|
Stabby McDamage posted:I have one, and I'd love to play with it, but the stupid DC jack on it is different than the one China uses on everything else, so I haven't been able to test it yet. By default it comes with no power supply, as the audience is people who will solder a pigtail from an RC battery or similar for it. I managed to power it up by jamming a mismatched connector into it enough to verify it turns on and could heat up. I've since gotten the right connectors, but I haven't had time/interest to play with it more. Quick question, are any parts of the cartridge magnetic? I've been thinking of building a nonmagnetic soldering iron, but haven't found any tips/heating elements without steel or nickel in them.
|
#
?
Jun 16, 2017 01:03
|
|
|
|
| # ? May 28, 2024 18:35 |
|
|
ANIME AKBAR posted:Quick question, are any parts of the cartridge magnetic? I've been thinking of building a nonmagnetic soldering iron, but haven't found any tips/heating elements without steel or nickel in them. You just mean ferrous, right? None of it is actually magnetic (i.e. attracting iron by itself). In any case, I checked it with a little magnet:  Why is this desirable?
|
|
#
?
Jun 16, 2017 01:34
|
|
|
BRB, soldering inside an operating NMRI machine.
|
|
#
?
Jun 16, 2017 01:37
|
|
|
ANIME AKBAR posted:Quick question, are any parts of the cartridge magnetic? I've been thinking of building a nonmagnetic soldering iron, but haven't found any tips/heating elements without steel or nickel in them. I know gently caress-all about what soldering irons are made of so I can't help you, but I'm curious about what you want such a thing for.
|
|
#
?
Jun 16, 2017 01:50
|
|
|
Sorry about the longpost, this is going to be pretty entry-level stuff probably but I have spent too much time on this thing to give up on it tl;dr why does my cheap diy scope give just a negative V? Also do good-ish cheap/small oscilloscopes exists at all or should I just buy a 2nd hand bench model? anyway. I bought one of those chinese DIY oscilloscope kits because 1) I kind of need a simple scope sometimes 2) kits are fun and they're well cheaper than any assembled oscilloscope, new or second hand will likely ever be. Now I knew it wasn't going to be perfect but it would do for measuring slowish signals that originate from my synth and things like that. Plus good troubleshooting practice if it fucks up. Since I can kind of solder but far from perfectly I got the one where all the SMD parts were already done and saved myself that trouble. Took an hour or two of very careful assembly to get it to go from  (If I'd just put those things a little straighter, this would have been the perfect start to an inspirational electronics amateur instagram account) to  I'm still missing the rotary encoder there mostly complete. I think there's plenty of zoom to cast doubt on my soldering 'ability' but hooking it up, it's easy enough to calibrate it and go. I stick it into various small noise/voltage generators and an LFO which yield the expected output waveforms.  It responded quite nicely to different inputs, timebases, screen repositioning. All for 20$!Spirits high, I decide it's time to stick it into the supplied plastic enclosure and finally finish something for once. This is where things go downhill, rapidly. First of all, the rotary encoder for the center rotating encoder kind of hovers over 4 main board breakout pins on its own little board. The plastic front plate attaches to the encoder using two screws, the encoder pins are then soldered to the main board. 4 solder joints and 2 screws to keep the front plate on seems... little  Anyway there are a lot of extra variables at that point too because while doing this the small LCD must also be kept in its proper space and the spacing between board and front plate should also be even. This suckss because you need to screw the plate onto the encoder reaaalllyyy loosely, much more than seems right, or else the plastic warps/screen bends a little. Warped plastic renders the pushbuttons perma-pressed a well. So I did that wrong, corrected it and inadvertently broke up a rotary encoder solder joint which was fine while testing, didn't see this until later, etc etc, took me ages to get it right/figure out where I went wrong. In the end I went on slightly annoyed at my own incompetence with finehandiwork. The rest of the case was so badly fitted that even with a little bit of force to get the parts into their place it wouldn't really close. I'd had enough so I screwed it shut with a mm to spare on both plastic sides, loosely but close enough to keep things together. The rotary encoder wasn't pushable anymore due to plastic warping again. I put the whole thing down for the night. After a good nights sleep I decided to not use the enclosure at all since the scope was going to be limited to desktop use anyway (its fed off a 9V jack). Screwed it out of the plastic and boot it up. I'm greeted by  Red wire's on the test signal