOS app but I think I am up for the challenge. I was a web application programmer for 8 years in my previous career so I have extensive experience with nearly all of the web related programming languages. I have only dabbled a little in C for personal amusement, but I read through all of the MIOS32 tutorials and found them very easy to understand. I also programmed some assembler for the Amiga back in the day.

I also know some basic electronics, enough to keep myself out of trouble building and flying electric powered R/C Aircraft. The MIDIBox hardware platform takes most of the hard work out of this aspect and I’m sure the community can help me get past any other problems I might have.

Probably the biggest obstacle for me in this project is going to be spare time to work on it. I have a wife, toddler and 4 week old baby and run my own small business to keep the roof over our heads. So don’t expect constant updates. If I have a prototype with all of the above features working by middle of 2010 I will be really happy!

So that’s my intro, I hope you like the Razer, I am really jazzed about it in case you can’t tell. :smiley:

Current status of the project is that I have just placed an order for MB kits from Smash TV and I’ve ordered the touch strips. I live in Australia (just north of Sydney), so if these arrive before January I’ll be pretty impressed.

Cheers everyone!

Sam

2

my concern is the casework. this is something that would need to be addressed simply because tooling for that would be quite a cost. the case really needs to be modulr in the sense that its going to be easy to design and lay out using say sheet steel or aluminium, when the casework is designed all components are usually built using the sheet steel layout as then all bends are then flattened out correctly prior to laser cutting the material and then placed into a cnc folding machine to make all the necessary bends.

if your looking to get it made out of plastic then it really becomes a cost, your looking at mold blanks male and female to be mahined, the initail blocks, tooling for injection moulding etc.. it goes on.

if you need help with a case then let me know as its what i do.

apart from that everything else is feasable

3

Wow, this one looks great. What material do you want to use for that case?

4

My plan for the case is that I was going to build it out of wood.

I have absolutely zero knowledge or experience with any sort of plastic or metal fabrication especially if it goes anywhere near a CNC machine. Even if I knew someone who could do it I know nothing about working with those sorts of materials and would probably waste loads of money on badly engineered designs.

At least with wood I can buy it at the local timber yard, work on it at home with tools from the local hardware store, and mistakes are comparatively cheap.

My initial thoughts on case design would be that I would start with a sheet cut out in the shape of the body to form a “backplate”. Mount all the parts to that and then build a cover that drops on over the top with holes for all the buttons to poke through, etc. The neck would be a seperate piece that bolts on to the body section.

In the prototype stage I would just have all the parts mounted on the backplate and leave it naked.

5

I bet Wilba could ponoko the hell out of this design.

6

hey if you pull this off, it’ll be a one of a kind midibox (to my knowledge). the new akai lpk25 is a fairly inexpensive midi controller that has a compact design, so if you ripped out the keys and guts you could probably just route it’s midi straight into the core.

ultra

7

The LPK25 is not a bad idea, but I don’t think I would like the mini-sized keys.

I actually have an old MIDI controller keyboard here that I am intending to butcher to get my 2 octave keyboard. Failing that I was going to buy an M-Audio Keyrig 25 or Oxygen 25 and pinch a keyboard out of that.

With the Oxygen 25 I’d get some nice knobs and wheels to use too.

8

I’m dunno why, but there are many Aussies around here.

I can recomend a good cheap laser cutter in Syndey that I use.

http://www.lcne.com.au/

Regards

Mike (Sydney)

PS, what about implementing wireless MIDI & battery power? (M-Audio MidAir)

You don’t want a controller like that on stage, with a big umbilical cord hanging off it.

9

Thanks for the recommendation. You guys are really keen on this CNC stuff aren’t you!

Unless the wood body is a dismal failure AND someone gives me some fantastic reasons to go with a machined design I just can’t see myself going down that road.

I’ve built R/C aircraft before so I have at least a little skill with wood and making strong, light structures. Not saying I’ll build this one out of balsa, but just saying that working with wood seems quite feasible and is within my skill set.

10

Regarding the Mid-Air, yes i had looked at that and considered it, you are right about not wanting a cord hanging off just begging to be stepped on and yanked out. The only negatives would be cost and finding somewhere to shoe horn it into the case.

A big plus though is that designing for battery power could also solve another problem.

