What's new

hatmoose

Champion
Joined
Mar 25, 2021
Messages
1,073
Reaction score
2,240
Location
New Zealand
This thread is to share ideas and learnings from building other peoples projects.
I've had a good look and I cant see another like it, mods, if there is please let me know and I'll join onto that instead

The idea came about when I wanted a really good Supergun. So decided to make my own based on FrankFJSs awesome Minigun Supergun from this very forum
Minigun Supergun 2.5
Then I saw the USB2DB15 and Saturn 2 neogeo projects
USB2DB15
Saturn 2 Neo Geo
Then i saw the CPS2digiAV project
CPS2digiAV
then I saw a load of others. I realised it was time to stop imagining how cool they could be and start making them

There is a tonne of domain specific "stuff" in each of these projects, programing arduino chips, converting .brd to gerber so PCBway will etch them, correctly crimping XH2.54 pins. A lot of this stuff is either assumed knowledge that many people dont have, or really hard to figure out, or so cross-domain that very few people will have all the skills.

The goal of this thread is to share ideas, tips and techniques, raise the quality of the stuff we build. And most importantly to help show the project owners/maintainers/contributors that their work is valued and appreciated

So I'll be starting with the Minigun Supergun advanced 2.5 (9 pin). I'd love to follow your builds, learn from your experences and share your ides too
 
Today I started work on Frank_FJS Minigun Supergun Advanced 2.5 (9 pin). I bought the PCB's from another forum member (thankyou @evilsim).

I had never done any SMD work before so I bought one of these hot-air guns
KSGER official store on Aliexpress
some solder paste from a local supplier
and a load of SMD practise boards
random SMD practise boards on Aliexpress

The hot air gun is a work of art, I love it. I learned that Solder paste needs to be refrigerated and has a short shelf life, and that SMD work is a tonne of fun with the right gear
IMG_5205.jpg
 
Still working on Frank_FJS Minigun Supergun Advanced 2.5 (9 pin)

An hour of practise with the SMD rework gun took me from a mangled mess like the ones in the middle and the left, to a slighly less mangled mess :) With a bit more practise I can now make a reasonable job of 0603 and 0402 sizes. 0402 is actually too small for me to really see. Luckily all of the SMD devices in this project are much larger 1206 - except for the THS7374 - more on that in a bit

practise 2.jpg


What helped me was
1) Using the right amount of solder paste
2) Using the right airflow so I didnt blow the component off the board
2) Using the right temp to melt the paste while avoiding melting the component, many SMD LED's were harmed in the making of this picture

The "hard" bit in the Minigun Advanced is the THS7374 chip - those legs and pads are absolutely tiny.

This was the right amount of paste for me to not have any bridges
paste.jpg


That THS7374 is really, really tiny - but wth the practise and the hot air it was much easier than expected. I used the "low" setting on the hot air at 350 degrees. I also used a really narrow nozzle on the gun so I could direct the air more towards the legs.
After that I didnt melt any more LEDs. I dont know if I cooked the THS7374 as I have no means of testing it until the project is complete, but all of the joins look nice under the microscope

The tweezers in the pic feel sharp enough to draw blood, but at this level of magnification you can see they are actually really blunt. That chip is tiny

out.jpg
in.jpg


So feeling pretty positive now that the hard bit is done :)
 
Still working on Frank_FJS Minigun Supergun Advanced 2.5 (9 pin)

I decided to assemble all of the SMD components (including the LED's) with hot air. In order to get the solder paste to melt "shiny" I had to turn the temp up really high 390 - Im guessing the calibration is off

The narrow (5mm) nozzle is great for SMD chips, but rotten for caps and resistors. The ideal with caps and resistors is that the paste melts on both sides at once and the component gets pulled nicely into place with surgface tension. With the small (5mm) nozzle I had to melt each side indivdually then line them up with tweezers. With the larger nozzle (8mm) it worked lke magic.
IMG_5254.jpg