of 1kHz, amplitude either 3.3V or 0.1 V. The frequency is displayed correctly but my wave trace is gone and the voltages have all dropped deep into the negative in AC, DC and the GND modes. The trace is sort of plotted on the very bottom om the screen but no amount of recalibration/factory resets gets it up again. Doesn't seem that strange, given the voltages are super negative, all the time. Just touching the measuring probe used to give a cool little jump in the trace, but now I see no movement. I start remeasuring all testing points (helpfully provided by JYEtech in understandable english - they got their poo poo together very well for a Shenzhen budget diy manufacturer) Everything well within tolerance. Got out the loupe and reflowed/redid any less than ideal looking soldering (of my own, couldn't see SMD flaws). I'm new to this, so I was kind of running out of options but decided to recheck resistance across all resistors, which was good as well and cap polarities (fine) and obvious short. Nada. Googled, jyetech even has an active support forum but I could not find my problem. Also did not feel like regging there since I did all the standard troubleshooting steps they provide. Reflashing the firmware is currently not an option as I don't have a USB-to serial thingy (just a few arduino's/teensies) and didn't feel like figuring that out before finding out which bit failed in this instance. So I ask you guys (that are still reading): What failing component has the effect of screwing with the measured voltage so badly? The relative voltages are decent but it's just much too low. Calibration fails, so then maybe software? In test mode at the start, even without hooking up the probes there was a signal (so entirely without the board that fits the probes. caps, resistors for measurements). That's definitely not the case now, so i the main board (which doesn't carry a lot of the measuring circuit the main suspect? Here it is, without the analog board and components. schematics (pdf).  Bonus question: I decided this 15$ scope would be an easy way to check whether I'd find uses for one and even after only a half hour of properly working I know I'm going to end up with a 'real' one. For hobbyist purposes, like I said mostly synths/audio/some arduino signal tracing, is it smart to get an old used analog bench model or are handheld/touchscreen ones between 100-150$ also servicable? I don't put a premium on portability, though a small footprint is a plus which kind of rules out properly old CRT ones. This ill-fitting plastic adventure reminded me it's way better to buy good quality stuff once instead of a cheap buggy ones that break over time - so how much is the lower limit of 'good' in US $?
|
|
#
?
Jun 16, 2017 02:40
|
|
|
Old Binsby posted:Kit scope Considering it was working outside the enclosure, I wouldn't suspect the firmware needed to be reflashed since it was working at one point. So rule that out. Breaking a solder joint when mechanically fastening the boards together leads me to believe you physically damaged something. It requires a lot of force to break a through hole joint like that, so maybe you torqued the screws down too much? In general, its best not to reflow SMD components more than once. This is increasing in importance the smaller the component is. None of the components look fine pitch enough to where I'd consider them to require hot air work, but I don't know how fine your tools are. This looks like a great board to practice soldering on though, as its got a mix of very common package types. This is a good exercise in what you value your time as considering you invested $20. You've likely already extracted more than $20 in experience soldering and assembling electronics, so consider that a win. If you really don't want to give up, buy another kit and go through your process again, but pay very close attention to when you fasten it all together. Good luck! Edit: Obligitory check your scope ground. Ne Cede Malis fucked around with this message at 06:56 on Jun 16, 2017 |
|
#
?
Jun 16, 2017 06:36
|
|
|
Stabby McDamage posted:In any case, I checked it with a little magnet: Platystemon posted:BRB, soldering inside an operating NMRI machine. Bingo! My adviser told me that way back he had an aluminum/bronze butane iron which was nonmagnetic. But nowadays every tool is built with cheap chinese steel.
|
|
#
?
Jun 16, 2017 13:59
|
|
|
Molten lead will dissolve most metals, but one of the only ones it can't dissolve is, weirdly, iron. I always thought it was odd that they'd make things that get so hot out of a metal that oxidizes so readily but I guess it's the only one that won't get eaten by the lead. If you're sure you'll only be working with lead-free solder they make copper soldering iron tips and that would probably make the iron much less magnetic (since I'm guessing most of the 'full attraction' and 'slight attraction' in that picture is just it getting attracted to the tip)
|
|
#
?
