Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
Shame Boy
Mar 2, 2010

babyeatingpsychopath posted:

What do you think is protecting the service wire at your house transformer from the 38KV @ "functionally unlimited" current? It's not a circuit breaker. Just a boring 19th-century technology.

Folks, selective coordination of circuit protective devices is an entire field of electrical engineering, is well-studied, and is very safe.

It's a large, special fuse that blows in a very bangy way that physically pushes it away from the live connection, I have seen one blow in person and was very glad it was up on a pole and not right next to me.

In between it and anything that can only take 10A is going to be a few circuit breakers and additional fuses. I'm not saying fuses aren't good, I'm saying the safety of power distribution relies on layering.

Adbot
ADBOT LOVES YOU

Zero VGS
Aug 16, 2002
ASK ME ABOUT HOW HUMAN LIVES THAT MADE VIDEO GAME CONTROLLERS ARE WORTH MORE
Lipstick Apathy

Marsupial Ape posted:

Why would the batteries in my BMS not be discharging equally? I have 3 16850 batteries in series wired to a small 3 cell BMS board. The positive side (left most) battery had the greatest drain with the other two consecutively higher. All batteries are the same model and amp hours and brand new. Charged with a 12.6 volt charger meant for 3 cell battery configuration. Is the BMS board bad? Should I try rewiring it?

3 Cell Lithium BMS start at.... $1.37 with free shipping on eBay, that's right a buck thirty-seven shipped: https://www.ebay.com/itm/402851424750

Might try that?

Marsupial Ape
Dec 15, 2020
the mod team violated the sancity of my avatar

Zero VGS posted:

3 Cell Lithium BMS start at.... $1.37 with free shipping on eBay, that's right a buck thirty-seven shipped: https://www.ebay.com/itm/402851424750

Might try that?

That’s pretty much what I am using. I’m just going to rebuild it from scratch. Thanks, though

longview
Dec 25, 2006

heh.

babyeatingpsychopath posted:

There are current-limiting 10A fuses that are capable of interrupting 20,000A at DC from 12-1500VDC. They're common industrial parts. They're not even expensive in an industrial sense. Large-scale DC PV arrays have made these things common.

Fuses are really impressive devices. They interrupt current by burning a wire in a bucket of sand. Safe, effective, not at all scary.

What do you think is protecting the service wire at your house transformer from the 38KV @ "functionally unlimited" current? It's not a circuit breaker. Just a boring 19th-century technology.

Folks, selective coordination of circuit protective devices is an entire field of electrical engineering, is well-studied, and is very safe.

Of course it's possible, controlling huge amounts of power relatively safely is a fundamental part of modern society.
But is it possible isn't the question here, the question was: should we hack up some ridiculous monstrosity of a power supply to act as a cheap battery charger to save the company a few bucks?
The only sensible answer to that question is no, battery chargers that run off mains are not that expensive and are nominally designed to be safe to use.
Further the way I interpreted the original question was "should we just start hooking up batteries directly to this power supply through some fuses and maybe a resistor to charge them?" In which case the answer is oh my god no.

Since you brought it up, how is the current limiting part of these 10 A fuses you mentioned performed? Standard melty fuses aren't generally known for their speed but I suppose the thermal-resistance properties of the metal would limit current somewhat.
Common automatic circuit breakers AFAIK do not limit current all that much except by virtue of that solenoid-like current-derivative detector (as opposed to the bimetallic strip thermal detector) which can act very quickly to limit fault current before the current can rise to crazy levels.

Acid Reflux
Oct 18, 2004

Just to add a little bit of detail - these battery packs are (per the manufacturer's maintenance manual) designed to be charged directly from a DC power supply set to a specific constant voltage and current limited, and not with a "proper" battery charger. Deviating from the prescribed procedure means that the units aren't being serviced in accordance with the manual anymore, which is a violation of numerous FAA regulations.

People above me see the BFPSU that's not being used anymore and ask me if we can do this thing with it. Very skeptical me decides to ask the people with more detailed knowledge about the thing, because this is power that's quite literally two orders of magnitude higher than I normally deal with and I don't think it's a very good idea. Goons say " oh gently caress no", completely confirming my initial hesitation, and I move on to making them buy me some much more reasonable bench supplies so as to not die a horrible flaming explosive death.

