|
Looking for ideas / inspiration. Recently moved into an old home with a laundry chute. Laundry machines are in the basement, living spaces on the second floor. Dropping clothes from the top is awesome, but bringing piles of laundry upstairs is not. Idea: Laundry chute elevator for bringing clothes back up. My idea of how to do this is with an electric hoist at the top attached to a custom made wooden box that would hoist the laundry back up. I figured buying a premade motor like that would be easier than crafting a motor and winch system on my own, but I'm open for alternative ideas. Any thoughts? Also, if I were to use a hoist like the above, it looks like i'd need to interface with that pendant switch. I found this documentation, but I do not fully understand how the relays work with the capacitors. Would someone mind clarifying how this system works for me in an ELI5 manner?
|
#
?
Jul 6, 2019 16:45
|
|
|
|
| # ? Jun 11, 2024 00:43 |
|
|
I think if you used regular mechanical relays for the motor control instead of SSRs (which will be fine, since you aren't going to be triggering them more than a couple of times a day) you don't need to deal with that discharging circuit. I would go that way. Make sure you use relays rated for the expected power draw, and notice that a given relay can switch much more AC than DC.
|
|
#
?
Jul 6, 2019 17:21
|
|
|
Kind of a long shot on this one but maybe someone knows something. I'm trying to track down an infuriating bug with a 3D printer where the x-axis motor randomly stops working. So far I have chased it down this path: - motor stops working suddenly in the middle of a movement - this is because the motor driver is getting reset to junk settings - this is because it's receiving bad data on its SPI control lines - this is because there are random voltage spikes on the control lines - this is because the entire printer's power supply is spiking - I cannot figure out where the gently caress these transients are coming from. Here's a scope trace:  1, 2 and 3 are SCK, MOSI and CS for the x-axis driver, respectively. 4 (dark blue) is connected directly to the PSU's 24v rail. As you can see, it's spiking up to 60 volts(!) and ringing back down to -14v for a total swing of 75 volts. I guess the reason that nothing has exploded is because the spike only lasts 20ns and the whole event is over in about 300ns. The spikes do not correspond with obvious actions like starting or reversing a motor. They only occur when a motor is moving, but they happen at the beginning, the end, and in the middle of a movement with equal frequency. All of the heaters and fans are disabled and there is nothing drawing a load other than the motherboard, the active motors, the LCD screen, and the leveling probe. Where is this coming from? How is it defeating the power supply's filter capacitors? How can I stop it from happening or at least soak it up? Sagebrush fucked around with this message at 01:17 on Jul 7, 2019 |
|
#
?
Jul 7, 2019 00:55
|
|
|
Mr. Bubbles posted:Looking for ideas / inspiration. Any chance you could put a really strong vacuum on the upstairs end of that tube and just suck all the clothes back upstairs instead?
|
|
#
?
Jul 7, 2019 03:45
|
|
|
Sagebrush posted:Kind of a long shot on this one but maybe someone knows something.
|
|
#
?
Jul 7, 2019 11:41
|
|
|
GnarlyCharlie4u posted:Any chance you could put a really strong vacuum on the upstairs end of that tube and just suck all the clothes back upstairs instead? That would be awesome but it's 4 walls of wood making the chute, not a tube so I don't think it'd have good enough of a seal.
|
|
#
?
Jul 7, 2019 15:31
|
|
|
Sagebrush posted:
PSU filter caps generally don't work at high frequencies. this is definitely something arcing. I'd try small capacitors (100 pf -> 1 nF range) between each line and ground. Barring that, there are differential SPI transceivers you can get to turn it into balanced data across cat5 pairs. I had to use that when I ran a 100 ft SPI line up to an accelerometer on my radio mast.
|
|
#
?
Jul 7, 2019 17:47
|
|
|
Mr. Bubbles posted:That would be awesome but it's 4 walls of wood making the chute, not a tube so I don't think it'd have good enough of a seal. Do you have room above the chute to house the motor? As someone who doesn't have a laundry chute, I am very jealous.
|
|
#
?
Jul 7, 2019 18:41
|
|
|
Assuming it's been built with the assumption that things are only ever going to go down it, make sure there aren't any weird lips or funnels anywhere that a motorized platform might get hung up on.
|
|
#
?
