|
Splode posted:well you could replace them with diodes that don't emit light? Thanks for being my rubber duck. When you put it that way I gave the circuit a bit more scrutiny. All it would achieve is putting a 1.1k resistance from the positive half of the AC to ground. Although the one I built previously has no hum, I don't imagine the diode and resistor would be beneficial. It might cause assymetric pulses in the ground plane. Sagebrush posted:Some sort of non-emissive light-emitting diode, you say? If only science could concieve of such a thing! But yeah, having the LEDs lighting up the tubes from underneath is kind of tacky. I know it's a dirt cheap kit, but I like building kits. The other one is mounted to the top of my Raspberry Pi 3 with the DAC board anyway, so it'd be nice to have a free range one. I saw cheap acrylic cases on Aliexpress for the preamps are available now. I really don't like acryllic cases but it's better than nothing. General_Failure fucked around with this message at 02:49 on Oct 10, 2019 |
#
?
Oct 10, 2019 02:45
|
|
|
|
| # ? May 9, 2024 07:45 |
|
|
Splode posted:well you could replace them with diodes that don't emit light? Technically all diodes emit light.
|
|
#
?
Oct 10, 2019 02:47
|
|
|
So a while back I mentioned I wanted to make a little circuit with starting off an arduino nano to get back into things and a cheap OLED module from aliexpress. the goal is to push a button which wakes up an ESP8266 and lists an average price of recently sold ebay listings of a particular item. Funnily enough the OLED I got was mislabeled SCK and not SCL. That was a bit of a head scratcher before ignoring and assuming c'hinesium' and going with I2C wiring. Someone mentioned there was an API for this sort of thing already out there, just having trouble finding it.
|
|
#
?
Oct 10, 2019 03:47
|
|
|
SCK and SCL are both abbreviations for serial clock. The other one should be SDA, for serial data. If it were SPI, you should be seeing MISO/MOSI/CS, (master in slave out/master out slave in/chip select) but you might also see SDI/SDO/CS (serial data in/serial data out) on some boards. Basically just don't confuse SCK and SCL and SDA and SDI and SDO and SPI and you'll be fine. Simple
|
|
#
?
Oct 10, 2019 03:58
|
|
|
Yeah nobody can agree on how to abbreviate clock
|
|
#
?
Oct 10, 2019 04:02
|
|
|
Humphreys posted:
Look for SSD1306
|
|
#
?
Oct 10, 2019 04:36
|
|
|
General_Failure posted:I can be helpful! Thanks, I got that bit going with adafruits library and oled_i2c. I was meaning the API for the ebay data scraping?
|
|
#
?
Oct 10, 2019 05:35
|
|
|
Ohhhh. Sorry. Yeah I've not scraped eBay before. Just other things I can't seem to recall. Sorry. Is it possible to set up an RSS feed of sales on eBay? Just a thought anyway. Never tried that.
|
|
#
?
Oct 10, 2019 06:12
|
|
|
General_Failure posted:Ohhhh. Sorry. Yeah I've not scraped eBay before. Just other things I can't seem to recall. Sorry. Is it possible to set up an RSS feed of sales on eBay? Just a thought anyway. Never tried that. You can save a search on ebay and it will periodically email you with new items that come from that search.
|
|
#
?
Oct 10, 2019 06:17
|
|
|
Cojawfee posted:You can save a search on ebay and it will periodically email you with new items that come from that search. They used to do the same with RSS feeds too. That's why I mentioned it. Don't know if they still do. It was useful. It'd be pretty simple to parse the RSS feed too, and a bit closer to realtime.
|
|
#
?
Oct 10, 2019 06:42
|
|
|
Cojawfee posted:You can save a search on ebay and it will periodically email you with new items that come from that search. I have that setup already for other stuff. Unfortunately not as realtime or 'on demand' that I'm chasing. But to be honest, I haven't tried with SOLD listings - will look at it. General_Failure posted:They used to do the same with RSS feeds too. That's why I mentioned it. Don't know if they still do. It was useful. It'd be pretty simple to parse the RSS feed too, and a bit closer to realtime. Hmmm haven't played with RSS feeds in quite some time. Another thing to check out. Over the weekend I'll have a read up on this: https://developer.ebay.com/devzone/finding/callref/findCompletedItems.html
|
|
#
?