ANIME AKBAR
Jan 25, 2007

afu~

Dr. Despair posted:

https://www.amazon.com/KORAD-KD3005...31072489&sr=8-3

This Korad is what I was looking at a bit ok... it has the distinction of being the cheapest brand of benchtop power supplies that digikey is willing to sell (which means you can go through them if you don't want to pay jeffy b). I wound up buying a used Kikusui off ebay instead though so I can't speak too much to it's quality... and it's slightly more than your price range anyways.
I highly recommend Korad supplies. My workplace had one of them when I joined (the KA6005D I think), along with a couple of the more typical triple output bench supplies. First time I used it, I hoarded to my bench and ordered two more.

The fourth digit on the current display is actually quite accurate, even near zero. The only other PSU I've used which can claim that is a Keysight. The UI is pretty good, the ones with two knobs for per output are very easy to use. All of them have an output on/off button so you can adjust your settings before enabling the outputs. Haven't accidentally blown up any loads by fatfingering the buttons yet.

I have one KD3305P (two variable outputs, one fixed). Haven't used it much but it seems to have the same quality as the single output models. Oh and the "fixed" output can be set to 5V/3.3V/2.5V, which is nice. Only issue I've had with it so far is that the "fixed" output's current limit isn't adjustable, and I don't think there's a way to display its actual output current, so you have to be careful using it.

My favorite is the KWR103. Similar form factor to the basic units, but its output range is 0-60V and 0-15A. Overall output is limited to 300W, but that still makes it versatile enough for 90% of my use cases. And it's still around $300 USD.

Overall they're by far the best budget lab PSUs I've ever used. I think my lab is up to six total, so far none have failed, but one has a very noisy fan. Wish they made a high voltage model.

edit: Oh and I bought all of mine from digikey. I would not be surprised if there were knockoffs floating around on amazon and ebay, so beware.

babyeatingpsychopath
Oct 28, 2000
Forum Veteran


longview posted:

Of course it's possible, controlling huge amounts of power relatively safely is a fundamental part of modern society.
But is it possible isn't the question here, the question was: should we hack up some ridiculous monstrosity of a power supply to act as a cheap battery charger to save the company a few bucks?
The only sensible answer to that question is no, battery chargers that run off mains are not that expensive and are nominally designed to be safe to use.
Further the way I interpreted the original question was "should we just start hooking up batteries directly to this power supply through some fuses and maybe a resistor to charge them?" In which case the answer is oh my god no.

Since you brought it up, how is the current limiting part of these 10 A fuses you mentioned performed? Standard melty fuses aren't generally known for their speed but I suppose the thermal-resistance properties of the metal would limit current somewhat.
Common automatic circuit breakers AFAIK do not limit current all that much except by virtue of that solenoid-like current-derivative detector (as opposed to the bimetallic strip thermal detector) which can act very quickly to limit fault current before the current can rise to crazy levels.

Ahh, I interpreted the question as: Given a 1000A 0-40VDC bus, can we connect current-limited outputs to it to charge batteries? I then took "current-limited outputs" to mean some kind of smart charge controller.

Current limiting in fuses: Standard slow-blow fuses are just a bit of calibrated wire in a tube. Glass is common. Higher-current versions use ceramic to prevent UV (or x-ray) from coming out. They limit dI/dT really well, because ultra-fast rise means ultra-fast heating and the wire burns quickly. Big current-limiting fuses are strips of metal with holes in them, the whole assembly encased in sand. When the short circuit happens, the holes make narrow segments in the current path that heat geometrically faster. When one segment burns through, the resulting plasma has a higher resistance, which means the other segments flow more current, and burn faster. The sand keeps the arc from bridging and the plasma from escaping the fuse.

Current-limiting power fuses will clear an AC short circuit event WELL under 1/4 cycle (before peak current is reached). Even good breakers in their instantanous-trip region (10x-20x rating current) only trip after 1/2 cycle. A 10A class RK1 fuse will clear a >200A "available fault current" in <1/4 cycle with a max let-through current of 100A, and is capable of interrupting up to 200,000-300,000A of available current @600V. Telecom and other specifically-developed low-voltage DC fuses (photovoltaic, DC vehicular, etc) can be better, but their characteristic time curves are becoming annoying to look up.

My application for this would be: Charger unit feeds some bus bars. The bus bars have a fuse holder on each leg supplying smart chargers. These fuse holders are rated at the full current available (>1000A @ 40VDC) and have suitably rated current-limiting fuses installed (rating based on the charger). The wire coming off of the fuse blocks goes to DC-DC battery chargers, who now don't know that there's practically unlimited current. They'll have a nameplate fusing recommended; using that fuse means the charger will never see anything it's not rated and listed for, and sizing the wiring based on that fuse means that the wire will never see current beyond its rating and listing.