Jul 8, 2019 00:38
|
|
|
I've got some neopixels I want to drive from a little BLE module. Problem is, the neopixels are the through-hole sort, which require both a 5v power supply and 5v logic level. The BLE board runs at 3v, and I'm powering this from a little 3v lipo battery. (And yeah, I tried running them on 5v supply and 3v logic, to no avail.) For providing power to the thing, I'm just going to use something like this: https://www.adafruit.com/product/3661...it's cheap, it's simple, and it provides enough power for my needs (I'm only driving a couple neopixels.) So that gets me a 5V supply, cool. For the logic, though, I'd like to not bother with a full on logic level converter, that seems overkill and wasteful. The communication is all one-way over a single wire, it's just bit banging the data down the line. I have a big bin of 74xx parts from days of yore, and I'm thinking I can probably make it work with those. I'm thinking a 7407 would be the most ideal for that, but I don't seem to have any. So, we make do: I'm thinking either a 7404 and just chaining a couple of the gates together to mimic a 7407, or I also have some 7408s and I could just tie the inputs on one of the gates together. Is this a dumb idea? The idea here would be to provide 5v on vcc, and just drive the logic input at 3v, which should be enough since it looks like they read high at 2v, standard ttl. Do I need to worry about screwing up my BLE board doing this? Should I throw a diode or something in there to protect it? It's been a loooooong time since I've done this sort of stuff, so I'm a bit rusty, and this was never my field anyhow.
|
|
#
?
Jul 9, 2019 04:17
|
|
|
I recently did some restoration work on a very old hand-cranked telephone and part of my compensation was getting to keep the dynamo. It's built like a brick shithouse and seems in great shape mechanically, aside from being dirty as hell and any possible brushes assumably being in poor condition. You turn a gear on the side and a rotor spins inside so that seems like half the battle. What's a good way to do some diagnostics and restoration on something this crude? I'm not even entirely sure where I'd attach the multimeter leads on this thing, or what to expect as a readout. My best guess is to dig up an old incandescent bulb and wire it between what i'd guess are teriminals and give it a spin and see what lights up, but incandescents are rather out of style and I have none on hand (and a boring ol resistor as load does'nt provide particularly useful immediate/visual feedback)
|
|
#
?
Jul 9, 2019 04:19
|
|
|
Hold the leads in your mouth and give 'er a spin never take my advice
|
|
#
?
Jul 9, 2019 04:22
|
|
|
Bad Munki posted:I've got some neopixels I want to drive from a little BLE module. Problem is, the neopixels are the through-hole sort, which require both a 5v power supply and 5v logic level. The BLE board runs at 3v, and I'm powering this from a little 3v lipo battery. (And yeah, I tried running them on 5v supply and 3v logic, to no avail.)
|
|
#
?
Jul 9, 2019 04:42
|
|
|
Bad Munki posted:I've got some neopixels I want to drive from a little BLE module. Problem is, the neopixels are the through-hole sort, which require both a 5v power supply and 5v logic level. The BLE board runs at 3v, and I'm powering this from a little 3v lipo battery. (And yeah, I tried running them on 5v supply and 3v logic, to no avail.) Check this for simplicity: https://hackaday.com/2017/01/20/cheating-at-5v-ws2812-control-to-use-a-3-3v-data-line/
|
|
#
?
Jul 9, 2019 12:09
|
|
|
Bad Munki posted:Hold the leads in your mouth and give 'er a spin never take my advice My 7th grade science teacher had one, and would give an A to anyone who could hold the leads in their hands for one full crank of the handle. Even with 10 or 15 of us in series it was quite the tingle.
|
|
#
?
Jul 9, 2019 14:44
|
|
|
One Legged Ninja posted:My 7th grade science teacher had one, and would give an A to anyone who could hold the leads in their hands for one full crank of the handle. Even with 10 or 15 of us in series it was quite the tingle. It was actually used as a torture device called the Tucker Telephone at one time so congrats on your science teacher for being one of the few brave enough to literally torture their students 
|
|
#
?
Jul 9, 2019 15:02
|
|
|
If you get waterboarded do you get an A+?
|
|
#
?
