|
BattleMaster posted:Yeah you mainly just stop when the count rate suddenly jumps after you step the voltage up, you don't go deep enough into the continuous discharge region to get arcing or whatever that can damage the tube. This is such a standard calibration procedure that they even taught us how to do it as students and let us do it on real (expensive) equipment without issue. BattleMaster posted:I just went and ran some old soviet end window tube I had (forgot the model) that operates at a very high voltage of around 2500 volts or so with my polonium-210 disc source and I wasn't seeing any visible glowing or arcing. It's not something that should be expected either. With the information you added either the voltage spikes are too high for the tube or maybe you're letting too much current flow through it. What does your circuit and component values look like? I went back and calculated the current that's likely flowing through it and yeah I'm thinking it's an order of magnitude too high. I've got a 10M resistor on the anode and a variable 0-50K resistor between the cathode and ground. That 10M resistor would be passing 50uA at 500V worst-case which is waaaay more than the 1uA the tube's spec says it should be getting. Oops.
|
#
?
Mar 8, 2017 05:02
|
|
|
|
| # ? May 8, 2024 11:14 |
|
|
Zero VGS posted:Has anyone used one of these analog voltmeters before? Was thinking of installing one in my EV: -Yes. -Watts = Voltage^2 / resistance. Hopefully it is high. Find a data sheet. -All needle gauges are, that is one reason why people moved to digital displays. - You are dealing with very high voltages and currents. It will help. a little. My only other advice is Please Don't.
|
|
#
?
Mar 8, 2017 05:14
|
|
|
Is this for a DIY EV or a commercial model? If it’s a commercial model, I wouldn’t gently caress with the battery. There’s already a built‐in voltmeter somewhere. Find some way to tap it (like, off the data bus) if you really want an analogue gauge on the dash. If you do make a connection to the battery, get good fuses. They’re more expensive than you’d think. Platystemon fucked around with this message at 06:42 on Mar 8, 2017 |
|
#
?
Mar 8, 2017 06:22
|
|
|
Platystemon posted:Is this for a DIY EV or a commercial model? It's an older factory made EV (OEM Ford Ranger Electric), it actually uses an analog needle to show you "miles remaining" on the dash but I'm swapping the Pb battery pack for Li-Ion which has a totally different voltage curve for SOC, and I figured it'd be easier at a glance to put in a meter for actual voltage. There are some cure 2-wire digital displays, but they only range up to 32 volts, beyond that you have 3-wire digital displays where you hook them to both the HV battery and a 12v battery, so I'd like to try for the simplest possible solution first. You might be right that it could have it's own voltmeter accessible by Canbus or OBDII or something, I'll go through the literature.
|
|
#
?
Mar 8, 2017 07:04
|
|
|
Beware that voltage is not the most reliable measure of li‐ion state‐of‐charge.
|
|
#
?
Mar 8, 2017 07:11
|
|
|
Zero VGS posted:- How many watts does it use when on? If it's just a watt or two I'll probably have it on 24/7 Very little. Most analog panel meters are 50uA - 1mA full scale. Voltage panel meters typically use the more sensitive movements to disturb a voltage source less (an ideal voltage meter would pass 0 current), though a high voltage meter like this one. Given that this one is designed for 500v, if we take worst case of 1ma and 500v it'd be half a watt, and is probably ~1/10th of that (and possibly less). As an aside, analog panel meters that measure higher currents put a shunt in parallel with the movement to handle the bulk of the current, but otherwise use the same movements. Meters used for measuring voltage put a resistor in series to drop it. Assuming the same 1ma full scale @ 500v, the resistance of the meter would be ~500000ohms, and about 499900 of those ohms would be from a resistor. Zero VGS posted:- Is it sensitive to movement? i.e. would driving my vehicle make the needle waver all over the place or does it tend to stay in place? Some are more sensitive than others, but in general they all move. That style is typically one of the more sensitive types. this style is typically less sensitive to movement Zero VGS posted:- If I put some very low amp fuses on this at it's own terminals and the terminals where it connects to the battery pack, that should keep things safe, right? Don't skimp on the voltage rating of the fuse. Overvolting a fuse can cause it to simply arc over and not actually fuse at all. Note that the fuse is to protect you, not the meter. The meter will have a very high resistance, and barring a manufacturing defect, be fine with the 500v. You aren't.
|
|
#
?
Mar 8, 2017 07:16
|
|
|
Platystemon posted:Beware that voltage is not the most reliable measure of li‐ion state‐of‐charge. If it isn't... then what is? The only other thing I can think of would be monitoring all the amps going in/out of the battery.
|
|
#
?