This meant I was able to do all the SMD on all three boards quite quickly.
Luckily for me I had not ordered all of the through hole components. The reason that was lucky is that I learned that i should attach the JAMMA edge connector BEFORE the through-hole components.
I needed quite a big soldering iron tip to do the JAMMA edge, and the clearance meant that i melted the JST socket - the damage is only cosmetic but I'll probably replace it when I order the rest of the throguh hole components anyway.
IMG_5257.jpg


So lesson learned for all future boards
  • IC first with the 5mm nozzle (because its hard)
  • SMD next with the 8mm nozzle (because hot air will melt anything plastic)
  • JAMMA edge next (because there is not enough clearance to solder it later)
  • Through hole components last (lots of these are plastic and will be damaged otherwise)
So here is one completed Minigun Supergun 2.5 Advanced (9 pin) and two more with all the SMD and Jamma edge done, waiting for my next Digikey order
IMG_5260.jpg


Tomorrow I'll start crimping pins for the power and kick harness. This is a mix of Hirose, JST XH2.54 and big molex power pins
 
Still working on Frank_FJS Minigun Supergun Advanced 2.5 (9 pin)

Today was "crimping stuff up" day.

I needed to make a CPS2 kick harness, and a 6-pin power cable

The CPS2 kick harness is HAS pinout on the Minigun end and standard pinout on the board end
the Power cable is custom pinout on the Minigun end and whatever you want on the power supply end (in my case I'll be soldering it directly onto a Pico-su

I used the minigun BOM from this post https://www.arcade-projects.com/threads/minigun-supergun-kick-harness-help.16300/
and the CPS2 parts BOM from this post https://wiki.arcadeotaku.com/w/HRS_DF1B

Power plug minigun end = Molex 6 rectangualr connectors 4.2mm
Power housing 1 x Molex 39-01-2060 WM3702-ND CONN RECEPT 6POS DUAL
Power pins 6 x Molex 39-00-0038 WM2501CT-ND CONN SOCKET 18-24AWG CRIMP TIN

Power plug other end = soldered directly to power supply in my case
No parts for this

Kick harness CPS2 end = Hirose 34 pin 2.5mm
Kick CPS2 end 1 x DF1B-34DS-2.5RC 2.5MM DOUBLE ROW SKT 34P HOUSING
Kick CPS2 end 6 x DF1B-2022SC 20-22AWG SKT CONTACT TIN PLATING

Kick harness minigun end = JST HX2.54
Kick supergun end 1 JST XHP-6 455-2218-ND CONN HOUSING 2.5MM XH 6POS
Kick supergun end 6 JST SXH-001T-P0.6 455-1135-1-ND CONN SOCKET 22-28AWG CRIMP TIN

And I had this crimping tool from Aliexpress
https://www.aliexpress.com/item/4000038571208.htm

The Hirose pins were an absolute dream to crimp onto 26AWG silicone wire so the Kick CPS2 end was done in no time

Unfortunatley that's where it started to go wrong

The JST pins on the kick supergun end are JST XH2.54 and my crimping tool could not do them - the criming tool description says "XH" but it was unable to crimp genuine JST pins - the jaws were machined 1/2mm too wide- cheap stuff on Aliexpress is cheap for a reason
https://www.aliexpress.com/item/4000038571208.htm
I must have destroyed 30 pins before I gave up and crimped them by hand with pliers.
ruined pins.jpg


So I was left with this kind of crappy looking HAS/minigun to CPS2 kick harness - it's good enough for testing
good enough for testing.jpg


I'll re-make the minigun end when these specialist HX2.54 crimpers arrive from AliExpress, and I'll make the next one super short for CPS2
https://www.aliexpress.com/item/4000412502024.html
When I'm done it will look like this
IMG_5294.jpg


I had no trouble crimping the big Molex power pins, but I realised that all the wire sizes I had (22AWS and 26AWG) were probably a bit fine for power, so i've ordered some 20AWG and 18AWG and will make my power connectors when they arrive - until then I'll use these ones that were spare when I replaced my PC power supply
completed power .jpg


So all in all crimping day was a bit of a failure
the XH2.54 thing caught me out so I'm waiting on tools
the wire gage thing caught me out so I'm waiting on wire