Jul 9, 2019 15:06
|
|
|
Bad Munki posted:Hold the leads in your mouth and give 'er a spin never take my advice I actually did manage to get something from what appeared to be the terminal posts- voltage never budged no matter what i did, amps never moved a hair if the dynamo was rotated below a sudden speed, but very quick/jerky turns, even a half-turn of the wheel done with a snap of the wrist, caused a very large and brief current spike that maxed out the usual ampera- i actually had to use the 10A unfused tap on the multimeter, which I've never had cause to touch, and I think it briefly exceeded the 10A max reading. No idea how to interpret that, my core electrical knowledge is thin at best, but it's interesting.
|
|
#
?
Jul 9, 2019 21:29
|
|
|
Ambrose Burnside posted:I actually did manage to get something from what appeared to be the terminal posts- voltage never budged no matter what i did, amps never moved a hair if the dynamo was rotated below a sudden speed, but very quick/jerky turns, even a half-turn of the wheel done with a snap of the wrist, caused a very large and brief current spike that maxed out the usual ampera- i actually had to use the 10A unfused tap on the multimeter, which I've never had cause to touch, and I think it briefly exceeded the 10A max reading. No idea how to interpret that, my core electrical knowledge is thin at best, but it's interesting.
|
|
#
?
Jul 9, 2019 23:44
|
|
|
Ambrose Burnside posted:I actually did manage to get something from what appeared to be the terminal posts- voltage never budged no matter what i did, amps never moved a hair if the dynamo was rotated below a sudden speed, but very quick/jerky turns, even a half-turn of the wheel done with a snap of the wrist, caused a very large and brief current spike that maxed out the usual ampera- i actually had to use the 10A unfused tap on the multimeter, which I've never had cause to touch, and I think it briefly exceeded the 10A max reading. No idea how to interpret that, my core electrical knowledge is thin at best, but it's interesting. If it's a telephone hand crank, then it's probably producing directly the ringing noise in the other phone. In later years, that would be standardized at 300VAC at not too much current and some frequency. Basically, a "ringing phone" sound is a bell solenoid directly acted upon by this crank. There may be an impulse coupling or some other mechanical method to ensure that a pulse doesn't go out unless of sufficient impetus to actually ring a phone.
|
|
#
?
Jul 10, 2019 03:00
|
|
|
Sir Bobert Fishbone posted:Check this for simplicity: https://hackaday.com/2017/01/20/cheating-at-5v-ws2812-control-to-use-a-3-3v-data-line/ I'm not 100% sure that'll actually work with the through-hole neopixels, they don't function quite the same as the regular strips do. Maybe it'd work! But my set of pixels for this project is a whopping two so I'm also not inclined to sacrifice the first one's performance. Anyhow, the 7408 worked fine. Next step is writing a better control app to drop on my phone. Thanks for the sanity check, y'all. e: 
|
|
#
?
Jul 10, 2019 17:02
|
|
|
Looks rad. I've got an interesting problem that is impossible to Google for. I'd like to take a standard fluorescent light ballast, and get a steady 24VDC out of it. It's just the most convenient power source. Pulling apart a ballast, it's completely potted, so I can't tell anything useful about how it's made. The faceplate says 600V out, and I haven't turned it on at the bench for more investigation because I don't want to die. Are there any cheap off the shelf solutions for this, or any design tips if I spin my own solution? I am, in general, a professional, so I can be reasonably safe about high voltage.
|
|
#
?
Jul 11, 2019 01:24
|
|
|
Bad Munki posted:I'm not 100% sure that'll actually work with the through-hole neopixels, they don't function quite the same as the regular strips do. Maybe it'd work! But my set of pixels for this project is a whopping two so I'm also not inclined to sacrifice the first one's performance. I hand-soldered a string of 340 ws2812b LEDs, and it worked OK for that, but then again when you've got 340 of them you don't give a poo poo about the performance of the first one. When you've got a string of...two...your priorities change a bit.
|
|
#
?
Jul 11, 2019 02:01
|
|
|
ante posted:Looks rad. If you have access to the ballast, don't you have access to the wires going into it (110VAC) that you could hook any normal screw-down power supply to? I'm sorry to be the guy questioning your question, but I'm imagining being on that ladder and I can't see why you'd want the ballast output instead of input.
|
|
#
?
Jul 11, 2019 05:31
|
|
|
Stabby McDamage posted:If you have access to the ballast, don't you have access to the wires going into it (110VAC) that you could hook any normal screw-down power supply to? I wanna be able to plug a thousand devices into a thousand existing fixtures, without having to rewire a thousand fixtures, twice
|
|
#
?