One concern I have with my design is noise on the analogue channels causing jitter with the touch strips, this is even more of a concern since I am using the Core32 which has only 3V analogue ports.

Using battery power, such as a 2 Cell Li-Po battey with linear voltage regulator, would eliminate any interference that might enter the system via a wired power supply.

EDIT: Just looked at the schematic for the Core32 and realised it has the voltage regulator built-in, so I can hook a 2 Cell LiPo straight onto J1!

11

The touch strips arrived today! I got 2x 500mm softpots and 2x 24" FSR (force sensing resistor) strips. My plan is to sandwich the two together so I have an FSR underneath each softpot, this how I plan to have both position and force sensitive touch strips.

I got them from Trossen Robotics

500mm Softpot

24" FSR Strip

Smash TV is waiting for backordered parts atm, and I am at the end of the list. Since we are about to be hit by the silly season I’m guessing my MIDIBox parts won’t come till some time in January.

12

I love the machine-gun styling; the only thing I could suggest would be to consider tilting the keyboard assembly a little to the right (from the player’s side) to put the keys at a more comfortable angle for your right hand.

13

Thanks for the complement Dr Bunsen. :slight_smile: With the angle on the keyboard, I was thinking that, when held like a guitar, the whole instrument would be already angled up at about 30 degrees or so from horizontal, this should hopefully put the keyboard at an angle that is just right for the right hand. We’ll see when I put together the prototype, but it’s certainly something I will look out for, thanks for the heads up.

This morning I finally had a chance to take a look at my Softpots and FSRs. I plugged them into a breadboard and hooked up my multimeter just to see how they respond to touch. I was very happy to find that both types of strips are sensitive to a fairly light touch. I was a bit concerned that they might require more force to activate than I would like, which would make it harder to play with finesse and cause fatigue, but I reckon they are just right. Very happy about this.

I also checked out that Ponoko site for the first time because of frailn’s comment. Interesting business idea! If someone could convince me that they have materials that would be suitably strong and durable AND hold my hand through the process of designing and submitting the plans to Ponoko, then I might consider it. As it is though, it’s a bit too foreign of a concept for me to get my head around at the moment.

14

Over the last week I’ve been working on the case design for my prototype. I am calling it the Razer 6mm, because I decided to make it entirely out of 6mm ply since I already have a sheet of it. I tried to keep things simple as this is only a prototype case.

It has become a bit fatter in all directions as I had to allow extra space all over the place so that everything fits in together.

I have made a series of screen shots to show how I am planning to build it. There is one piece of the case that is not 6mm ply and that is the block that joins the neck to the baseplate. I will use wood screws to screw the neck and the baseplate onto that block.

As I am typing this I have just realised that I haven’t yet modelled the screws and posts that will allow the top plate to be removed. So if your wondering how I will get internal access, well the whole top plate will come off. I will have about a dozen threaded-tube riser posts coming up from the base and the top plate will be attached by screwing into those posts.

The Base

The side walls, note that the end caps on the body protusions are perferated metal plates. On the top right you can also see the neck attachment block.

The top plate and the bottom layer of the neck.

The raised area behind the keyboard, the top layer on the neck, and the handle. The case is complete at this stage.

Here are all the electronic bits installed.

This shot is looking from the other side with the keyboard removed. You can see the LCD on the right. The white rectangle inside is the Core32 module.

This is another internal shot near the neck. You can see how the handle (square aluminium tube) will be screwed to the base. It also gives you a close-up of the LED lit momentary switches I plan to use.

15

Here’s a quick TODO list of case design issues/concerns that I haven’t properly dealt with yet.

* Weight. I predict this 6mm ply version will be a bit of a boat anchor. Not sure what material I am going to switch to for prototype #2 if that turns out to be the case. I’d like to be able to do a fibreglass shell for it but I know diddly squat about working with fibreglass.

* Keybed attachement. Until I have the keybed in my hands I won’t know the exact number and position of the attachment points. I was originally thinking about harvesting 24 keys from an old 4 octave controller keyboard that I have here, and for the sake of cost I will probably still do that for the prototype. However I think in the long run it would better for the sake of repeatability if I harvested a whole keyboard from a cheap 25 key controller such as the M-Audio Oxygen 25.