Good enough for testing - but I'll need to do a lot of rework when the parts and tools arive

Tomorrow I'll 3D print some cases
 

Attachments

  • IMG_5292.jpg
    IMG_5292.jpg
    209 KB · Views: 120
Still working on Frank_FJS Minigun Supergun Advanced 2.5 (9 pin)

Today was a pretty lazy day printing 3D printing Cases - I laid them all out so they would print in one job which will take 14 hours at good quality

I have a Ender Creality 3 pro which is about the cheapest 3D printer that isnt complete rubbish - it prints PLA and PETG well in stock form, everything else including ABS is pretty marginal.

Anyway, for cases I'll be trying these two
this one https://www.thingiverse.com/thing:3807401
and this one https://www.thingiverse.com/thing:4629721

The notes for both say that they are updated for the advanced 2.5 (9-pin). But others on this board have said that they dont line up quite right - I'll print both and report back
Screen Shot 2021-05-01 at 7.35.53 PM.png
Screen Shot 2021-05-01 at 7.37.14 PM.png

I'll update when the print is done and we'll see how they fit the 2.5
 

Attachments

  • Screen Shot 2021-05-01 at 7.35.53 PM.png
    Screen Shot 2021-05-01 at 7.35.53 PM.png
    227 KB · Views: 94
3D printing day met with good sucess, needs some minor tweaks.

Printed these with generic PLA on a Creality Ender 3 Pro
I flipped the models to lie flat so it didnt need supports, .2mm layer height, 100% infill

Print 1 - the pacifier by Printmaker
As others on this board have noted, V3 of the Pacifier project here https://www.thingiverse.com/thing:3807401 didnt fit some of the the new component placement in the advanced 2.5 9-pin board.

Top half - Everything fits perfectly except
  • LPF hole needs to move left to align with LPF switch
  • Needs hole for the sync header
  • Service and Test holes could maybe move right and up a tiny bit (OCD is playing up)
  • R,G, B holes could maybe move right a tiny bit (OCD is playing up)
  • I was sad that the voltmeter mouting posts in the render didnt make it into the .stl - that would have been sweet
IMG_5427.jpgIMG_5426.jpgScreen Shot 2021-05-03 at 2.10.49 PM.png

Bottom half - everything fits perfectly except
  • Hole for audio plug needs to move a bit closer to the 9-pin hole, I clipped this one with side cutters to make it fit
IMG_5425 (1).jpg

This remix says it has adjusted the hole placement (no volt meter mounting posts). I will print that and see what happens. If it comes out good I'll leave it there, otherwise it's time to learn how to edit .stl files :)
https://www.thingiverse.com/thing:4822249

Print 2 - SuperGun MiniGun v2.5 Case for Black Jamma Connector by CrazyDart
Everything here fit really well. My printer and slicer printed some of the support posts too fat, but when I filed them back to tolerance it was very lovely.
IMG_5419.jpg

I was worried that the edge of the jamma connector is inset about 2.5mm from the front of the case, but the jamma connector sits proud by 3mm, so this is a feature to protect the edge, not a bug
IMG_5420.jpgIMG_5421.jpgIMG_5423.jpg
No need for a second round with this one - perfect on the first go. If I was making a bunch I would tweak my slicer settings to get that post to print perfectly, but it was 3 minutes work with a file...

So back onto the 3D printer to see how we go with that remixed pacifier top - 4.5 hours...
 

Attachments

  • IMG_5425.jpg
    IMG_5425.jpg
    132.3 KB · Views: 106
Still working on Frank_FJS Minigun Supergun Advanced 2.5 (9 pin)

Ooooh boy, those 4.5 hours turned out to be a lot longer than I expected :) This was more like 20 hours work

My OCD was acting up and there was a big gap between "good enough" and "as good as I wanted it to be". Took me several days to learn enought 3D modeling to make it as good as i wanted it to be.

In order to close that gap I had to teach myself 3D modeling with absolutely no prior experence. Then design, tweak, experiment, print and prototype each part until I got it perfect. I ended up using Sketchup Make.