Jun 16, 2017 17:21
|
|
|
ANIME AKBAR posted:Bingo! If you really want one, you could take a regular tip to any metal-working lathe-haver, and they should be able to turn a rod of whatever metal you want into that shape.
|
|
#
?
Jun 16, 2017 20:20
|
|
|
His Divine Shadow posted:Anyone have any experiences with the TS100 soldering iron, seen some youtubes that pimp it pretty hard. Also the Hakko 888 is high up on my list since I found it in 230V: That looks awesome, so I went and got one on eBay, $58 shipped from New Jersey and includes the proper barrel connector: http://www.ebay.com/itm/311837827808 Then I also went and got a 0-24v power bank for $46 shipped from China, which holds eight 18650 cells, so I can can see how that does with all the spare cells I have around: http://www.ebay.com/itm/182272692602 Edit: Here's some Amazon reviews of the power bank (called the QD-188 pr the QD 188-ALT), seems like the real deal, and one guy on YouTube even wired an extra eight 18650 cells in parallel: https://www.amazon.com/Portable-Emergency-Computer-External-Rechargeable/dp/B01EK2MDTY/ https://www.youtube.com/watch?v=q-kQOptcrrw Zero VGS fucked around with this message at 02:27 on Jun 17, 2017 |
|
#
?
Jun 17, 2017 02:07
|
|
|
MRIs are expensive enough that I wouldn't hack together something and then find out exactly how ferromagnetic it is. Also, you don't want to be the guy that got a soldering iron stuck to the magnet and had to have it shut down. Apart from cost/downtime, there's coolness factor to consider. One of the NMRs at my school's still on the same shot of current from the 70s. Vintage electricity's way better 
|
|
#
?
Jun 17, 2017 03:52
|
|
|
Anyone have recommendations for a good flashlight to mod into a battery holder/switch/handle assembly for a toy project? I was planning to cobble together separate parts, but I figure this is probably easier and faster for rev 1, to the extent that someone else might already have a favorite model to tear down. No other parts have been specced yet, so no hard constraints on voltage, etc.
|
|
#
?
Jun 17, 2017 04:23
|
|
|
KnifeWrench posted:Anyone have recommendations for a good flashlight to mod into a battery holder/switch/handle assembly for a toy project?
|
|
#
?
Jun 17, 2017 05:22
|
|
|
KnifeWrench posted:Anyone have recommendations for a good flashlight to mod into a battery holder/switch/handle assembly for a toy project? I was planning to cobble together separate parts, but I figure this is probably easier and faster for rev 1, to the extent that someone else might already have a favorite model to tear down. The Graflex 3-cell flash assembly has long been a go-to for toy projects. Getting hard to find anymore. For a real answer, I like the Harbor Freight 3xAAA flashlight. It's well enough made that it won't immediately fall apart but comes apart very easily, and there's usually a coupon for "free with purchase" of anything else. The one I got has a bare bones voltage regulator/constant-current power supply ahead of the LEDs, so it's easy to just scrape the LEDs off and have some reasonably stable power.
|
|
#
?
Jun 17, 2017 06:10
|
|
|
babyeatingpsychopath posted:The Graflex 3-cell flash assembly has long been a go-to for toy projects. Getting hard to find anymore. Fortunately for me, it's not that toy project. I told my son I wasn't paying $16 for this, but that we could build one together as a summer project, and thus be able to customize it and learn about circuitry, etc.
|
|
#
?
Jun 17, 2017 07:40
|
|
|
I'm looking for a switch that will toggle on/off, but will default back to off when power drops. Not sure what to call that for googling. Latching? I dunno. The actual application: I'm putting a pump for a sprayer on my mower, it'll run off the 12VDC supply from the mower, and when I turn it on, I want it to stay on indefinitely, until I either flip the switch off, or shut the mower down. I don't want it to come back on if I forget to shut it off when powering down the mower and then fire it up later. Also, since it's going on the mower, it needs to be somewhat robust and sealed against dust and water (not waterproof, just water resistant for hosing the whole thing down from time to time.)
|
|
#
?