Mar 8, 2017 07:42
|
|
|
Zero VGS posted:If it isn't... then what is? The only other thing I can think of would be monitoring all the amps going in/out of the battery. That’s what your phone does. It’s called “coulomb counting”.
|
|
#
?
Mar 8, 2017 07:45
|
|
|
Platystemon posted:That’s what your phone does. It’s called “coulomb counting”. Don't think my phone bothers with that. I've repeatedly seen it decide I've gained 10% of my battery capacity back just from cooling down.
|
|
#
?
Mar 8, 2017 07:47
|
|
|
Zero VGS posted:If it isn't... then what is? The only other thing I can think of would be monitoring all the amps going in/out of the battery. Check out the graph on this page, of a set of typical discharge curves for a Li-Ion cell (because I don't feel like reuploading): https://learn.adafruit.com/li-ion-and-lipoly-batteries/voltages The two things you should notice about this are 1) The capacity changes a lot while the voltage remains relatively flat. This means your measurement needs to be very accurate even in the best case, just to get a good rough estimate of capacity. 2) Different curves for different discharge rates. In other words, voltage also depends on load. So in the middle of your highly sensitive flat portion of the graph, you could have swings of 10, 20, 30 percent depending on whether you're exerting and how much. There are two ways to solve this, but both require that you measure current. The first, coulomb counting, has already been mentioned, and it's great, but arguably more complicated to implement. I would implement a coulomb counter if I was concerned with battery degradation and wanted a mechanism for autonomous self-calibration. That might be overkill, or it might be extremely shrewd in your application. The other way is to characterize the battery. This will front-load your effort but be easier to implement. Load will sag the voltage, but by how much? Well, V = IR, so if you're measuring voltage and current, you just need to know the battery's impedance. You can do this by using an artificial load of a known or measured current to discharge the battery, measuring its voltage periodically, and then disconnecting the load and measuring the voltage at 0 load. This will give you a set of points for known capacities (because you can hand crank the coulomb counting) where you know deltaV and deltaI, so you can calculate R. R will probably be small, but may change with capacity, so you can either take the average (which will be good enough in most cases) or generate a lookup table. In either case, you now can augment your rudimentary Voltage-only capacity measurement with the expected change due to the instantaneous load. You are measuring voltage V' = V_0A + I_load * R_batt. Use the measured quantities to solve for V_0A and then reference into your table you built during the battery characterization, and you'll have a much more stable and accurate capacity estimate. If you have the budget for it, battery characterization can be made very straightforward with a programmable load like this: http://www.digikey.com/product-detail/en/b-k-precision/8500/BK8500-ND/1135014 but it's absolutely not required. An ammeter, a voltmeter, and a fixed load are all you need if you have the patience to babysit your experiment for many hours. (Oh, and of course, don't overdischarge your battery) KnifeWrench fucked around with this message at 20:35 on Mar 8, 2017 |
|
#
?
Mar 8, 2017 15:54
|
|
|
ate all the Oreos posted:I went back and calculated the current that's likely flowing through it and yeah I'm thinking it's an order of magnitude too high. I've got a 10M resistor on the anode and a variable 0-50K resistor between the cathode and ground. That 10M resistor would be passing 50uA at 500V worst-case which is waaaay more than the 1uA the tube's spec says it should be getting. Oops. 10M should be fine, in fact I use 4.7M in some of the stuff I've built. You'd need an incredible count rate to go over 1 uA or a dead short, and if you're getting a dead short the tube is probably dead anyway. I was mostly wondering if you were using something ridiculous like 100k but that doesn't seem to be the case. Try it with a power supply that's more stable and see if it still glows.
|
|
#
?
Mar 8, 2017 17:20
|
|
|
How are the alignment holes on the tail of an OLED supposed to be used?
|
|
#
?
Mar 9, 2017 18:55
|
|
|
Are you speccing out a hotbar solder?
|
|
#
?
Mar 9, 2017 21:40
|
|
|
If I'm crimping some heavy gauge lugs, there's no harm in adding a dab of this paste to the cable end first, right?: https://www.amazon.com/MG-Chemicals-Conductive-Assembly-Synthetic/dp/B00RE0QP1W I read you're suppose to crimp lugs on to heavy duty cable, but never solder them because it makes them brittle against vibration. I figured the paste would prevent corrosion and give a bit of extra insurance.
|
|
#
?