Oct 10, 2019 08:02
|
|
|
Zero VGS posted:I'm well stocked on consumables, connectors might be a good idea.
|
|
#
?
Oct 10, 2019 11:52
|
|
|
General_Failure posted:But yeah, having the LEDs lighting up the tubes from underneath is kind of tacky. Yeah I've always wondered why people do it to tube amps, I assume it got cargo-culted from nixie clocks at some point, where the LED provides a baseline amount of energy in a dark room to allow the ionization to start consistently when rapidly multiplexing them on and off. Even then though I think it looks tacky and is probably not necessary unless you plan on running your nixie tubes in pitch black conditions all the time... I think my favorite case of this was in a friend's weird AV receiver, they put a couple tubes up front in a window so you could see them to establish how High End it was, except when I looked closely at them it was clear that their filament wasn't even running and the glow was coming from an orange LED under each one 
|
|
#
?
Oct 10, 2019 14:48
|
|
|
Shame Boy posted:Yeah I've always wondered why people do it to tube amps, I assume it got cargo-culted from nixie clocks at some point, where the LED provides a baseline amount of energy in a dark room to allow the ionization to start consistently when rapidly multiplexing them on and off. Even then though I think it looks tacky and is probably not necessary unless you plan on running your nixie tubes in pitch black conditions all the time... Because actually powering the tubes means you have to replace them when they burn out. Easier to stick a couple cheap tubes behind plexi to hide the fact that you're using a T amp.
|
|
#
?
Oct 10, 2019 17:21
|
|
|
I finished the preamp sans LEDs and driver resistors. The hard part was finding where I'd put the power supply I use because I'd rearranged a couple of times in the past year. I thought I'd found it initially but it sounded terrible. I'd managed to grab a 12V DC instead of the 12V AC one. Once I found the right one and tried it, it was all good. I was very relieved, mostly because one of the tubes looked like it had been pulled roughly and had bent pins. This preamp uses 6J2s instead of 6J1s. Not a drat clue what the difference is. My quick test was using my phone as a source and a pair of headphones. Like the first one I made there's definitely an audio quality improvement, at least perceived. And no drat blue LEDs. Much better. I think I said but my original is attached to the top of a Pi3 with a DAC board. It's the best quality sound my crappy ears can appreciate. Of course that's used with headphones too. The output is good enough to drive them thankfully.
|
|
#
?
Oct 12, 2019 07:36
|
|
|
Dawncloack posted:Thanks a bunch! I'll also get the book now. For many opamps with very low bias currents or well matched bias currents it can be left out or should be left out. You want it when you care a lot about offset and the opamp has larger bias currents than offset currents. Or to take it to the next step you'd want to fully match the impedance, not just the resistance of the two opamp inputs. So matching resistors and capacitor.
|
|
#
?
Oct 13, 2019 15:13
|
|
|
edit: wrong thread
Zero VGS fucked around with this message at 20:49 on Oct 22, 2019 |
|
#
?
Oct 14, 2019 03:49
|
|
|
sensor Q for a job i'm pitching in on: any thoughts on the sorts of sensors/sensor configurations that could be used to verify, through multiple/redundant sensing means, that a backup self-priming bilge pump in a ship running an automated maintenance cycle (just operating for a couple of minutes once per week/month/etc) is operating correctly? So far I've got - voltmeters/ammeters measuring the pump power draw - flowmeter in pump outlet - pressure sensor in pump outlet - float sensor in bilge, to check fill level - laser/ultrasonic rangefinder in bilge, as nonmechanical backup for float - column of moisture-sensitive conductive sensors in bilge, again as bilge redundancy - moisture sensors at overflow/topping-over point in bilge, for a final "bilge pumps are totally overwhelmed" alert - possibly a magnetic rotary encoder that tracks pump impeller RPMs/rotation status, if that's even possible to set up without replacing/overhauling the entire pump - MAYBE induction/capacitive non-contact proximity sensors??, if they permit reliable detection of a pipe being filled with water vs. full of air directly through a PEX/PVC pipe sidewall baselines for sensors more complicated than true/false checks will prolly be established with a 'teach' mode where the pump is operated as normal for an extended time while the sensors gather data and establish averages and ranges. also, despite the context we don't have to worry about marine suitability or maritime regulations or anything like that, nobody's safety or well-being will actually depend on this device, and we're casting a wide net before we narrow the possibilities down, so any and all weird ideas in this regard are welcome Ambrose Burnside fucked around with this message at 04:27 on Oct 14, 2019 |
|
#
?