Jul 11, 2019 05:33
|
|
|
ante posted:I wanna be able to plug a thousand devices into a thousand existing fixtures, without having to rewire a thousand fixtures, twice Like the thing you build will replace the flourescent tubes? Is this to retrofit LEDs, because if so, I think there may be a commercial product for that. If it's for something more custom/weird, I'd start by seeing if the 600V is AC or DC. If it's AC, you could get close with a transformer, then use a normal AC/DC converter. How much current do you need at 24VDC?
|
|
#
?
Jul 11, 2019 05:45
|
|
|
Stabby McDamage posted:Like the thing you build will replace the flourescent tubes? Is this to retrofit LEDs, because if so, I think there may be a commercial product for that. Yes and yes, but also weird custom stuff. Like an ESP32 or something in there as well. Need about 2A. I thiiiink it'll be AC, and 600-ish volts to ignite the flourescent tubes, and then drop down to 100ish AC to maintain them - But as I said, this is just a guess, because internet searches on how the ballasts/flourescent tubes work are completely poisoned by electrical supply websites stating that ballast are magical black boxes that make your lights go. Likewise searches for existing power conversion modules that do what I want. I could possibly built my own solution with a transformer and rectify it, but I suspect it's not quite that simple, and would rather avoid the huge amounts of R&D required.
|
|
#
?
Jul 11, 2019 06:24
|
|
|
Bad Munki posted:I've got some neopixels I want to drive from a little BLE module. Problem is, the neopixels are the through-hole sort, which require both a 5v power supply and 5v logic level. The BLE board runs at 3v, and I'm powering this from a little 3v lipo battery. (And yeah, I tried running them on 5v supply and 3v logic, to no avail.) I would just spend the $0.40 on a purpose-built shifter instead of gluing two chips together.  You could also use one 7404 and invert the data you send if you're willing to dig down that far into the neopixel library (possibly the pin driver in your microcontroller can also be configured to do logic 0 -> high voltage, logic 1-> low voltage).
|
|
#
?
Jul 11, 2019 07:34
|
|
|
Foxfire_ posted:I would just spend the $0.40 on a purpose-built shifter instead of gluing two chips together. I mean, it's just the one component with four external connections, two of them being power, so it's really no different than a purpose-built module, except that I already have the IC on hand. Here's the whole thing:  Orange is the 3v input, green is the 5v output. Bad Munki fucked around with this message at 18:41 on Jul 11, 2019 |
|
#
?
Jul 11, 2019 18:38
|
|
|
um excuse me posted:Oscilloscopes. Something I last used about a decade ago with equipment from the 80s or 90s. I remember the one I had was a Tektronox 2 MHz analog. I used to use it with a signal generator and a logic tester. Is the market pretty much the same? I saw there are some USB based ones, are those useful? What about the multimeter ones? Are ones under $100 trash? Are there combination scope/signal generator/logic testers out there? I'm super out of touch but want to be able to test ICs. I was told to ask about this here. For context I want to use this on sensors and ECU on my car. So 12v max stuff. I was also told the Rigol 1054 is the generic answer. But a few options would be great.
|
|
#
?
Jul 11, 2019 19:38
|
|
|
High end hobbyist, it's hard to beat the Rigol Lower end, I was more impressed with the DSO Nano than I expected, and the form factor is really sexy. I think it's obsolete, but Seeed has a newer, better thing in the same kind of market
|
|
#
?
Jul 11, 2019 20:44
|
|
|
The only thing that steered me away from a USB scope is the somewhat low max voltage ratings. I almost definitely would have got one if I planned to only work with logic level stuff. E: with regards to two vs four channel, I didn't think I wanted more than 2 channels but I was super happy to have 4 when troubleshooting some SPI communication: Two for signal, one for clock, and one for chip enable. CopperHound fucked around with this message at 00:46 on Jul 12, 2019 |
|
#
?
Jul 12, 2019 00:42
|
|
|
CopperHound posted:E: with regards to two vs four channel, I didn't think I wanted more than 2 channels but I was super happy to have 4 when troubleshooting some SPI communication: Two for signal, one for clock, and one for chip enable. I've got a Salea 4ch logic analyzer for tasks like that. Relatively cheap, protocol analysis somewhat baked in.
|
|
#
?