* Position of the Core32 module. The headline feature of the Razer is the analogue strips on the neck, I need these to be as smooth and accurate as possible to maximise their playability. Due to the STM32 having 3.3V analogue inputs, analogue jitter is a real concern. The typical advice to minimise jitter is to make the cable runs as short as possible between the analogue devices and the core. So far, in my design, they are about as far apart as they can get. I am intending to use battery power for the razer so there shouldn’t be any PSU switching noise to deal with which is a plus. I will just have to experiment here and there is not a lot I can do until I get the Core32 built. Worst case scenario here is that I will just have to make the case thick enough that the Core32 module can sit underneath the keyboard right next to the neck attachment block.

16

well i ran it through solidworks and even with the tightest margins on a cnc bender you would have to have breaks in the fold tabs. however there is another option which is the have it all cut out of acrylic, a nice gloss red acrylic and then bond the sides to the base to give you the upright side panel sections, similar to the ponoko way but with fewer tabs. and the edges of the sides can be chamfered so they too can be bonded together, thereby making a very strong lower case, the main face can be an aluminium or acrylic face. if you go aluminium you can have some clinch/pem studs inserted and then the panel can be primed and painted or powdercoated.

if i get time this week i will do a little set of renders to explain the sectional bonding process with a few little panels shown.

in the meantime heres a quick render of the case where i tried a marginal cnc bend.

17

Crikey! That render looks like a photo! Very nice!

I know nothing at all about CNC or manufacturing in general, so I can barely follow what you are saying, but I am very grateful that you want to help as you seem to know A LOT more than I do about this stuff.

You have probably guessed that I use Google Sketchup for my designs. It’s the only 3D program that I can get my head (and wallet) around. Since you are offering to help, is my Sketchup file any good to you? Happy to post it if you want it.

Acrylic and aluminium sound like great options, if only I had any idea at all of where to even buy them! :frowning: I’ve seen perspex at Bunnings (local hardware warehouse store) but it’s like $60 for a 3mm sheet that is about half a square meter which seems like a horrible rip off.

Also, I’ve looked extensively for 1mm or 2mm sheet aluminium in the past for a previous project and the only place I could find that would sell direct to the public only sold sheets that were 3m x 1.2m which was far more than I wanted and was too big to fit in my car anyway! Maybe I’m going about it all wrong, but I just can’t seem to find places that sell this stuff in hobbyist sizes and quantities.

18

well the acrylic case can be cnc cut and then bonded, the face plate can also be cnc cut, if you go with the acrylic face then its all done at the same time using the same comapny, i can find a company near to you that can take solidworks/dxf/dxg or iges files etc.

as for aluminium you can use any cnc cutting service as long as the files are supplied and correct they will just dump them into the machines and cut it. i can also locate all the pem holes etc.

now to do this case properly is going to need precise measurements and tech drawings of the components/parts being used, by this i mean the data sheets for the parts.

the case is buildable, its just the acrylic way seems the best option for you to get the lightness and glossyness you want. once bonded the things is realy strong, just dont drop it!

19

OK, that all sounds pretty cool. BTW what is a pem hole? Do you mean the holes for the switches and pots to poke through?

For now I think the first step is to focus on getting the electronics and firmware working properly so I know I am not wasting your time on the case work. I’ll still just build my cheap wooden prototype to hold everything together while I tinker.

Really the only reason I am doing so much work on the case at the moment, instead of the electronics and software, is because I am still a long way down SmashTV’s backorder list for my MIDIBox kits.

20

Looking into wireless MIDI solutions, I’ve been rather dissapointed to see that every product I’ve found so far is only a 1 way connection. What’s more, they are all designed to carry MIDI FROM the battery powered remoted unit TO the base station, so you can’t even use 2 sets in tandem to get a 2 way connection because the receiver units are not battery powered.

What’s more, the Razer is a MIDI device that is going to emit a lot of CC data, probably on several channels at once, so it would really be preferrable to use the USB MIDI connection of the Core32. For that reason I am probably going to use one of these for my wireless solution.

USB 2.0 Wireless Extender

It’s not cheap but it provides a high speed USB 2.0 connection to about 30m using an 802.11g connection.

Now I can have high data rate MIDI and also get MIDI feedback from my DAW.

EDIT: Yes I know the WiRanger remote unit is not battery powered, but it’s external PSU is 5V DC so providing power won’t be a problem.

BTW here is the manufacturers product page.