Oct 14, 2019 04:21
|
|
|
You could do what every septic transfer or sump setup does and have three float sensors taped to a pvc pipe. Low for cutoff, pump start, and high level. The control starts the pump when the middle sensor makes, shuts it off when the lower one breaks. The high level is attached to a screamer or alarm of some kind. usually these sensors are simple floating orientation sensors, with a ball bearing and two contacts, when they are suspended in a fluid they go vertical and the contacts make. If you really get fancy you set a run timer if the pump starts but doesn't drain the sump in x minutes it also trips the alarm.
|
|
#
?
Oct 14, 2019 04:38
|
|
|
Client wants significantly more data/feedback than just float sensors can provide, particularly direct data about pump operation independent of the bilge; off-the-shelf float sensors will form the core of the system b/c yeah they're simple and reliable and already used everywhere, but they can't provide all of the required feedback. It's also a configurable product built to the future customer's spec, so having an overabundance of sensor options is preferable to help cover unanticipated edge cases or weird implementations (or just selling more poo poo). This will get dramatically simplified and all those sensors would never be used all at once, but this is the "throw poo poo at the wall and see what sticks" phase, so 
Ambrose Burnside fucked around with this message at 05:11 on Oct 14, 2019 |
|
#
?
Oct 14, 2019 05:08
|
|
|
edit: wrong thread
Zero VGS fucked around with this message at 06:17 on Oct 14, 2019 |
|
#
?
Oct 14, 2019 05:45
|
|
|
Ambrose Burnside posted:Bilge stuff I'd probably code in some sensors together to let someone know if the bilge pump is clogged. Combine the pump power draw sensor with the flow meter. If it's pumping and there's no/little flow, then either the pump is clogged or the bilge is empty. Add a moisture sensor to eliminate that last condition and shut the pump off if there's nothing to pump to save wear and tear on its motor. You know, bilge pumps and their automatic floats are available for small boats at any boating supply store for fairly cheap, if you're the type that likes to fiddle with examples to get an idea of what to build. Edit: you might also want to look into sending units for cars as well. That's the float inside the fuel tank, designed to operate with electricity next to fuel vapors and not explode. Yes, that's a thing to worry about with bilges too. Oil and its derivatives float, while bilges are the lowest point in a hull and are usually almost completely enclosed. All those vapors get trapped. Get the fume to air ratio just right then add a spark... You might want to include an exhaust fan on a timer as well, along with a draw sensor and alarm. Edit2: you could do other fun stuff like multiply the flow rating of the pump by how long it's been switched on to get an idea of how much water the boat took on. It could let the owner know to look for a leak. kid sinister fucked around with this message at 16:58 on Oct 14, 2019 |
|
#
?
Oct 14, 2019 16:33
|
|
|
I don't know anything about tank stuff, but I had a teacher in college that was a chemical sensor engineer before he left to teach and I remember him saying that ultrasonics were the standard fluid level sensor they used pretty much everywhere, especially if it was for something remotely hazodous. That was about 20 years ago though.
|
|
#
?
Oct 14, 2019 18:46
|
|
|
Nuclear non contact coolant sensors use temperature probes, and idea you could borrow if there is enough dT between the ocean and vessel at the sensing location. Strain gauges at known load bearing locations calibrated to test fills, also non contact and very reliable. Bernoulli flow gauges, contact but no moving parts. Basically limit as much poo poo that moves as possible for higher reliability.
|
|
#
?
Oct 14, 2019 19:18
|
|
|
Crossposting from the physics thread. I want to make a vortex mixer for my miniature / model tank paints. Basically what I want to make is a small platform where I place my paint bottle, and then an RC car motor (or something similar) connected to an Arduino rotates it at x RPM for a minute, hopefully creating a vortex inside the paint bottle and mixing up the paint good and proper. I know this is a thing, because a lot of places are selling vortex mixers, but the prices are quite high, and the actual tech seems like a fun DIY project for a lot less money. While I'm fairly decent with electronics, I have little experience with rotating motors and basically have no idea what I should be looking for as far as RPM and torque go. I've been looking at small DC brush motors available at my local electronic shops (much like this one: https://product.mabuchi-motor.com/detail.html?id=31), but is this anything like what I'd want?
|
|
#
?