The Big Scary Charger has its own input and output protection built in; we're not messing with that. The smart chargers have their own input and output protection; we're not messing with that.

Laopooh
Jul 15, 2000

Hi nerds! First time in this thread I think. So playing the game Shenzhen I/O got me on a programming and electronics kick and I decided to try to make a dog treat dispensing robot and it kinda works! 😀🦴 🐶

When a doggo gets within 50cm it lowers the treat tray, and then pulls it back up after snacking is complete. I added a safe/arm switch, an LCD readout, lights, and a cute animated display for fun and to learn. Next up to add is a manual control override with a joystick.

(It doesn't pull back up all the way in the vid because I forgot to account for the weight of two uneaten Milk Bones on the retrieval 🦴🦴💦)


https://youtube.com/watch?v=s0EuD2UsAOo
(not embedding for me, ?)










Anyway, button timing and coding to get clean input toggles, and nested loops is throwing me for a curve! I'm good at googling and kludging stuff together but is there a succinct lesson to teach me clean ways to have multiple while/if loops monitoring buttons or whatever even while the main loop is executing? Without delaying the loop or hanging on waiting for an input? I don't have much formal coding or electronics training.

Laopooh fucked around with this message at 22:16 on Sep 12, 2021

PDP-1
Oct 12, 2004

It's a beautiful day in the neighborhood.

Laopooh posted:

Anyway, button timing and coding to get clean input toggles, and nested loops is throwing me for a curve! I'm good at googling and kludging stuff together but is there a succinct lesson to teach me clean ways to have multiple while/if loops monitoring buttons or whatever even while the main loop is executing? Without delaying the loop or hanging on waiting for an input? I don't have much formal coding or electronics training.

Looks like a fun project! One way to avoid hanging up the main loop while still doing a set of steps is called a Finite State Machine (FSM) where you have a finite set of states where you can only be in one at a time. For whichever state you happen to be in at the moment you can either do a few simple tasks or make a decision to transition to another state. When you transition to another state you might also want to do a few simple tasks but these only execute once per transition. Typically you'd list out all of your states as an enumerated type and choose the active one with a switch statement something like in this pseudocode:

code:
enum State {waitingForDoggo, loweringBasket, doggoIsEating, raisingBasket, waitingForReload};

State currentState = waitingForDoggo; // this is our starting state
int timer = 0;

void UpdateFSM()
{	
	switch(currentState)
	{
		waitingForDoggo:
			// wait until doggo is detected then start lowering the basket
			if(doggoDetected)
			{
				StartLoweringBasket();
				timer = 0;
				currentState = loweringBasket;
			}
			break;

		loweringBasket:
			// lower the basket for five seconds, then stop
			if(timer > 5seconds)
			{
				StopBasket();
				timer = 0;
				currentState = doggoIsEating;
			}
			else { timer++; }
			break;

		doggoIsEating:
			// let doggo eat for 10 seconds, then raise basket
			if(timer > 10seconds)
			{
				StartRaisingBasket();
				timer = 0;
				currentState = raisingBasket;
			}
			else { timer++; }
			break;

		raisingBasket:
			// raise basket for 5 seconds 
			if(timer > 5seconds)
			{
				StopBasket();
				currentState = waitingForReload;
			}
			else { timer++; }
			break;

		waitingForReload:
			// wait for user to reload a bone in the basket and push the 'reloaded' button
			if(userPushedReloadButton)
			{
				currentState = waitingForDoggo;
			}
			break;

		default:
			Error! We should never get here!
	} // switch
}
The forums kind of mangled that code but hopefully you can see that each time UpdateFSM() gets called from your main loop the switch statement hops to whatever state you're in and executes only a very few lines of non-hanging code before returning control to your main program.

petit choux
Feb 24, 2016


Resembles my naming conventions

Laopooh
Jul 15, 2000

PDP-1 posted:

Looks like a fun project! One way to avoid hanging up the main loop while still doing a set of steps is called a Finite State Machine (FSM) where you have a finite set of states where you can only be in one at a time. For whichever state you happen to be in at the moment you can either do a few simple tasks or make a decision to transition to another state. When you transition to another state you might also want to do a few simple tasks but these only execute once per transition. Typically you'd list out all of your states as an enumerated type and choose the active one with a switch statement something like in this pseudocode:

code:
enum State {waitingForDoggo, loweringBasket, doggoIsEating, raisingBasket, waitingForReload};

