|
PoizenJam posted:Sadly, I tried asking Pioneer for a repair guide or specs. They don't release those. Let me introduce you to a very powerful magic spell. Say it along with me now: "service manual"! Presto! Page 81 lists all of the parts on that board. Googling a few of the part numbers shows that some people sell factory parts. Whether or not they have them in stock is another issue. kid sinister fucked around with this message at 21:32 on Feb 6, 2020 |
#
?
Feb 6, 2020 21:26
|
|
|
|
| # ? May 9, 2024 13:09 |
|
|
God bless those people that scan and upload manuals. I have used them for lawn mowers, old tractors, washing machines, and electronics over the decades. Even if I have the manual myself I just look it up online.
|
|
#
?
Feb 6, 2020 22:26
|
|
|
Does anyone know what that kind of semi-transparent plastic is that you put in front of LED 7-segment displays to improve the contrast? Can I just buy a sheet of that somewhere to cut pieces I need? e: This EEVBlog thread (that I couldn't find via google until I had already made this post for some reason) has some suggestions, such as a circular polarizing filter or just printing various densities of grey onto a transparency sheet with a laser printer, I guess I'll try those. Shame Boy fucked around with this message at 17:36 on Feb 7, 2020 |
|
#
?
Feb 7, 2020 17:30
|
|
|
if I'm thinking of the same thing you are, I think it's literally just deep red clear tinted plastic. The tint blocks like 70% of the ambient light reflecting off the LED surround but allows the emitted light to shine through mostly unobstructed. Try some theatrical gels or even just a red plastic report cover from Staples
|
|
#
?
Feb 7, 2020 17:37
|
|
|
I think it's just stage light gels.
|
|
#
?
Feb 7, 2020 17:40
|
|
|
Sagebrush posted:if I'm thinking of the same thing you are, I think it's literally just deep red clear tinted plastic. The tint blocks like 70% of the ambient light reflecting off the LED surround but allows the emitted light to shine through mostly unobstructed. Try some theatrical gels or even just a red plastic report cover from Staples Cojawfee posted:I think it's just stage light gels. Yeah that thread I found said that's what cheaper things use, especially if they only need to transmit one wavelength. They also suggested transparent vinyl colored tape just stuck to some clear acrylic which seems easy enough. However I have a few bi-color components so I'd need something that can transmit at least red and green, so I'm guessing the only option there is gonna be polarizers or grey filters.
|
|
#
?
Feb 7, 2020 18:23
|
|
|
I have some clocks with bare 7-segment displays and I use vinyl headlight tinting foil, costs nothing on eBay.
|
|
#
?
Feb 7, 2020 18:27
|
|
|
longview posted:I have some clocks with bare 7-segment displays and I use vinyl headlight tinting foil, costs nothing on eBay. Oh that stuff looks perfect, thanks!
|
|
#
?
Feb 7, 2020 18:31
|
|
|
I'm not sure if this is useful, but I came across this adafruit project where someone built 7 segment LEDs with neopixel LED strips: https://learn.adafruit.com/ninja-timer-giant-7-segment-display
|
|
#
?
Feb 7, 2020 18:31
|
|
|
Someone was doing something similar to that in the arduino thread and seemed to have nothing but problems with it.
|
|
#
?
Feb 7, 2020 18:37
|
|
|
Grey tinted plexiglas is better than red, I have a paper on this, let me dig a moment. Well I couldn't find it. Once upon a time we switched from red to grey lenses for my companies red LED displays due to the paper. The gist is that there is better contrast with grey. We use tinted plexiglas but window tint or stage gels should work great. They also look more professional in grey IMO. taqueso fucked around with this message at 19:27 on Feb 7, 2020 |
|
#
?
Feb 7, 2020 19:13
|
|
|
Shame Boy posted:I'd need something that can transmit at least red and green, so I'm guessing the only option there is gonna be polarizers or grey filters. Or yellow, if it uses the correct dyes
|
|
#
?