Oct 14, 2019 20:41
|
|
|
What RPM do you think you want to run at? Google for possibilities if you don't know. What size of paint bottles? If they're anything like the old tiny Warhammer paints I remember from the 90s, then torque won't matter so much, pretty much anything will be able to spin something the size of a film canister. Figuring out appropriate motor sizes on paper is really challenging, you'd be better served Just Building It and then adjusting afterwards as necessary
|
|
#
?
Oct 14, 2019 20:56
|
|
|
Get a geared motor that runs at the RPM you want so it's as simple as an on/off switch or a 555 timer and a mosfet. An RC motor is wildly more power than you'd need. https://www.servocity.com/motors-actuators/gear-motors/standard-duty-gear-motors
|
|
#
?
Oct 14, 2019 20:58
|
|
|
Der Shovel posted:vortex mixer if you're referring to the type of vortex mixers used in labs to mix test tubes, they don't actually "spin" but are fixed in one direction and move around the outside of circle rapidly, if you just spin the bottle really fast you run the risk of all of the paint slamming up against one of the sides without actually mixing. I've seen these go on ebay for about 40 bucks before, so something to consider once you work out how much all of the mechanicals will cost.
|
|
#
?
Oct 14, 2019 21:07
|
|
|
kid sinister posted:I'd probably code in some sensors together to let someone know if the bilge pump is clogged. Combine the pump power draw sensor with the flow meter. If it's pumping and there's no/little flow, then either the pump is clogged or the bilge is empty. Add a moisture sensor to eliminate that last condition and shut the pump off if there's nothing to pump to save wear and tear on its motor. Yeah, combining sensor data with a scheme like to give the user much more system feedback than they could provide individually is the general idea. That's the other reason the client wants a diverse sensor spread involving both bilge/plumbing and pump, it'd extend the system's feedback net to watch more of the bilge-draining bits overall while also providing more specific info about fault states. And yeah, we're def interested in extended applications of the feedback beyond just maintenance and emergencies, leaks are one example. We're also considering stuff like hydrocarbon ratio sensors coupled w GPS for geolocation data, so bilge pump operation can be restricted based on location and maritime law and the bilge contents (i.e. there are often oil contaminant PPM limits for water discharged within X country's coastal territories), a mini weather station aboveboard to detect a possible correlation between bilge pump operation and bad weather, etc kid sinister posted:You know, bilge pumps and their automatic floats are available for small boats at any boating supply store for fairly cheap, if you're the type that likes to fiddle with examples to get an idea of what to build. I believe this side of things is already planned for, but it's not my problem; actual engineers fully handle the more directly safety-concerned stuff that's largely dictated by maritime regs. It's nice to only have to do the fun half of product design with someone else handling the math and the parts selection and the due diligence. um excuse me posted:Nuclear non contact coolant sensors use temperature probes, and idea you could borrow if there is enough dT between the ocean and vessel at the sensing location. Strain gauges at known load bearing locations calibrated to test fills, also non contact and very reliable. Bernoulli flow gauges, contact but no moving parts. Basically limit as much poo poo that moves as possible for higher reliability. Temperature sensors are smart, thanks- we can def assume a significant differential at at least one or two locations, or include a small power resistor in the module design to help stabilize temps and exaggerate the differential. We're aiming for limiting moving parts for obvious reasons, yeah- altho I didn't know there were flowmeters w no moving parts, hopefully we can source some for "cheaper than the nuclear industry" prices.
|
|
#
?
Oct 14, 2019 22:31
|
|
|
I'm pretty sure that untreated machinery space bilges aren't something you can discharge regardless of if you're doing it through an oil content meter or not.
|
|
#
?
Oct 14, 2019 22:52
|
|
|
No offense but I'm a marine engineer and every aspect of this project gives me cost, regulatory, and maintenance nightmares. If this is for a large fully unmanned ship your client needs to be looking much more at ground up reconceptualization of their plant design than a gigantic mess of wiring.
|
|
#
?