State currentState = waitingForDoggo; // this is our starting state
int timer = 0;

void UpdateFSM()
{	
	switch(currentState)
	{
		waitingForDoggo:
			// wait until doggo is detected then start lowering the basket
			if(doggoDetected)
			{
				StartLoweringBasket();
				timer = 0;
				currentState = loweringBasket;
			}
			break;

		loweringBasket:
			// lower the basket for five seconds, then stop
			if(timer > 5seconds)
			{
				StopBasket();
				timer = 0;
				currentState = doggoIsEating;
			}
			else { timer++; }
			break;

		doggoIsEating:
			// let doggo eat for 10 seconds, then raise basket
			if(timer > 10seconds)
			{
				StartRaisingBasket();
				timer = 0;
				currentState = raisingBasket;
			}
			else { timer++; }
			break;

		raisingBasket:
			// raise basket for 5 seconds 
			if(timer > 5seconds)
			{
				StopBasket();
				currentState = waitingForReload;
			}
			else { timer++; }
			break;

		waitingForReload:
			// wait for user to reload a bone in the basket and push the 'reloaded' button
			if(userPushedReloadButton)
			{
				currentState = waitingForDoggo;
			}
			break;

		default:
			Error! We should never get here!
	} // switch
}
The forums kind of mangled that code but hopefully you can see that each time UpdateFSM() gets called from your main loop the switch statement hops to whatever state you're in and executes only a very few lines of non-hanging code before returning control to your main program.


PDP, this is amazing. Thank you so much! Your use of enum, switch, and break (which I don't understand yet, but can see kinda how they work) have given me so many leads to follow (psuedocode, I know). Plus the "doggoIsEating:" containing thing, etc. I can see there are coding structures I'm unfamiliar with that could help a lot. I'll update once I figure them out (*´ω`*)

This is my slowbrains understanding of switches. I'm enjoying the connection between program logic and physical stuff but it's hard. How long is the contact point conductivity measured before moving to the next step in the program? Once the program moves on, do the switches stop being monitored? I think I'm a small/large step away from understanding how this works :sweatdrop:


Laopooh fucked around with this message at 00:33 on Sep 13, 2021

Fanged Lawn Wormy
Jan 4, 2008

SQUEAK! SQUEAK! SQUEAK!
Since it came up recently, any recommendations for lead free solder and using it? I bought some for the shop a bit back, but it’s such a pain in the rear end to get flowing I’m falling back in my leaded solders instead.

Mostly unsolder LED tapes and some circuit boards.

In most cases with the leaded stuff I can tin the wire and the pad, and then do a quick touch to join t he m without issue. I generally don’t need extra flux unless I’m doing something like RGBW led tape - often the ribbon wires for that are aluminum, which doesn’t behave as well, and the small pads make it tougher to do.

Dominoes
Sep 20, 2007

I've been using arbitrary wire. ChipQuick or MG Tin/Bismuth/silver eutectic is great for paste.

What iron and tips are you using? What temp? What sorts of things are you soldering?

Hakko chisel tips at 360C work great for a number of uses. For example, when soldering 2.54mm pitch pins onto something THT, I hold the iron on the pin (Sometimes flush against the pin, sometimes angled more to heat the board), wait a second or two, then tough the opposite side with wire.

Forseti
May 26, 2001
To the lovenasium!

Laopooh posted:

PDP, this is amazing. Thank you so much! Your use of enum, switch, and break (which I don't understand yet, but can see kinda how they work) have given me so many leads to follow (psuedocode, I know). Plus the "doggoIsEating:" containing thing, etc. I can see there are coding structures I'm unfamiliar with that could help a lot. I'll update once I figure them out (*´ω`*)

This is my slowbrains understanding of switches. I'm enjoying the connection between program logic and physical stuff but it's hard. How long is the contact point conductivity measured before moving to the next step in the program? Once the program moves on, do the switches stop being monitored? I think I'm a small/large step away from understanding how this works :sweatdrop:



The other very cool thing about finite state machines is they can be translated directly to hardware logic, so you could implement the same algorithm in e.g. 7400 series ICs if you wanted :)

babyeatingpsychopath
Oct 28, 2000
Forum Veteran


Laopooh posted:

PDP, this is amazing. Thank you so much! Your use of enum, switch, and break (which I don't understand yet, but can see kinda how they work) have given me so many leads to follow (psuedocode, I know). Plus the "doggoIsEating:" containing thing, etc. I can see there are coding structures I'm unfamiliar with that could help a lot. I'll update once I figure them out (*´ω`*)

This is my slowbrains understanding of switches. I'm enjoying the connection between program logic and physical stuff but it's hard. How long is the contact point conductivity measured before moving to the next step in the program? Once the program moves on, do the switches stop being monitored? I think I'm a small/large step away from understanding how this works :sweatdrop:

If you pop on over to the arduino thread i can make a brief effortpost about how switch works, how control flow works in arduino, what you can expect to happen when, and how some easy-mode polling works.

Fanged Lawn Wormy
Jan 4, 2008

SQUEAK! SQUEAK! SQUEAK!

Dominoes posted:

I've been using arbitrary wire. ChipQuick or MG Tin/Bismuth/silver eutectic is great for paste.

What iron and tips are you using? What temp? What sorts of things are you soldering?

Hakko chisel tips at 360C work great for a number of uses. For example, when soldering 2.54mm pitch pins onto something THT, I hold the iron on the pin (Sometimes flush against the pin, sometimes angled more to heat the board), wait a second or two, then tough the opposite side with wire.

Two Hakko FX-888Ds are my workhorses, usually with The "Hoof" type tips (conical w/ a flat spot on it) or chisel tips. I think my temps tend to be around 700-ish, but I don't always re-calibrate for new tips, so that may vary some. Actually, looking it up online, I might be running a little too hot in general, lol.

Again, typically some circuits - usually Thru-hole, but on rare occasion I'll fix smd stuff. Soldering to LED tape is probably the biggest in terms of time though, which is usually 18 AWG or smaller.

Dominoes
Sep 20, 2007

Hmm. Maybe try a new tip?

Fanged Lawn Wormy
Jan 4, 2008

SQUEAK! SQUEAK! SQUEAK!
I’ll experiment some more. I doubt it’s a new tip issue - I did some replacements relatively recently.

Dominoes
Sep 20, 2007

Are you making sure the tip is in contact with the base of the pin for at least a second (maybe more) before applying the wire? Does the wire have a rosin core?

Rescue Toaster
Mar 13, 2003
I've found even with leaded solder to be happy with the joint I always have to add flux every single time I heat the joint if I'm not feeding flux-cored solder. And it's doubly so with lead-free. I keep a syringe of quick chip gel flux right next to my iron at all times.

Every time I'm like 'oh I'll just touch this up because it dragged a bit of a point' or whatever, it always ends up looking even worse if I don't add flux.

Forseti
May 26, 2001
To the lovenasium!

Rescue Toaster posted:

I've found even with leaded solder to be happy with the joint I always have to add flux every single time I heat the joint if I'm not feeding flux-cored solder. And it's doubly so with lead-free. I keep a syringe of quick chip gel flux right next to my iron at all times.

Every time I'm like 'oh I'll just touch this up because it dragged a bit of a point' or whatever, it always ends up looking even worse if I don't add flux.

I also use lots of flux even with lead bearing solder. Part of it is because a lot of what I'm soldering is tiny things to a non-tinned copper PCB and I use the "wrong" technique of putting a bit of solder on the tip of my iron and then bringing it to the part because I only have two hands. I'm using my other hand to hold the tiny capacitor or whatever in place with a toothpick so it doesn't stick to the tip of my iron instead of the board.

I've heard great things about Multicore brand solder, it has flux in multiple cores within the solder instead of only the center. I bet that really would work better for lead free soldering, though I haven't tried it myself. It's crazy expensive though: https://www.mouser.com/ProductDetail/Loctite/737688?qs=sGAEpiMZZMtyU1cDF2RqUAE8D5Z89X%2Fj1kAD%2FIEsvgQ%3D

I checked my favorite though (Kester 63/37) and it's also a lot more expensive than when I bought it :(. Great, now I have the urge to hoard it...

KnifeWrench
May 25, 2007

Practical and safe.

Bleak Gremlin
I'm struggling to find the search terms for what I need. Maybe someone here has a recommendation?

I'm trying to set up a connector for reasonably high current DC (~25A) such that there is an interlock or pilot contact preventing a user from unplugging while current is flowing. Either a second set of contacts that makes contact at a different depth of mating, or a switch that can only be actuated when the connector is mated, and prevents the connector from being removed unless the switch is off.