Top Part - prototypes from left to right. v0.9 pictured on right, v1 is on the printer now
IMG_5433.jpg


Changes to the top part
Moved all of the holes around to align properly with the components on the 2.5 advanced mini-din9 board
Crafted some push buttons for the service and test switches - it would have been possible to acheive the same effect with different switches, but I didnt want to change the BOM
Adjusted the trimpot holes to accomidate a possible variation to the BOM that Frank talked about a while back
https://www.digikey.co.nz/product-detail/en/bourns-inc/3362P-1-102TLF/3362P-1-102TLF-ND/2536611
Crafted a light chanel post for the voltmeter to carry the light to the surface of the case - I printed this in clear to carry the light
Changes the location and added mounting posts for the voltmeter
adjusted the labels to my liking - might not be to everyones taste, looks a little like a geocities website from 1997

Bottom part - prototypes from left to right v1.0 is complete on the right, but as you can see the edge printed wonky so I need to print another one
IMG_5432.jpg


Changes to the bottom part
Changed the writing to reflect that this is v2.5 which is the final version of the minigun project according to frank
Adjusted the location of the Audio-out hole so that it aligns correctly with the audio jack on 2.5 board
added labels to clarify that this supergun outputs RGBS (cant really see it in these pics, its on the side)
added labels to clarify what pin is what on the power connector (cant really see it in these pics, its on the side)
adjusted the nut-socket to accommodate a standard M3 nut (this reduces the number of line items in the BOM by one, M2.5 nut no longer required, all nuts are now standard M3)

The (almost) finished article
V1.0 is printing now to fix that volt meter alignment, add some more accurate labels to the "Sync" jumper hole, and extend the length of the "service" and "test" buttons so they sit proud of the case by 1mm
IMG_5435.jpgIMG_5434.jpgIMG_5436.jpgIMG_5437.jpg

So the BOM for the new and improved(?) case is
2 x M3 20mm cup head bolts
2 x M3 14mm flat head bolts
4 x M3 Nuts
2 x M2 6mm flat head bolts

I'll post some final pics and the STL's when it is actually perfect and when I can get a perfectly printed top and bottom to assemble - hopefuly in the next day or so of my printer is feeling kind
 
Still working on Frank_FJS Minigun Supergun Advanced 2.5 (9 pin)

Nice easy one today - finishing up the kick harness

I was having terrible trouble crimping the X2.54 pins for the kick harness, in the end I gave up and used pliers which was very fiddly and gave a poor resul . My X2.54 crimping tool arrived from china today - does an absolutely perfect job.

The tool itself is amazing, crimping these pins has gone from being a pain to a pleasure.
The fake pins/connectors/sockets that came with it are worse that garbage - I'll never use them for anything except practise.
To give you an idea genuine JST pins cost roughly 15-25 cents each - that whole box of pins, housings and sockets was $2 of aliexpress

IMG_5440.jpg

But the tool is great - The crimps are perfect every time, it's single action so onlt a few seconds each. I'll probably do another post later comparing the the quality of genuine JST pins/housing to the chinese stuff, and the superb quality of a connection crimped with this compared to my poor efforts with pliers :)

IMG_5444.jpgIMG_5443.jpg

But anyway - Kick harnesses now done at top quality with top quality parts - sweet!

Tomorrow more 3D printing of that case...
 

Attachments

  • IMG_5445.jpg
    IMG_5445.jpg
    275.2 KB · Views: 118
Still working on Frank_FJS Minigun Supergun Advanced 2.5 (9 pin)

I'm finally "done" with the case for this - it's very, very vaguely based on the awesome Pacifier case here https://www.thingiverse.com/thing:3807401 with a tonne of changes to satisfy my OCD. About a million years ago I worked for Sun Microsystems (remember them?) doing human factors engineering. A lot of the changes are to address the human factors things that were bugging me.