Feb 7, 2020 21:05
|
|
|
I've seen stick on films for that. They're used to covering power LEDs that are a little too bright. You can buy a sheet of them in different shapes on Amazon.Unperson_47 posted:God bless those people that scan and upload manuals. I have used them for lawn mowers, old tractors, washing machines, and electronics over the decades. Even if I have the manual myself I just look it up online. God bless the old farts at companies that have been around 90+ years (HP, GE, etc) that go into the file cabinet archives and scan all their old manuals to put them online. PoizenJam, service manuals are magical tomes that contain disassembly instructions, exploded views, part numbers, troubleshooting tables and more. kid sinister fucked around with this message at 21:13 on Feb 7, 2020 |
|
#
?
Feb 7, 2020 21:07
|
|
|
kid sinister posted:I've seen stick on films for that. They're used to covering power LEDs that are a little too bright. You can buy a sheet of them in different shapes on Amazon. I remember working at a metal stamping place and we made some component for GE circuit breakers. The print we had had an initial approved date in the 40s, and revision dates in the 60s, 70s, 80s, 90s, and 2000s. All the revisions were "verified product compliance." It was still dimensioned in fractional inches and three generations of engineers (including the Metrification ones) said "send it."
|
|
#
?
Feb 8, 2020 12:09
|
|
|
Holy poo poo- those diagrams are incredible. And thanks for the advice- ‘blow out diagram’ was vocabulary I did not know, and I did not think that service manuals would be leaked like that! In any case... I don’t think the LEDs are actually labelled in the manual- unless I’m missing something. Which the manual indicates means ‘not replaceable’. But I disassembled it regardless and they are indeed 4 leg through ole, dome type. 1 ground and 3 colours each soldered to a (what I believe to be) an smd style resistor.
|
|
#
?
Feb 8, 2020 21:59
|
|
|
I'm trying to 3D Print some frames to place over LED strips. Any ideas for ways to stop the light from bleeding through the material? Even 100% infill doesn't really help. It's so small that adding metal reflectors seems really difficult. Is there some kind of paint-on metallic paint that reflects 100% of light? This is similar to the LCD-like segmented numeric digits y'all were talking about a few pages ago so I figured I'd ask. edit: I'm going to try just adding some black acrylic paint and see what happens. Unperson_47 fucked around with this message at 02:42 on Feb 9, 2020 |
|
#
?
Feb 9, 2020 02:39
|
|
|
Unperson_47 posted:I'm trying to 3D Print some frames to place over LED strips. Any ideas for ways to stop the light from bleeding through the material? Even 100% infill doesn't really help. Fran made a 3D printed LED display thing and she just used duct tape and white paint for that: https://www.youtube.com/watch?v=M3jyMX8EXAQ In my own personal experience, yeah black paint worked completely fine for a light box thing I made. I think I put two coats of matte latex paint on and it worked fine. I'd think white paint would be better since it would reflect instead of just absorb, but in my case I also wanted to tame down internal reflections.
|
|
#
?
Feb 9, 2020 02:50
|
|
|
I'm trying to get started in learning FPGA programming, and I've been reading up on different types of architectures and interfaces which led me to a weird idea for a special type of CPU cache or something. I'm just wondering if this is a thing that's already done (or if not, why its a dumb idea): The idea is to implement a stack for a processor directly in hardware as a bunch of serial-in serial-out, bi-directional shift registers. So you would chain them serially as deep as you want, and then parallel them as wide as a machine word. Then a special register in CPU would be wired directly to the top of the stack for push/pop operations. Seems like it could be really efficient (and low latency?) to not require any sort of addressing or stack pointer etc., but I don't really know poo poo about IC design.
|
|
#
?