This probably doesn't exist, and even if it does, it's probably too expensive, but I want to know what's out there before I decide "process to keep people from being idiots" is the best option.

Forseti
May 26, 2001
To the lovenasium!

KnifeWrench posted:

I'm struggling to find the search terms for what I need. Maybe someone here has a recommendation?

I'm trying to set up a connector for reasonably high current DC (~25A) such that there is an interlock or pilot contact preventing a user from unplugging while current is flowing. Either a second set of contacts that makes contact at a different depth of mating, or a switch that can only be actuated when the connector is mated, and prevents the connector from being removed unless the switch is off.

This probably doesn't exist, and even if it does, it's probably too expensive, but I want to know what's out there before I decide "process to keep people from being idiots" is the best option.

The XT90 can be had with spark arrestors and is common for eBikes: https://www.amazon.com/XT-90S-Connector-Silicon-RC-Battery/dp/B08L3S2ZK8

I'm not sure if that solves the problem? You can still physically disconnect it under load.

Not sure what a generic search term would be for the interlock you're describing, but implementation wise I would think maybe a small solenoid that moves a pin that can physically prevent disconnecting it while it's powered. Seems complicated and expensive though.

KnifeWrench
May 25, 2007

Practical and safe.

Bleak Gremlin

Forseti posted:

The XT90 can be had with spark arrestors and is common for eBikes: https://www.amazon.com/XT-90S-Connector-Silicon-RC-Battery/dp/B08L3S2ZK8

I'm not sure if that solves the problem? You can still physically disconnect it under load.

Not sure what a generic search term would be for the interlock you're describing, but implementation wise I would think maybe a small solenoid that moves a pin that can physically prevent disconnecting it while it's powered. Seems complicated and expensive though.

Spark arrestors may be a smart/not-over-engineered solution. The main thing is I don't want to have a human do something that I wouldn't want a switch to do (arc/bounce).

Thanks!

KnifeWrench fucked around with this message at 17:49 on Sep 13, 2021

babyeatingpsychopath
Oct 28, 2000
Forum Veteran


KnifeWrench posted:

I'm struggling to find the search terms for what I need. Maybe someone here has a recommendation?

I'm trying to set up a connector for reasonably high current DC (~25A) such that there is an interlock or pilot contact preventing a user from unplugging while current is flowing. Either a second set of contacts that makes contact at a different depth of mating, or a switch that can only be actuated when the connector is mated, and prevents the connector from being removed unless the switch is off.

This probably doesn't exist, and even if it does, it's probably too expensive, but I want to know what's out there before I decide "process to keep people from being idiots" is the best option.

CEE plugs have pilot pins that form electrical interlocks. Basically, you run your contactor coil voltage through the pilot pin connection and it makes after all the power pins are fully seated and breaks before all the power pins are fully unseated.

https://en.wikipedia.org/wiki/IEC_60309#Pilot_contact

Shame Boy
Mar 2, 2010

I'm looking at MG Chemicals' site a lot lately cuz I've been using their conductive paints to make a 3D printed box shielded and they have a delightful animation at the top of the electroless plating section:

https://www.mgchemicals.com/category/electronics-prototyping/electroless-plating/

e: And if anyone's wondering, no you can't electroless-tin plate the conductive paint. It only seems to work on copper, and the paint is either silver-based or silver-coated-copper, depending on how much you wanna pay for it. Was really hoping I could just dunk a part in it and watch as a hard metal shell of tin magically formed around it but no, it just kinda sat there doing nothing and then I had a mess to clean up :v:

Shame Boy fucked around with this message at 15:40 on Sep 14, 2021

Forseti
May 26, 2001
To the lovenasium!