Mar 9, 2017 21:53
|
|
|
Afaik properly crimped connections are supposed to be gas tight so theoretically there would be no benefit to doing something like that - if anything it may prevent the cold weld from forming.
|
|
#
?
Mar 9, 2017 22:13
|
|
|
JawnV6 posted:Are you speccing out a hotbar solder? Hmm, I didn't even know that was a thing. That is a good term to google. I was hoping to pull off reflow soldering in a modified toaster oven with the tail pinned in place by the tooling holes.
|
|
#
?
Mar 10, 2017 01:13
|
|
|
Apologies if I'm in the wrong thread (though I don't believe I am) I'm building an arcade game that needs fancy-rear end special controls and has to go into a custom cabinet and trying to rig up all sorts of things and putting the cart before the horse, as they say. Anywho, this cabinet is going to be controlled with four 3" trackballs, two per player. Trackballs are loving expensive but I managed to snag 7 of them off a dude who owns a bar and swaps them out of his Golden Tee machine every few months for about $12 a pop. Score. Little bit of windex and some elbow grease on the rollers and they're basically brand new trackballs at this point. Then I decided that if I'm going to have four drat trackballs in a dark room they might as well light up, of course. Brand new lit trackballs are... expensive, to say the least. Fortunately, contact juggling balls are only like $11 on Amazon and are, best as I can tell, a perfect replacement for the white balls that I have. So now on to my issue: There's a hole in the bottom of the trackball assembly that I can easily pass some light through. What do y'all wager would be the best setup for this? Multiple standard LEDs in a ring? One big fuckoff 9W LED from adafruit under a diffusion screen? Something else? Only thing I had convenient to test the juggling ball was the LED hat on my raspberry pi that's usually just sitting in a corner drawing pretty pictures, but it did a relatively good job: https://www.youtube.com/watch?v=Nw9pLE7fnuM Just wondering if I can alter my setup in any way to get better light show going ( I should probably start a thread for this whole project ) e: Thinking about it now, having made this post, I guess my real problem is I'd want a sorta cloudy ball, rather than this perfectly clear guy I've got going on. Anyone got a good source for mostly-transparent acrylic balls on the cheap? Sockser fucked around with this message at 07:06 on Mar 10, 2017 |
|
#
?
Mar 10, 2017 07:02
|
|
|
taqueso posted:Hmm, I didn't even know that was a thing. That is a good term to google. I was hoping to pull off reflow soldering in a modified toaster oven with the tail pinned in place by the tooling holes. You can also get little ZIF sockets for that style of connector.
|
|
#
?
Mar 10, 2017 08:14
|
|
|
Sockser posted:e: Thinking about it now, having made this post, I guess my real problem is I'd want a sorta cloudy ball, rather than this perfectly clear guy I've got going on. Anyone got a good source for mostly-transparent acrylic balls on the cheap? If you want it looking frosted as opposed to "swirling patterns shot through the ball"-style cloudy and the ball is acrylic, hit it with some very fine sandpaper, like 2000 grit. That should give the surface a frosted look and diffuse any light going through it fairly well.
|
|
#
?
Mar 10, 2017 15:01
|
|
|
csammis posted:If you want it looking frosted as opposed to "swirling patterns shot through the ball"-style cloudy and the ball is acrylic, hit it with some very fine sandpaper, like 2000 grit. That should give the surface a frosted look and diffuse any light going through it fairly well. Not a bad call, and not an expensive experiment if things go south. Going to give this a poo poo later today, thanks.
|
|
#
?
Mar 10, 2017 16:09
|
|
|
Try manually placing a few LEDs under the ball and see if certain arrangements mix nicely and would work well for swirling patterns. I feel like a ring of 5 around the bottom would be decent. When I was messing around with a deep-red super-bright LED for a bike tail-light, using sanded acrylic* as a diffuser, it didn't take much power to make it glow. And it was pretty easy to get too much light and make it hard to look at. You probably don't need a really powerful LED. *It was actually Acrylite, acrylic that has diffusion beads embedded in it. I don't think they make spheres, but this stuff is pretty awesome at spreading out the light.
|
|
#
?
Mar 10, 2017 16:21
|
|
|
If you're going with an LED ring I've had good luck with the NeoPixel rings: https://www.adafruit.com/product/1463 Each LED is individually addressable too so you can make a bunch of neat multi-color patterns
|
|
#
?
Mar 10, 2017 16:40
|
|
|
ate all the Oreos posted:If you're going with an LED ring I've had good luck with the NeoPixel rings: Seconded. You'll need some kind of controller, but they get brighter than you'll need, and once you have the controller figured out, you have lots of flexibility. Don't miss that there are different sized rings and even a 7-LED "jewel" so you have options for your physical layout.
|
|
#
?