Oct 14, 2019 22:56
|
|
|
Der Shovel posted:Crossposting from the physics thread. Have you considered just spending 40 bucks on a magnetic stir plate and a bunch of stir bars? Unless you really just want to make something.
|
|
#
?
Oct 14, 2019 23:01
|
|
|
um excuse me posted:Basically limit as much poo poo that moves as possible for higher reliability. This right here. Bilges are filthy places. It's not so much wear that will be the failure of these parts. It's all the crap in the water that will jam poo poo up. That's why I was talking about coding sensors together to check for clogs.
|
|
#
?
Oct 15, 2019 00:37
|
|
|
kid sinister posted:This right here. Bilges are filthy places. It's not so much wear that will be the failure of these parts. It's all the crap in the water that will jam poo poo up. That's why I was talking about coding sensors together to check for clogs. Vibration, heat (commonly 50C+, infinite salt/SOx/humidity), even the air is filthy with oil and particulates, the requirement to build to an incredibly tight price point, shoddy shipyard wiring and installation, 400 coats of paint, etc. The fact that sometimes your bilges flood with fuel (don't forget it's solid at temps up to 38C but will be 100C+ when it hits the bilge on a bad day) and not water. Or acid (used to clean everything).
|
|
#
?
Oct 15, 2019 01:05
|
|
|
Ambrose Burnside posted:hopefully we can source some for "cheaper than the nuclear industry" prices. Oh you wont even need to come close to that kind of certified sensor. They use temperature probes because they are screwed into sealed testing ports welded into the coolant lines so there is zero possibility for a system leak.
|
|
#
?
Oct 15, 2019 01:19
|
|
|
You can also use the aircraft safety approach and use triple redundancy. If one sensor goes bad, you'll know because it'll disagree with the other two, allowing it's data to be discarded until you replace it. Obviously this approach is best suited to cheap and easily replaceable sensors, but it does let you get away with less reliable parts
|
|
#
?
Oct 15, 2019 02:26
|
|
|
Is this for Boeing?
|
|
#
?
Oct 15, 2019 02:44
|
|
|
Some version of these get used in gasoline storage tanks. http://www.mtssensors.com/products/liquid-level-transmitters/index.html
|
|
#
?
Oct 15, 2019 02:54
|
|
|
Der Shovel posted:Crossposting from the physics thread. Imagine holding the top of a tube still with your hand and tracing a circle with the bottom. That's the motion a vortex mixer makes. If you spin the tube around its central axis, you've built an axial centrifuge instead and it'll separate the contents instead of mixing (densest stuff out on the walls of the tube). You won't need more than a few thousand RPM, but will probably want that to be adjustable, so design your electronics accordingly. To figure out what a torque will do to the bottle, you need the moment of inertia for the axis you're rotating around. This will be yucky because the axis of rotation isn't aligned with the bottle's axis. Overestimate it by approximating the bottle as a point mass that is at the bottom of the tube so that it traces the full circle of the mixer. That will end up overspecing the motor. Moment of inertia for a point mass going around a circle is mr^2 where m is the mass and r is the radius of the circle example: - Suppose you have a 40g paint bottle and your mixer is moving the bottom around a 50mm radius circle - Overestimate of moment of inertial is 40g*50mm*50mm = 1E-4 m^2 kg - The motor you linked has a stall torque of 8.71E-3 N m - (8.71E-3 N m) / (1E-4 m^2 kg) = 87 rad/s^2 = 5000 deg/s^2 = 833 rpm/s^2 so it will start off accelerating pretty fast. Motor torque will drop as RPM increases until eventually it reaches 0 at the no load RPM (actually less than that since you'll have drag from the bottle and losing energy in sloshing)
|
|
#
?
Oct 15, 2019 05:20
|
|
|
|
| # ? May 9, 2024 07:45 |
|
|
Hmm, does anyone have any thumbwheel encoders that are reliably sourceable? This is pretty depressing: https://www.digikey.ca/products/en/...y=1&pageSize=25 Some of those are available through Alibaba, but for how long?
|
|
#
?
Oct 15, 2019 11:01
|
|