Jun 17, 2017 16:31
|
|
|
Bad Munki posted:I'm looking for a switch that will toggle on/off, but will default back to off when power drops. Not sure what to call that for googling. Latching? I dunno. Does it have to be a toggle? Normally, you'd do something like this with a relay that gets bootstrapped on with a momentary button. You could similarly force it off with a second button. Edit: I guess the cleanest would be a switch to enable the bootstrap, and a button to initiate it. That way you could turn it off, and be confident it would stay off, but it wouldn't turn on without the button Edit 2: something like this:  Supplied without warranty; I just woke up KnifeWrench fucked around with this message at 17:14 on Jun 17, 2017 |
|
#
?
Jun 17, 2017 16:45
|
|
|
Bad Munki posted:I'm looking for a switch that will toggle on/off, but will default back to off when power drops. Not sure what to call that for googling. Latching? I dunno. Classic Relay problem. This is a start/stop station.
|
|
#
?
Jun 17, 2017 22:03
|
|
|
It's not that I'm looking for the circuit, that's pretty straight forward, I'm more looking for a pre-baked solution, I'm sure someone out there manufactures such a switch that meets this need, all buttoned up in a nice IP67 rated package. Just not sure what to call it to find it! 
|
|
#
?
Jun 17, 2017 22:12
|
|
|
Look for drill press or saw latching switches. I've seen those as add-on boxes.
|
|
#
?
Jun 18, 2017 14:24
|
|
|
Bad Munki posted:It's not that I'm looking for the circuit, that's pretty straight forward, I'm more looking for a pre-baked solution, I'm sure someone out there manufactures such a switch that meets this need, all buttoned up in a nice IP67 rated package. Just not sure what to call it to find it! They're pricy.
|
|
#
?
Jun 18, 2017 16:56
|
|
|
Ahahahaha yikes.
|
|
#
?
Jun 18, 2017 17:03
|
|
|
Today's episode of "it never even occurred to me that this would be attempted  " -- heat shrink cable joints for 11kV 3-core XLPE cable.https://www.youtube.com/watch?v=lpy1MvXP9tU It's utterly snore-inducingly boring, and yet interesting at the same time; over half of the 20 minute runtime is spent getting all the various layers of insulation and shielding off.
|
|
#
?
Jun 19, 2017 06:07
|
|
|
ullerrm posted:Today's episode of "it never even occurred to me that this would be attempted This is an accurate review. I watched the whole thing and don't regret it, but I do kind of wonder what got pushed out of my brain for this new knowledge that I'll never use.
|
|
#
?
Jun 19, 2017 08:26
|
|
|
ullerrm posted:Today's episode of "it never even occurred to me that this would be attempted Those kits are like $3000 each, too. To get certified as a HV splicer, you have to do one kit with an instructor, and two more in the field under supervision within six months. Ten grand in parts alone just to be allowed to connect wires together without someone else watching. There are also cold-shrink kits if you want to see something that's even more intricately demanding.
|
|
#
?
Jun 20, 2017 00:34
|
|
|
I have a pair of Argon 6340A active speakers for my PC which have some garbage "energy saving" feature where they'll shut themselves off if there's no sound input. The problem is the input detection is way too insensitive and even though the manual says they'll shut off after 20 minutes, they often shut off after ~30 seconds if the input volume drops too low, like during a quiet part of a song or game. It's pissing me off and I want it gone. I haven't cracked the speakers open yet, but there's an audible click when they turn off and on, so I'm assuming there's a mechanical relay that just cuts power to the amplifier. I figure if I can get to it, I can just bypass the relay to have the speakers active all the time. They still have a power switch in the rear to manually turn them off if needed. Alternatively I had the idea that I could run a USB lead from the computer to the speakers and use the power from the USB port to activate the relay so they'll stay active as long as the computer is on, but turn off when I turn the computer off. Without knowing the exact specs of the relay yet, do you think this would be a reasonable approach? Collateral Damage fucked around with this message at 11:12 on Jun 20, 2017 |
|
#
?
Jun 20, 2017 11:09
|
|
|
Collateral Damage posted:Alternatively I had the idea that I could run a USB lead from the computer to the speakers and use the power from the USB port to activate the relay so they'll stay active as long as the computer is on, but turn off when I turn the computer off. Without knowing the exact specs of the relay yet, do you think this would be a reasonable approach? This last part depends on the relay's driving current. If it switches on 5V at less than 500mA, this could work. If it were me, I'd just bypass the sucker.
|
|
#
?