Shame Boy posted:

I'm looking at MG Chemicals' site a lot lately cuz I've been using their conductive paints to make a 3D printed box shielded and they have a delightful animation at the top of the electroless plating section:

https://www.mgchemicals.com/category/electronics-prototyping/electroless-plating/

e: And if anyone's wondering, no you can't electroless-tin plate the conductive paint. It only seems to work on copper, and the paint is either silver-based or silver-coated-copper, depending on how much you wanna pay for it. Was really hoping I could just dunk a part in it and watch as a hard metal shell of tin magically formed around it but no, it just kinda sat there doing nothing and then I had a mess to clean up :v:

:lol: that really is a fantastic animation

Shame Boy
Mar 2, 2010

Oh man there's an animation for this category too, which I assume means there's one for every category, and they're all excellent, how have I never noticed this before:

https://www.mgchemicals.com/category/electronics-prototyping/



I like the implication that their ammonium persulfate and positive developer solutions are both convicted felons

ante
Apr 9, 2005

SUNSHINE AND RAINBOWS

Shame Boy posted:

I'm looking at MG Chemicals' site a lot lately cuz I've been using their conductive paints to make a 3D printed box shielded and they have a delightful animation at the top of the electroless plating section:

https://www.mgchemicals.com/category/electronics-prototyping/electroless-plating/

e: And if anyone's wondering, no you can't electroless-tin plate the conductive paint. It only seems to work on copper, and the paint is either silver-based or silver-coated-copper, depending on how much you wanna pay for it. Was really hoping I could just dunk a part in it and watch as a hard metal shell of tin magically formed around it but no, it just kinda sat there doing nothing and then I had a mess to clean up :v:

Thanks for that edit, that's exactly what I started thinking about while reading the blurb

Shame Boy
Mar 2, 2010

ante posted:

Thanks for that edit, that's exactly what I started thinking about while reading the blurb

There is actually a process for electroless-plating 3D printed plastic I found, but it's much more involved and requires a bunch of steps including "bathe the whole thing in acid that eats the plastic to increase the surface area" and "coat in an initial catalyst solution containing gold or platinum" which puts it a bit out of reach of the casual hobbyist. I've seen some people get good results with normal electroplating on top of the conductive paint though, and I guess if you lay down copper first you could immersion-tin it after that. Apparently with electroplating you can use the much cheaper graphite-based paint too, so that'd be nice. However it's a bit more work than the "dunk it in the liquid tin I have lying around" process I was hoping for, and the conductive paint is more than good enough for what I'm trying to do by itself, so meh.

Leandros
Dec 14, 2008

Are (lithium) battery-powered devices designed with the capabilities of the sourced batteries in mind? By which I mean, I know that there's a fair bit of variation in e.g. 18650 chemistry/quality/capacity/current capabilities, and I would like to replace a set of them in my vacuum with new ones.
Ideally I want to go for the highest capacity/current combination, but I was wondering if that might blow up my batteries because their internal resistance is lower than expected by the design, or something like that.

Hexyflexy
Sep 2, 2011

asymptotically approaching one

Forseti posted:

:lol: that really is a fantastic animation

I'm posting these to my friends at the moment, they're brilliant :)

Hexyflexy
Sep 2, 2011

asymptotically approaching one

Leandros posted:

Are (lithium) battery-powered devices designed with the capabilities of the sourced batteries in mind? By which I mean, I know that there's a fair bit of variation in e.g. 18650 chemistry/quality/capacity/current capabilities, and I would like to replace a set of them in my vacuum with new ones.
Ideally I want to go for the highest capacity/current combination, but I was wondering if that might blow up my batteries because their internal resistance is lower than expected by the design, or something like that.

Not my field but my bf builds electronic gadgets, art things, and he's spent a stupid amount of time the last couple of years working out how different types of battery chemistry and manufacturer deal with low power draw conditions, because it makes a huge amount of difference to how a thing can sit at idle on a shelf before you turn it on ocasionally.

I've refurbished a few bits of equipment e.g. 1980s fluke multimeters and I've been careful to get something as similar as possible to the original batteries, because I'm wary the engineer who designed the battery maintainance circuit was expecting a specific internal resistance or something.

KnifeWrench
May 25, 2007

Practical and safe.

Bleak Gremlin

Leandros posted:

Are (lithium) battery-powered devices designed with the capabilities of the sourced batteries in mind? By which I mean, I know that there's a fair bit of variation in e.g. 18650 chemistry/quality/capacity/current capabilities, and I would like to replace a set of them in my vacuum with new ones.
Ideally I want to go for the highest capacity/current combination, but I was wondering if that might blow up my batteries because their internal resistance is lower than expected by the design, or something like that.

My gut, which is informed but not an authority, tells me the biggest issue you would see due to a different internal resistance would be a lower accuracy on the reported capacity for any device that's compensating for load when using a lookup table. That's a pretty minor effect on a pretty small set of devices -- doubtful your vacuum is one of them.