So here is the final result.
IMG_5493.jpgIMG_5496.jpg

I'm really pleased with the way the buttons and light post came out, the way the voltmeter is perfectly flush with the top of the case, and a bunch of other minor tweaks.
IMG_5498.jpgIMG_5491.jpgIMG_5494.jpg

I'd never used any 3D moderling software before. So this was an awesome learning experence. Behold the massive pile of prototypes and experiments to get to this stage.
IMG_5499.jpg

At this point I'll consider the case "done". I might do one final revision for my own personal satisfaction with extra-long trimpots - but from a usability standpoint this is at an acceptable level of quality and functionality.

Tomorrow I start work on testing and bringing it all together
 

Attachments

  • IMG_5492.jpg
    IMG_5492.jpg
    168.8 KB · Views: 114
  • IMG_5495.jpg
    IMG_5495.jpg
    208.5 KB · Views: 108
  • IMG_5497.jpg
    IMG_5497.jpg
    182.1 KB · Views: 105
I'm not a fan of using a hot air station to solder components.
I rather use a soldering iron with a bevel tip and some flux.
 
Still working on Frank_FJS Minigun Supergun Advanced 2.5 (9 pin)

I have some NeoGeo CD controllers coming in the mail, but in the meantime I have created this this monstrosity
It started life as an Icade Core - which is utterly useless for anyhing except ipad games by the way. Now completley stripped and refitted with Sanwa parts with DB15 plug :)
The Stick is an Sanwa JLF which is susposed to be the best a man can get - But I am not really loving it. I much prefer the Semitsu LS38.
I'll probably swap the spring for a stiffer 2lb one, get a shorter 60mm shaft, and replace the square restrictor with an octagon one - at which point it will basicly BE an LS38...
IMG_5500.jpg

Nearly ready - just need some power and waiting on my mini-din9 cables to arrive. The MVS is an Aliexpress special, the cart is a boot, and the TV was the cheapest on the market back in the day - I'll save my good gear until I'm done testing.
IMG_5501.jpg

I'm not a fan of using a hot air station to solder components.
I rather use a soldering iron with a bevel tip and some flux.
Totally agree - the leg-pitch on that 7374 was beyond my abilities to do with the iron. But who knows one day with enough practise :) I'm terrified that I'm going to cook one of them with the hot air in the meantime - havent lost any yet. But it will happen sooner or later I'm sure.
I do have some T12 BC1 BC2 and BCF2 tips coming, so the SMD situation should improve soon
 
Don't laugh, but I'm a little proud of this hack job. My handiwork is a sloppy mess but this gbscontrol works, clock generator and all. Not bad for a cheaper DIY upscaler/downscaler. I definitely could have set this up much neater, used different wire, and mostly just improved my soldering skills.
 

Attachments

  • IMG_3302.JPG
    IMG_3302.JPG
    207.7 KB · Views: 121
Don't laugh, but I'm a little proud of this hack job. My handiwork is a sloppy mess but this gbscontrol works, clock generator and all. Not bad for a cheaper DIY upscaler/downscaler. I definitely could have set this up much neater, used different wire, and mostly just improved my soldering skills.
Nice! I had not actually heard of the GBS control until you posted this - looks like an awesome solution, especially at the price!
 
Still working on Frank_FJS Minigun Supergun Advanced 2.5 (9 pin)

I realised that I wasnt happy with the Power indicator on the previous version of the case - it was super dim.
I tried various "clear" filament but none of it was clear enough. In the end I bought some clear plastic rod to carry the light to the surfsce of the case.
I had to adjust the hole size because the smallest rod I could easily get was 4mm - but now all of the holes on the bottom left are 4mm, and it looks nicer.

I decided to print the new top shell in "CPS2 blue" - the color is a really good match for the original shell. I polished this one to get the texture to match the CPS2 shell. So just like a CPS2 A+B set the top of the case will be blue, the bottom will be black and the buttons for service and test will be that blue on black contrast that I love on CPS A boards.
IMG_5545.jpgIMG_5546.jpg

so now I'm 100% happy with the layout and human factors of the top shell.