Feb 9, 2020 09:58
|
|
|
What if you actually want to address the stack? RTOSes in particular do some fuckery with the stack pointer to maintain a different stack for every thread. I don't see any barrier to you implementing that on your own design(not a CPU architecture wizard though), but optimisations often have flexibility trade-offs, making the CPU less useful as a whole
|
|
#
?
Feb 9, 2020 10:08
|
|
|
I'm rusty on the CPU architecture but that sounds like a run of the mill stack register? It sounds like a simple way to code it in verilog but I don't think its much of a resource saver. Embedded rams are one clock access and a mux structure for addressing is also really efficient in luts. In-fact the synthesizer is always looking for ways to stuff things in ram and might infer one on its own in this case.
|
|
#
?
Feb 9, 2020 18:21
|
|
|
PoizenJam posted:Holy poo poo- those diagrams are incredible. And thanks for the advice- ‘blow out diagram’ was vocabulary I did not know, and I did not think that service manuals would be leaked like that! You'd be surprised how far back you can get documentation for stuff. I've been working on old radios recently. It's fun seeing schematics dated pre-WW2. Service manuals are available for sale to the public a lot of the time. Sometimes they're only for sale to authorized service centers, but those do have a tendency to leak... Indeed, "service manual" is a very useful phrase to know. Like Unperson_47 mentioned, it's not just for electronics. You can get them for appliances, lawn equipment/small engines, cars... You'd be surprised the amount of things you can fix if you Google the make and model along with "service manual". babyeatingpsychopath posted:I remember working at a metal stamping place and we made some component for GE circuit breakers. The print we had had an initial approved date in the 40s, and revision dates in the 60s, 70s, 80s, 90s, and 2000s. All the revisions were "verified product compliance." It was still dimensioned in fractional inches and three generations of engineers (including the Metrification ones) said "send it." Let me guess... You all were stamping lock out slides to keep the breaker actuators from being switched on during electrical work, right? I'd believe the compliance updates. The modern breaker panel hasn't changed much in 50+ years. Even the new larger breaker types like GFCI and AFCI are still made to fit the old panel dimensions. Those old, larger dimensions make cramming modern SMD components inside the breaker housings even easier. kid sinister fucked around with this message at 19:48 on Feb 9, 2020 |
|
#
?
Feb 9, 2020 19:40
|
|
|
One more tip for finding documentation for really old stuff if you can't find any online and the company is still in business. Email their customer service. Get an actual human being. You'll have an 90-100% chance of getting help. If that company has since been bought out, try emailing the new owners. In this case, you got <50% chance of getting help. It's better than nothing. The chance varies with how old the thing is and how popular its model was.
|
|
#
?
Feb 10, 2020 02:12
|
|
|
kid sinister posted:Let me guess... You all were stamping lock out slides to keep the breaker actuators from being switched on during electrical work, right? Nope, this was something internal. It was brass and like 3/16" wide and 7/8" long with a #4-40 tapped hole in one end and a #6-32 in the other. The dies were ANCIENT for this thing. Trying to keep a #6 hole centered in a piece of brass that tiny was challenging.
|
|
#
?
Feb 10, 2020 12:31
|
|
|
Unperson_47 posted:I'm trying to 3D Print some frames to place over LED strips. Any ideas for ways to stop the light from bleeding through the material? Even 100% infill doesn't really help.
|
|
#
?
Feb 10, 2020 13:11
|
|
|
evil_bunnY posted:I'd use metal foil or matte paint. Copper tape works pretty well, too.
|
|
#
?
Feb 10, 2020 13:50
|
|
|
kid sinister posted:One more tip for finding documentation for really old stuff if you can't find any online and the company is still in business. Email their customer service. Get an actual human being. You'll have an 90-100% chance of getting help. If that company has since been bought out, try emailing the new owners. In this case, you got <50% chance of getting help. It's better than nothing. The chance varies with how old the thing is and how popular its model was. Finnex support told me which capacitor to desolder and replace inside the power supply to stop my aquarium lights from flickering 
|
|
#
?