Jun 20, 2017 13:33
|
|
|
Collateral Damage posted:I haven't cracked the speakers open yet, but there's an audible click when they turn off and on, so I'm assuming there's a mechanical relay that just cuts power to the amplifier. I figure if I can get to it, I can just bypass the relay to have the speakers active all the time. They still have a power switch in the rear to manually turn them off if needed. You have the switch in the back if you're concerned with draw while the computer is off. Using a relay as a wire is silly and wasteful. Speaking of which, if the click is audible, make sure you disconnect the coil while you're inside
|
|
#
?
Jun 20, 2017 14:38
|
|
|
Are you sure it's actually a relay and not just the "pop" that speakers make when they turn on or off anyway?
|
|
#
?
Jun 20, 2017 14:45
|
|
|
Nah, it's a pretty distinct click, not the pop you get when you turn a poorly made amplifier off.KnifeWrench posted:You have the switch in the back if you're concerned with draw while the computer is off. Using a relay as a wire is silly and wasteful. Collateral Damage fucked around with this message at 14:56 on Jun 20, 2017 |
|
#
?
Jun 20, 2017 14:54
|
|
|
I've got a simple project in mind that I'm thinking a Pi Zero will fit the bill. I need to download HTTP data from the internet, analyze it with a python script, and use it to display specific colors on three LEDs (red, green, or blue) depending on the data downloaded. I should be able to get three RGB LEDs and power them from GPIO ports (would use 9 IO ports in total), have a common ground, and resistors in series with each circuit. I'm thinking I'll use these LEDs https://www.digikey.com/product-detail/en/WP154A4SUREQBFZGC/754-1615-ND/3084119 Is there a better way to do this? Any issues with running three of the above LEDs at once as far as current/voltage for the GPIO ports? And if I am using red, green, or blue and no shades in between, I won't need to use PWM, correct?
|
|
#
?
Jun 24, 2017 23:56
|
|
|
Does anyone know where I can get affordable fans around 150mm diameter that can reverse direction? I'm trying to make a heat recovery ventilator like this dude: https://www.youtube.com/watch?v=pmtC4LkHUac There's some San-Ace fans on Mouser that reverse direction, but they are around $50 (!) each.
|
|
#
?
Jun 25, 2017 00:10
|
|
|
Mr. Bubbles posted:I've got a simple project in mind that I'm thinking a Pi Zero will fit the bill. I need to download HTTP data from the internet, analyze it with a python script, and use it to display specific colors on three LEDs (red, green, or blue) depending on the data downloaded. 2 of your 3 colors have a Vf equal to your supply voltage (3.3v if my quick check is correct). This means your current limiting resistors will be difficult to set reliably and your LEDs may not illuminate at all. In general, it's preferable to use GPIOs to sink LED current or toggle an external driver, rather than source the current directly. I haven't looked at the specs on the Zero, so I can't say if this is specifically a bad idea in your case. If I'm reading what you're saying correctly, though, you plan to wire multiple outputs to each other to increase your current output, which IS explicitly a bad idea. You are correct that you shouldn't need PWM, though.
|
|
#
?
Jun 25, 2017 00:26
|
|
|
KnifeWrench posted:2 of your 3 colors have a Vf equal to your supply voltage (3.3v if my quick check is correct). This means your current limiting resistors will be difficult to set reliably and your LEDs may not illuminate at all. Could you recommend me a driver that would be suitable for this use and that I could control with the 3.3V GPIO pins? Or if not using a driver and instead sinking the LED current, do you mean using the 5V RPi rail as the + voltage for all of the LEDs and toggling the GPIO pins as ground? After a quick search, looks like 16mA is the max current recommended per pin to sink with, and the LEDs above are tested at 20mA. I guess I could bring it down to lower current with a resistor, but would light still be visible? Sorry for my naivety.
|
|
#
?