Mar 10, 2017 16:47
|
|
|
You can also buy strings of the same addressable LED's (or even individual LEDs on breakout boards) on the cheap from china on EBay if you don't mind waiting 1-4 weeks, I got a big string of 100 of them for like $10 that way. The Adafruit rings are much nicer quality, though
|
|
#
?
Mar 10, 2017 16:49
|
|
|
1. Had some 400 grit sandpaper lying around and gave it a quick scuff and it worked great! 2. I think the babiest of neopixel rings will probably be perfect, I'm gonna swing by microcenter and grab one or two tomorrow while I'm there. Should mount nicely under the hole in the trackball panel without shooting a million lumen beam straight through, blinding anyone who puts their face above it (The above/video before are both lit with an 8x8 grid of neopixels which is totally overkill, but just what I have handy)
|
|
#
?
Mar 10, 2017 20:13
|
|
|
Nice! Glad the sanding worked 
|
|
#
?
Mar 10, 2017 20:47
|
|
|
.
taqueso fucked around with this message at 04:59 on Mar 12, 2017 |
|
#
?
Mar 10, 2017 21:23
|
|
|
A 12 neopixel ring literally could not fit a 3" happ trackball any better, it's incredible, really https://www.youtube.com/watch?v=NfrfayB2PZE
|
|
#
?
Mar 12, 2017 07:32
|
|
|
Sockser posted:A 12 neopixel ring literally could not fit a 3" happ trackball any better, it's incredible, really Cool, that thing looks rad as gently caress Anyway, does anyone know what the hell this thing is? I found it in a random google image search and if it's what I think it is (a big breadboard with embedded power supply/supplies) that seems quite useful: 
|
|
#
?
Mar 13, 2017 02:04
|
|
|
I don't meant to be a dick, but the thing is clearly labeled: It's an NI Elvis II
|
|
#
?
Mar 13, 2017 02:15
|
|
|
Corla Plankun posted:I don't meant to be a dick, but the thing is clearly labeled: It's an NI Elvis II Oh. Obviously. I spent like 10 minutes trying to figure out what that text in the middle said (L1S2 I think?) and managed to not even see that text 
|
|
#
?
Mar 13, 2017 02:27
|
|
|
And of course, because it's NI, it costs 5x more than "too much". Edit: I guess it has more in it than it looks like, but still.
|
|
#
?
Mar 13, 2017 02:34
|
|
|
KnifeWrench posted:And of course, because it's NI, it costs 5x more than "too much". Yeah that plus it's meant for colleges, who will happily pay a ton, oh well. I know my dad made a power supply / breadboard combo thing himself a while ago, maybe I'll just do that.
|
|
#
?
Mar 13, 2017 02:39
|
|
|
Here's another breadboard, power supply, and integrated other stuff. Much less capable than that ni thing. https://www.amazon.com/Global-Specialties-PB-503-Digital-Workstation/dp/B005S3SC0E Still aimed at educators so it's still too expensive imo.
|
|
#
?
Mar 13, 2017 03:34
|
|
|
Aurium posted:Here's another breadboard, power supply, and integrated other stuff. Much less capable than that ni thing. That seems a bit too involved for what I want though, thinking about it I guess I just need something that has 3.3v/5v/12v rails automatically so I don't have to hook up extra power supplies when I'm trying to use my bench power supply for some weird other voltage which has come up a lot lately for some reason. I can probably make it out of a computer PSU (which I already have plenty of) and some breadboards now that I'm thinking about it so never mind, thanks anyway 
|
|
#
?
Mar 13, 2017 06:11
|
|
|
Digilent makes these bricks that are a little pricey but pretty slick. They take USB micro as input and fit boards with 0.1″ spacing.
|
|
#
?