Unfortunatley I'm still not there with the printing - different colored filament - even the same brand - print differently. In black the words are cleanly cut and clear, in blue they are a bit mushy around the edges. The blue filament has a lower melting point than the black and really sharp edges, like the edges of small letters - tend to run together.

To fix this i've adjusted the font and incresed the size of the letters 50%, I'll print another blue case and see if i can get it perfect\

I've also ordered enough parts from Digikey to complete the other two miniguns - one will be plain black, one will be CPS2 blue, and if I can get a good color match I'll see if I can print a japanese green one too.
 
Any plans to share your case files for others to use?
Of course! It would be a sad day indeed if I used an enjoyed someone elses hard work, had my fun with it, and didnt contribute my work back for others to use and enjoy too. The entire point of this thread is to share what we learn making these projects :)

I'm pretty sure this is the final final (of course I was absolutely sure that the previous 7 were also the final one...) - it's printing now - will know in 8 hours
Screen Shot 2021-05-15 at 11.36.04 AM.png


I'll probably publish as follows to thingverse as a remix to the Pacifier case
Top Minigin Advanced 2.5 (probably 2 dozen fixes, added bottons etc)
Bottom Minigun Advanced 2.5 (audio out now fits, adjusted nut size to reduce BOM part count by 1)
Pushbuttons (new)
BOM for nuts/bolts/posts
Will post a link to the thingverse here when it's up
 
Still working on Frank_FJS Minigun Supergun Advanced 2.5 (9 pin)

There is a problem in the mesh of the top shell that causes the print to fail half way every time - cura will slice it fine, but i wont print. My 3D modeling buddy reckons I probably have some leftover internal shapes and will hve a look for me.

Anyhoo - in the meantime back to finishing off the boards.

I've been using Isopropyl alchol to clean my boards until now, and while it does clean off the flux it always leaves a sticky residue. Acetone does a better job with the residue but it absolutely stinks and damages ABS plastic and some components too. I needed a better plan
IMG_5658.jpgIMG_5660.jpg
So today I got some specialist PCB board cleaner. This stuff gets the boards really, really, really clean - I love it. Im amazed that this stuff is still legal, I guess the amount of PCB manufacturing that goes on in this country is small enough that they havent bothered to legislate.
IMG_5662.jpg IMG_5659.jpg
It makes the boards so clean that I'm actually ashamed to touch them afterwards :) When they go into the cases I'll know that there arent any grubby fingerprints on the boards inside.

Tomorrow it's back to printing more cases - sigh...
 
Still working on Frank_FJS Minigun Supergun Advanced 2.5 (9 pin)

The evolution of the Minigun Supergun Advanced 2.5 9-pin case. This tool a LOOOOOOONG time for me to get right, and I was not even starting from scratch, but I learned a huge amount about 3D modeling, 3D printing and materials science in general.

The top half of the shell took 13 revisions to get "right" - and the slicer still hates it
The bottom half of the shell took 9 revisions to get right
IMG_5752.jpg


So here is the finished article.
IMG_5754.jpgIMG_5753.jpgIMG_5755.jpgIMG_5756.jpgIMG_5757.jpg

My 3D printer absolutely hates printing these, and my slicer (Cura) complains long and loud every time, but after much swearing I've finally managed to get "perfect" prints.

Key lessons learned
1) Thermoplastics are - thermally sensitive, if your print area is too hot or cold they will curl and run the print
2) PLA absorbs water from the air - it needs to be kept in a constant humidity environment to deliver consistent prints
3) Sketchup Make is a massive pain in the ass
4) Really slow first layer print speeds are the trick to getting really clear embossed writing on the top/bottom - the finals were done at 25mm/sec
5) Small layer height means the writing on the sides looks nicer - the finals were done at .12mm
6) Infill is a trade-off between strength and print reliability, especially on big flat pieces like these - I eventually got these to print at 100% infil by putting the printer into a temprature controlled, humidity controlled, draft-free environment (two larg cardboard boxs, printer in one, dehumidifier/heater in the other, offset pipe between them)

This is probably as good as its going to get for this particualr project - now I can start assembly and see if I got all the tolerances right :)
 
Back
Top