Feb 10, 2020 14:48
|
|
|
insta posted:Finnex support told me which capacitor to desolder and replace inside the power supply to stop my aquarium lights from flickering That is often both very good and very bad. E: Usually much better that sales people doing customer support though. CopperHound fucked around with this message at 19:06 on Feb 10, 2020 |
|
#
?
Feb 10, 2020 19:03
|
|
|
FISHMANPET posted:Feeling like I'm in crimping hell... So following the advice here and trying to find some good ends. I found what I thought would be a good set on Amazon that were no-name but at least trying to emulate the full flexible jacket of the nicer brands. Crimped well enough for me except they were the wrong tab size (my fault). The terminals I'm trying to connect to are 2.8mm/.11 and there's nothing that size at Home Depot, and even on Amazon it's garbage. So I'm looking to Mouser. The "terminals" category is pretty overwhelming with over 10k options and so many size variations (2.79mm, 2.80mm, 2.81mm) for what is basically the same size, but I think I've narrowed it down to these two from TE (formerly Tyco): https://www.te.com/usa-en/product-7-520366-2.html These are 2.79 mm x 0.51 mm for 22-26AWG with a purple sleeve https://www.te.com/usa-en/product-2-520084-2.html Also 2.79 mm x 0.51 mm but for 18-22AWG with a red sleeve There's also this third option: https://www.te.com/usa-en/product-2-520273-2.html 2.79 mm x 0.81 mm (so accommodates thicker tab or would be looser than above) but still with the red sleeve. All 3 of these are from the same product line (page 3). I'm sure the only way to be sure if I need .51 or .81 is to get a digital calipers so I guess that's just a gamble, but .51 seems much more standard.. I'm using 22 AWG solid core wire which is why I'm kind of on the fence about the red or the purple. Not sure if my basic crimpers would even be small enough to crimp the purple connector, or what kind of crimper I could even get for that that wouldn't cost $500. I think I've talked myself into the red 2.79 mm x 0.51 mm but just want to see what anyone else thinks because I'm loosing my mind trying to connect these terminals. And learning the importance of good specs.
|
|
#
?
Feb 10, 2020 19:41
|
|
|
e: sorry, i have bad reading comprehension apparently That type of terminal is pretty basic and you should be able to crimp it with a < $40 tool. Pick something that has your wire size in the listed range and it should crimp ok. However, purple is a new one to me. I would assume those switches use the thinnest material they can get away with, I'd go .51. If those switches are IDEC A6 clones, I think the datasheet says they are 0.3mm thick? 
taqueso fucked around with this message at 20:16 on Feb 10, 2020 |
|
#
?
Feb 10, 2020 20:07
|
|
|
Holy poo poo that's amazing. Yeah that looks to be what it's a copy of (or a Kacon K16 which appears to be identical). Anyway both the A6 (AL61-M1) and K16 (K16-281) spec sheet say the terminals are 2.8 x 0.5. It's not shown in your particular picture but I've got what IDEC calls a SPDT Illuminated or what Kacon is calling Illuminated 1C. That's exactly what I needed!  
|
|
#
?