Mar 13, 2017 06:53
|
|
|
KnifeWrench posted:Check out the graph on this page, of a set of typical discharge curves for a Li-Ion cell (because I don't feel like reuploading): Thanks for all that, I've been studying up. I was talking to an engineer who sells a high voltage battery charger for an electric car (the kind that converts AC from the grid into DC for the battery). I want to charge the 350v battery I have via that charger, but I also want to use a 350vdc solar panel array attached to the roof (this is a box van style truck so I have lots of surface area) to charge the batteries when the sun is out. I asked if there was a dependable way to disconnect the solar charge controller when charging from the grid, but he said I actually don't have to, because I can connect the solar charge control straight to the battery and it'll "pitch in" while the AC charger is going. I can't wrap my head around how that would work. I thought that if the batteries are at a low state of charge (say they are at 315v) and you apply 350vdc from the solar charger, then when the AC charger checks it will see 350v from the solar charger and become confused. But the engineer was saying that the AC charger will only see the battery's 315v, because the 350v from the solar charger won't increase the battery voltage until enough amps have gone into the battery. I assumed tapping a multimeter to the battery negative and positive would see whatever the highest charger voltage was at the time. Is there some kinda Youtube or something that explains how this works? Is it that the 350v charge from the solar is getting "sucked in" to the battery so fast that the other chargers don't see it?
|
|
#
?
Mar 13, 2017 17:33
|
|
|
Zero VGS posted:I can't wrap my head around how that would work. I thought that if the batteries are at a low state of charge (say they are at 315v) and you apply 350vdc from the solar charger, then when the AC charger checks it will see 350v from the solar charger and become confused. But the engineer was saying that the AC charger will only see the battery's 315v, because the 350v from the solar charger won't increase the battery voltage until enough amps have gone into the battery. You are right. The AC charger will not see 315 V. The solar charger has to apply somewhat more than 315 V to the terminals to get any current to flow into the battery and charge it. But it’s also probably not going to be applying 350 V, the nominal full voltage of the battery. For one thing, too much current would flow if you applied full voltage to an empty battery. For another, the solar panels don’t have enough power to manage it anyway. So instead some intermediate voltage will be applied, let’s say 320 V. (I just made this number up. The exact value depends on the composition of the battery and the power of the solar panel.) 320 V is what the AC charger would see. Now, the AC charger is probably going to operate in constant current mode. It wants to put, say 100 amps into the battery (again, ignore the specifics of this number). It will apply whatever voltage results in 100 A. As the battery fills, the charger ramps up the voltage to maintain 100 A. It switches to constant voltage for top‐off. See these curves:  The problem is that the current the AC charger applies will be in addition to whatever the solar charger is doing. The battery won’t be charged at a rate of 100 A. It will be charged at 100 A + whatever the solar charger is capable of. This could result in charging the battery at a rate higher than is healthy. This is all assuming a minimally smart charger. A car charger might well be smarter and get more confused, like if it’s trying to ascertain the health of the pack. Let us now assume that the engineer you talked to isn’t an idiot. You were talking about lithium ion batteries, right? Some other chemistries can be charged at constant voltage, in which case it doesn’t matter if if the solar charger is connected or not. Or maybe he knows that it’s a big battery and that even both chargers at maximum power will not charge it too fast. He presumably knows something about the product he sells. But it’s definitely not true that “the AC charger will only see the battery's 315v”. That is, at best, an approximation for a weak solar charger on a big battery.
|
|
#
?
Mar 13, 2017 19:03
|
|
|
Thank you, that makes a ton more sense. I have an entire 75kwh Tesla battery pack, which at a Tesla Supercharger station will take 120kw without breaking a sweat. The AC charger I'm getting is 6.6kw and the solar panels will put out at maximum 5kw (this is if I construct some fold-out canopies for them to triple the surface area and I'm in direct sunlight. The charger engineer said others have repurposed the Tesla packs to upgrade other EV's like I have, and the packs are so efficient that they don't even need cooling while charging until you get up to around a 20kw rate. I have an Orion BMS that will directly control the AC charger and can tell how exactly to charge, so maybe I can have it account for what the solar contributes. One other question, you said that if the pack is at 315v, the solar charger might toss 320v at it, and the AC charger would see that. What would prevent the AC charger from then saying "OK, I'll charge at 325v" and the solar charger follows up with 330v, as they try to leap frog each other?
|
|
#
?
Mar 13, 2017 19:25
|
|
|
|
| # ? May 8, 2024 11:14 |
|
|
ate all the Oreos posted:That seems a bit too involved for what I want though, thinking about it I guess I just need something that has 3.3v/5v/12v rails automatically so I don't have to hook up extra power supplies when I'm trying to use my bench power supply for some weird other voltage which has come up a lot lately for some reason. I can probably make it out of a computer PSU (which I already have plenty of) and some breadboards now that I'm thinking about it so never mind, thanks anyway I have one of these Dangerous Prototpyes ATX power supply breakout boards and it's pretty good. Can't say I've ever used it with breadboards though. They have a nice acylic case available for it too.
|
|
#
?
Mar 13, 2017 19:40
|
|