Jul 12, 2019 00:52
|
|
|
Yeah, logic analyzers are great, and using a scope for it is annoying. There are combined units.
|
|
#
?
Jul 12, 2019 04:00
|
|
|
4 channel scopes are really really nice for any remotely complicated setup, though. Logic analysers good for, well, analysing logic.
|
|
#
?
Jul 12, 2019 04:25
|
|
|
A good electronics setup has both. Personally I would suggest getting the recommended Rigol 1054, which can do a good amount of logic analysis on its four channels while also showing the analog weirdness that sometimes is the real source of your problem; and also pick up one of these ten dollar knockoff Saleaes and use it with the Saleae software for the times when you need to record really long sequences of communication. Yeah it doesn't have the same performance as the original and using it makes the original developers cry but as a hobbyist I think you're fine. I have one of the knockoffs and it does just fine for hobbyist-speed logic. Incidentally, my previous post is a real good example of why you can't replace a scope with a logic analyzer, even if you think you're only going to work on digital logic: That's a trace (on a Rigol 1054, natch) of an extremely short event (look at the timebase) occurring across the CS, MOSI and SCK lines for one stepper motor driver on a 3D printer, and also the 3D printer's main power rail. I am running SPI decoding on it, because I assumed it was a communications problem, but it turned out to be static electricity -- the extruder carriage moving around on its plastic wheels was building up a charge that would periodically discharge into the printer's electrical system and pull all sorts of lines up and down and generally wreak havoc. Logic analyzers are not meant to look at 75v 10ns ringing transients and I doubt my knockoff Saleae would have caught it; certainly you don't have the same fine-grained waveform display and trigger control that you'd get on a scope. e: while the 1054 is an unbeatable value, it does still have a few little Chinesium things going on. 276.000000ns 
Sagebrush fucked around with this message at 05:08 on Jul 12, 2019 |
|
#
?
Jul 12, 2019 05:04
|
|
|
Sagebrush posted:A good electronics setup has both. Personally I would suggest getting the recommended Rigol 1054, which can do a good amount of logic analysis on its four channels while also showing the analog weirdness that sometimes is the real source of your problem; and also pick up one of these ten dollar knockoff Saleaes and use it with the Saleae software for the times when you need to record really long sequences of communication. Yeah it doesn't have the same performance as the original and using it makes the original developers cry but as a hobbyist I think you're fine. I have one of the knockoffs and it does just fine for hobbyist-speed logic. Just use Pulseview / Sigrok for software - it's open source and kinda designed to handle those "clones" (not actually related to the Saleae in any way, hardware-wise), and owns bones. Maybe slightly less than the Saleae software, but you're also being in the moral right, and I had to spend a lot of time loving around with my drivers to get the Saleae software to recognise it. Much better just using Sigrok. But yeah, those crazy-cheap devices are awesome for 99% of the logic work you'll need them for.
|
|
#
?
Jul 12, 2019 05:21
|
|
|
I recently ordered a 1054 at work for light lab use and it's a lot of scope for the money. The preset signal filters are really nice and make logic analysis pretty good, too. It's nice to have options. I use the Salea for extended recordings of some digital signals. It's perfect for that. Funny story about why I ordered the 1054 in the first place: We have multiple really nice Tektronix scopes sitting around, but they don't have any connectivity options (maybe GPIB?) and no storage except for a 3.5"" floppy drive. A lot of our applications involve recording and exporting data. That was the sole reason for buying a new scope - the 1054 has USB and network control and USB mass storage.
|
|
#
?
Jul 12, 2019 05:26
|
|
|
Bad Munki posted:I mean, it's just the one component with four external connections, two of them being power, so it's really no different than a purpose-built module, except that I already have the IC on hand. It's just so chunky and huge. If you're going to box it up and make it a permanent thing, give it a decoupling cap and tie off the unused inputs so they're not floating.
|
|
#
?
Jul 12, 2019 05:35
|
|
|
|
| # ? Jun 11, 2024 00:43 |
|
|
Definitely get the Rigol 1054 if you can afford it, it's great and you can software-unlock double the bandwidth and all the features for free using an online keygen, though I certainly don't advocate doing that because that would be a crime 
|
|
#
?
Jul 12, 2019 07:00
|
|