Feb 10, 2020 21:13
|
|
|
OK, looking for some more part recommendations on this project. As a summary it's all flash and no substance, I just want as many knobs and triggers and blinking lights as possible. Inspired by the creator of the project, I want to emulate the needing to turn two keys like launching a missle (as seen in the intro of Wargames). As far as implementing it, I've got one free GPIO pin on my raspberry pi so I think I'd just get two key lock or selector switches and wire them up like this: Each would be spring return so there would need to be a hand on each to actually complete the circuit, and I'd just treat that as a button press in my code and act accordingly. But I'm not really sure what to actually use for the switch. I can find plenty of inexpensive key lock switches, but finding one with lights is proving to be difficult. I'd even settle for a spring return selector switch if it was illuminated, but still can't even find that. I've found something on AliExpress, and searching the part number takes me to a Chinese company called OnPow that in theory makes a spring return 2 position key switch, but I'm not sure if there's any way I could actually get it in single quantities for myself. I can't find any design like this from any of the "bigger" names I've come across while looking for parts on this project. I could also just get a non-illuminated key, and use some neopixel rings to provide the light (which the project is already using, so would be easy for me to control). The other piece of that is I'm basically looking for the most ridiculous connector to use for this, but also one that isn't likely to be used for power. I don't want someone to find some random power adapter and plug it into the socket and fry everything by shoving the wrong voltage in. So I'm thinking a standard DIN connector, unless anyone has any suggestions for "weirder" looking connectors. I'd be building these key boxes completely from scratch so making everything myself so I'd have complete freedom to do whatever here.
|
|
#
?
Feb 12, 2020 23:16
|
|
|
I have no part recommendations, but I would try really hard to find a second gpio so they can be non momentary keys that you have to turn simultaneously. Or have two entirely separate command stations connected by a cable (or zigbee if you want to be extra flash) that need simultaneous keys
|
|
#
?
Feb 13, 2020 01:28
|
|
|
Is there a chance you can throw a tiny breadboard in the box as well? Get a 74**08 (HC or LS depending on your logic level) and connect the switches to the inputs of an AND gate, and the output to your GPIO.
|
|
#
?
Feb 13, 2020 02:29
|
|
|
Just use two GPIOs and do it in software. Yeesh. Then you can also have the system work like the real thing does, where either key starts a brief timer during which the other one must be turned or it resets. As is, you could turn one key and jam it in place somehow then run across the room and turn the other and that's just not acceptable 
|
|
#
?
Feb 13, 2020 03:02
|
|
|
I've definitely been about the best way to do this. There will be a board in the project already so I can put in a chip if I need to, though I'm not sure what putting the signal through an AND gate gets me that I don't get from just making it a single loop, in this application. It seems like even with the AND gate, you could turn one of the switches on and hold it, and then once the other one turns on the gate would activate. I'm not good enough with Python and threading yet to figure out how exactly I would implement this entirely in code with two GPIO pins, and I also have to figure out how to add additional GPIO pins (it looks like the GPIO library I'm using doesn't support the normal port expander so I need to use a different library for those pins etc etc). None of those are unsolvable problems but I want to keep the scope of this project from spiraling out of control if possible. For now at least, I think I'm fine with the compromise of both switches being momentary, so if you're not holding onto them they'll revert back to off. Since they're just boxes connected with wires there's no guarantee that they won't be next to each other, but in theory it will take two hands to turn a switch - one to hold the box and the other to turn the dial - so even if they were next to each other it would be difficult for one person to trigger both switches.
|
|
#
?
Feb 13, 2020 04:09
|
|
|
You don't need threading to check the state of two buttons, or to read them, store the time, and then check the time in the next iteration through the loop
|
|
#
?
Feb 13, 2020 04:11
|
|
|
I'd simply hold the boxes together with my feet and use one hand per key.
|
|
#
?
Feb 13, 2020 04:16
|
|
|
Radio time of flight sensor requiring the boxes to be 5+ meters apart e: Turns out people actually make RF modules with time of flight + phase difference measurement circuits for ~$10, so you could totally actually do that. Foxfire_ fucked around with this message at 05:42 on Feb 13, 2020 |
|
#
?
Feb 13, 2020 05:16
|
|
|
Float one key at 100V relative to the other so if someone touches both at the same time they die. Simple!
|
|
#
?
Feb 13, 2020 07:04
|
|
|
|
| # ? May 9, 2024 13:09 |
|
|
Splode posted:Float one key at 100V relative to the other so if someone touches both at the same time they die. Simple! This is the answer. You just need two level converters.
|
|
#
?
Feb 13, 2020 07:08
|
|