Jun 25, 2017 00:53
|
|
|
Mr. Bubbles posted:I've got a simple project in mind that I'm thinking a Pi Zero will fit the bill. I need to download HTTP data from the internet, analyze it with a python script, and use it to display specific colors on three LEDs (red, green, or blue) depending on the data downloaded. How complex is the data, and does the program have to be Python? This sounds like something that could be easily achieved with an ESP8266 board like the NodeMCU or WeMos D1, and three WS2812 NeoPixels. You'd have to write Arduino C but the electrical complexity is nil (plug in three wires) and the total cost would be like 10 bucks. https://www.amazon.com/Makerfocus-NodeMcu-Development-ESP8266-ESP-12F/dp/B01N3P763C/ https://www.amazon.com/HiLetgo-5pcs-8-WS2812-Driver-Development/dp/B014F5VLN0/ Sagebrush fucked around with this message at 01:21 on Jun 25, 2017 |
|
#
?
Jun 25, 2017 01:19
|
|
|
So I cracked the speakers open to have a look. This is the main transformer. Power comes in on the left, the power switch is on the right with an inline fuse after the switch.  This is the relay board. There's a relay in the top left, but the back of the board is all epoxied over so I can't just short it out. So my new plan is to just bypass the board completely. The 4-pin connector at the bottom goes to the signal board. The 2 pin connector goes to the power LED, but I had to disconnect it to get the board out of the case.  I'm thinking this board doesn't do 220V rectifying, right? So I'm not sending AC into a tranformer that expects DC by shorting over the two connectors? The small transformer on the relay board outputs 11V AC and I assume it does low voltage rectifying to feed the input detection and power LED. Collateral Damage fucked around with this message at 18:26 on Jun 25, 2017 |
|
#
?
Jun 25, 2017 14:33
|
|
|
Mr. Bubbles posted:Could you recommend me a driver that would be suitable for this use and that I could control with the 3.3V GPIO pins? First off, don't apologize for naivety unless you're unwilling to admit it/ask questions. If I came off as terse or impatient, I apologize; it's just because I'm on my phone. The quickest way to test whether the light would be visible at a lower current is to buy your parts and try it out. Datasheets will list millicandela ratings, but ultimately your ambient conditions and personal perception matter the most. 75% sounds like plenty to me, but if you want to be safe... ...A driver circuit can be as simple as a BJT transistor. Blinking an LED is one of the most common introductory projects for learning electronics, so lots of resources should exist if you just search for "BJT LED driver". Here's one from the first page: http://www.learningaboutelectronics.com/Articles/LED-driver-circuit.php Just be aware that exact values will differ both from this example, and between the discrete diodes of your RGB part. And the cool part is, once you get this working, if you decide you want to figure out how PWM works, you can make any color you want!
|
|
#
?
Jun 25, 2017 16:32
|
|
|
The 4-pin connector was glued in place so I just cut it off. Now the speakers are always on unless I decide to switch them off. Only drawback is the power LED doesn't work any more, but I can live with that. Screw your energy saving regulations. We have nuclear. 
|
|
#
?
Jun 25, 2017 18:26
|
|
|
I've got a 12V car battery on a solar panel, and also a "20V" (18V) li-ion battery. I want to charge the 20V battery using the 12V battery. I can use an inverter to run the 20V battery charger, but this seems inefficient. Is there a way to use a constant current (or constant voltage?) power supply and/or buck converter to charge a battery?
|
|
#
?
Jun 26, 2017 06:16
|
|
|
|
| # ? May 28, 2024 18:35 |
|
|
So my "scroll wheels in a box" project has evolved into trying to make a simpler, cheaper version of this overpriced bullshit: https://palettegear.com/ Yeah motherfucker that's $50 for a goddamn single rotary encoder Anyway to start it's probably still going to be built into the same box, but I'm trying to come up with a protocol that will allow arbitrary nodes to be added later (probably just via a cable, not the fancy magnetic connectors these things have). What I've come up with so far is I2C, with a trimmed down version of SMBus's address resolution protocol (since I don't need most of the "device ID / vendor ID / etc" stuff). I don't think it needs to be hot-swappable or anything but I do want it to be able to set up the bus on its own without DIP switch fuckery. Am I on the right path or should I look into a different protocol? e: Additional context: right now each "node" other than the master is going to be an ATTINY20 or a PIC16, I bought a handful of both to see which works better Shame Boy fucked around with this message at 14:43 on Jun 26, 2017 |
|
#
?
Jun 26, 2017 14:40
|
|