Important things to match/exceed:
Max voltage: should be the same for the same chemistry, but make sure it's at least as high as the old cells
Max charge rate: if the new cells have a different capacity, you might need to do a little math, as this is frequently cited in C rather than Amps (for a 2Ah cell, 1C = 2A). Make sure the new cells can handle at least the charge rate of the old cells.

Again, I could absolutely be missing something, and the charger might make some crazy assumptions that could be a problem, but it's not unfair to assume the designer used a standard design rather than go out of their way to make it bespoke to the particular cell manufacturer, so I think it should probably be okay.

Forseti
May 26, 2001
To the lovenasium!

Leandros posted:

Are (lithium) battery-powered devices designed with the capabilities of the sourced batteries in mind? By which I mean, I know that there's a fair bit of variation in e.g. 18650 chemistry/quality/capacity/current capabilities, and I would like to replace a set of them in my vacuum with new ones.
Ideally I want to go for the highest capacity/current combination, but I was wondering if that might blow up my batteries because their internal resistance is lower than expected by the design, or something like that.

The really big one is don't use an 18650 that's not meant for a vape in a vape because the ones meant for vapes are made to minimize the failure mode of "blows up in your god drat face".

Other than that, I think it's just match the C rating (which tells you the current capacity) to your needs. I guess you should watch the charge current but the vacuum probably doesn't have a very fast charging circuit anyway.

If you want to scavenge batteries, ones that come from a lithium tool battery pack should be fine in your vacuum I would bet.

BattleMaster
Aug 14, 2000


That's a Raspberry Pi with a card edge connector added :eyepop:

Mr. Bubbles
Jul 19, 2012
Could anyone held a simpleton like myself decrypt this diagram? Context: Control of an electronic curtain rod.

https://www.dropbox.com/s/tjw9n54omrxf2ch/RJ%2012%20Port%20for%20Home%20Automation-1.pdf?dl=0

I don't know what the 1/2 and 3/4 logic is referring to. I hope to power this with ESPHome on an ESP8266 / Home assistant server using a few relays to open and close these switches. Do I choose one of the two (1/2 or 3/4) logic methods to control it, or use both? Is the 1/2 the simpler one for control with relay where pin 1 is the common wire and either pin 5 or 6 is closed via relay to complete the circuit to open or close the curtain? Thanks

Leandros
Dec 14, 2008

^^^Thanks all, makes sense. I was gonna buy some supposedly unused scavenged ones with more capacity (the ones in the vacuum were 2000mAh vs these 3000), just as a nice to have. Considering these are 4.2V charged I doubt it'll be a big issue to match max voltage, no?

xzzy
Mar 5, 2009

BattleMaster posted:

That's a Raspberry Pi with a card edge connector added :eyepop:

It's a pi3 layout, but all the ports are on the wrong edge. It's like they had the thought to avoid copyright issues, but then decided it didn't matter and slapped the logo on anyways.

Adbot
ADBOT LOVES YOU

poeticoddity
Jan 14, 2007
"How nice - to feel nothing and still get full credit for being alive." - Kurt Vonnegut Jr. - Slaughterhouse Five

Mr. Bubbles posted:

Could anyone held a simpleton like myself decrypt this diagram? Context: Control of an electronic curtain rod.

https://www.dropbox.com/s/tjw9n54omrxf2ch/RJ%2012%20Port%20for%20Home%20Automation-1.pdf?dl=0

I don't know what the 1/2 and 3/4 logic is referring to. I hope to power this with ESPHome on an ESP8266 / Home assistant server using a few relays to open and close these switches. Do I choose one of the two (1/2 or 3/4) logic methods to control it, or use both? Is the 1/2 the simpler one for control with relay where pin 1 is the common wire and either pin 5 or 6 is closed via relay to complete the circuit to open or close the curtain? Thanks

Someone who's not currently running a fever may have better information, but here's my take:

In the 1/2 control, you either have a pair of buttons or a toggle switch with NC dead zone in the middle.
This is only suitable for supervised manual control or when you have some other halt condition that prevents you from pushing past fully closed/open.
I would personally avoid this unless you have a very good reason to use it.

The 3/4 control has a third line (SW) which can be used with a microswitch, hall effect sensor, or something else to incorporate and end-stop.
Unless you have some external safety mechanism or your moving parts can handle being driven past intended end stops, this is the set up I'd use for anything other than carefully monitored manual control (and even then I'd want the end-stop cut-off switches).

If someone else has a different take on this, please correct me, especially if there's some safety consideration I've overlooked.

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply