what you think is the right thing may well be the opposite N0ZYC I've been wanting to build my own high powered electronic load for low voltage, like 12vdc N0ZYC Great Scott did a nice project on YouTube that I'm considering adapting N0ZYC https://vftp.net/N0ZYC/hobbies/electronics/test%20equipment/electronic%20loads/variable%20loads/constructing%20my%20own/Great%20Scott%20adjustable%20load%20project/full/schematic.png N0ZYC it's arduino controlling a TC4420 (a mosfet gate driver) driving a IRFZ44 mosfet N0ZYC I've got MUCH larger mosfets here on hand. will the TC4420 still be able to drive larger mosfets? or several in parallel? splud with an Allegro ACS712 hall-effect current sensor. N0ZYC yes I realize I will need a bigger current monitor splud it is a 6A high-speed driver. N0ZYC I have a pack of these STE180NE10 splud so, haven't seen you indicate how many watts you expect your kit to dump. splud thinks it'd be kewl if YOU linked to datasheets. N0ZYC I also have a large cube shaped heat sink, with flat sides for mounting the mosfets, the fins are on the inside and are meant to be cooled with air forced from end to end, with fans mounted on the ends splud "here's a part number, anyone thinking of discussing, go find it" N0ZYC https://vftp.net/N0ZYC/hobbies/electronics/raw%20parts/mosfet/high%20current%20mosfets/STE180NE10%20180A%20100v%20360w%20power%20mosfet.pdf splud so, you might want your load to have it's own supply (separate from whatever you're loading), so that it functions more purely as a resistive load. splud and can run to very low current without the added load of the mcu and other bits. N0ZYC yeah I am replacing a purely resistive load I made *decades* ago, but I want this one to be smoothly adjustable, and having an mcu to select and maintain a fixed current (or power) load would be so nice. I've just resigned myself to requiring a separate power N0ZYC my commercial electronic load requires plugging into ac power, and has a screen you about need a magnifying glass to read. NOT ideal. it also caps out at 400w fredp has left IRC (Ping timeout: 252 seconds) N0ZYC I want my load to be good to 1kw if possible, (or more) and to run on internal batteries to run the MCU and mosfets N0ZYC so its portable splud BTW, IME gate driver = PWM (or at a minimum, digital, on or off). Whhich means pulsed power. N0ZYC yep I haven't dived too deeply into the arduino program but I assume its feeding PWM to that driver chip N0ZYC the author of that program is using direct port calls in the program, making it very hard to read and not easily used on a different model of arduino, so I will probably be rewriting it from the ground up. well within my capabilities splud so you have an "average" load. Usually would expect the load fet to be driven linearly. N0ZYC I assumed that's what the RC network in the lower right was for, to average out that PWM splud gate druver INPUT pin is tied to D9.. splud and no RC, so even if D9 is outputting PWM, it's still digital input to the gate driver. N0ZYC I was commenting on U2's output N0ZYC R1 and C3 B-| has joined (~b-pipe@user/B-/x-9249737) djph N0ZYC: do you have the code by chance? N0ZYC https://vftp.net/N0ZYC/hobbies/electronics/test%20equipment/electronic%20loads/variable%20loads/constructing%20my%20own/Great%20Scott%20adjustable%20load%20project/F49TEF8JL6USSSR.ino splud yea, though that then seems less something you need a driver for, doesn't it? splud pictures 6A through RC N0ZYC the arduino's dpins are pretty low current. I assumed U2 was being used as a buffer N0ZYC I don't work with mosfets much, I don't know what they require. splud why not use an opamp? mbiscuit has left IRC (Quit: WeeChat 4.6.2) N0ZYC that seems like a more obvious solution N0ZYC but then why is there a such thing as a "gate driver ic", if an op amp works just as well? I've never used a gate driver ic before splud Also, your desire for 1kW loading is ... demanding. N0ZYC I assume the gate driver ic offers some benefit? the datasheet constantly is talking about "complete immunity from latch-up", but I don't know what that is splud you could use a series of power resistors. N0ZYC I have a fair collection of power resistors djph gate drivers keep you from torching an MCU pin from weirdo capacitive effects FETS can have as you start driving them "fast" N0ZYC I was using something like that previously. automotive brake lights actually. switched in 1, 2, 4, 8, and 16 amp series N0ZYC a bit of a pain to use. poor cooling. failed spectacularly on Friday ;) daugaard has joined (~daugaard@user/daugaard) splud think sort of like a binary progression, increasing values that you switch on or off (not PWM), and do trimming with a PWM'd FET (or just linearly driven). N0ZYC yep splud PWM or Linear. that FET is where the dissipation happens. N0ZYC I'd rather have a dial than a row of switches ;) splud dweeno does the switching. You still dial. N0ZYC yep splud 1kW is basically a space heater. splud and FTR, heater coils are a place to dump. N0ZYC my previous load: https://vftp.net/N0ZYC/hobbies/electronics/test%20equipment/electronic%20loads/N0ZYC%2012v%20DC%20load/index.html N0ZYC https://vftp.net/N0ZYC/hobbies/electronics/test%20equipment/electronic%20loads/N0ZYC%2012v%20DC%20load/bannerpic.JPG splud way more complexity, but less wasteful - battery charger with FET trimming. N0ZYC testing arrays of solar panels and also batteries Jesterboxboy has joined (~Thunderbi@195.140.136.207) N0ZYC I had it dumping 24 amps (@12v) for about 5 minutes the other day. something bought the farm. probably the relay N0ZYC long overdue to be replaced N0ZYC I built that from scrap around 1988 ;) kabe has joined (~kabe@user/kabe) N0ZYC anyway, getting back on topic. I'm wondering if I can just replace that IRFZ with one of my big mosfets, (and upgrade the current sensor of course) or is it going to require other things to be bulked up? djph sure, as long as you have the voltage to drive it djph (in general) splud "big mosfets" like? N0ZYC https://vftp.net/N0ZYC/hobbies/electronics/raw%20parts/mosfet/high%20current%20mosfets/STE180NE10%20180A%20100v%20360w%20power%20mosfet.pdf N0ZYC I have a pack of 6-8 of those Jesterboxboy has left IRC (Ping timeout: 252 seconds) N0ZYC I don't anticipate needing more than 2-3? N0ZYC they're rated 180 amps N0ZYC I suppose technically if I "only wanted 1kw, at 12v, just one (at 180 amps) would suffice splud remember, you'll need to wiring in there to handle that (and the connections to your box). Irony is, the mosfet has a very low Rds(on) -- this is ideally suited to switching power _TO_ something. N0ZYC since mosfets are voltage driven, not current, will this one require "more drive" than the IRFZ44? dbkb found another totally dead battery https://quad.pe/e/JLig8WDP4E.jpg djph threshold is apparently 3 volts, but seems "full on" is approaching 20 ... if I'm reading it right N0ZYC I think i have sufficient wiring to handle things N0ZYC yep i saw the 20v "max" djph you have a 100 amp power supply to drive this load? splud gate driver is about switching on and off FAST, so that the FET doesn't spend much time in the linear range. Gate charge is the figure from your fet you'd want to check. N0ZYC it's designed to *test* power supplies djph N0ZYC: and are you *TESTING* 100 amp power supplies? N0ZYC for me that means things like car batteries, 100Ah LIFE batteries, 1200w solar arrays, etc N0ZYC so yes I have enough power ;) N0ZYC my problem right now is my loads are *insufficient* for the power I am trying to test splud if the FET itself is supposed to be doing the dissipation, simply driving the gate in a linear fashion (i.e analog voltage, say, the output of an opamp) would do it. N0ZYC that's the plan joerg throws some DEATH batteries at N0ZYC kabe has left IRC (Quit: WeeChat 4.6.3) kabe has joined (~kabe@user/kabe) N0ZYC 1200w of portable solar power https://vftp.net/N0ZYC/power/solar%20power/my%20portable%20solar%20grid/evolution%20of%20solar%20power/full/2023.06.24%20Summer%20Field%20Day.JPG N0ZYC (I could do about double that in a pinch) splud IF you were using the fet to drive a load (say a large fairly low value resistor), that's be different - fast PWM switching would make more sense there so that the FET itself isn't dissipating much. joerg batteries suffocated in a Polymer pouch N0ZYC I have some big resistors but honestly the fet is easier to cool joerg the LiPo DEATH batteries N0ZYC that was one alternative I was considering tho N0ZYC I've got a box of 0.72 ohm 300w power resisistors splud eh, a big resistor has lots of surface area, or a nice Al housing. N0ZYC these have giant esposed ribbons joerg !battery electrobot ICR (Lithium Cobalt Oxide) 2.5V-3.7V-4.2V, IFR (Lithium Iron Phosphate) 2V-3.2V-3.65V, IMR (Lithium Manganese Oxide) 2.5V-3.9V-4.2V, NCR (Lithium Nickel Cobalt Aluminium Oxide) 3.0V-3.6V-4.3V, INR (Lithium Nickel Manganese Cobalt Oxide) 2.5V-3.6V-4.2V, LTO (Lithium Titanate) 1.6V-2.3V-2.8V joerg NOC what the "IFR" means, I call them LFP N0ZYC I thought I had pics of the resistors on my web site but I don't see them, hang on kabe has left IRC (Client Quit) nki has joined (~nki@193.5.234.59) kabe has joined (~kabe@user/kabe) splud https://i.imgur.com/k3Swg2z.jpg mercenary if a load is a PWM-pulsed mosfet, can one count that as purely resistive? splud pretty much, but it's not a _constant_ load. splud driving the FET linearly would be, splud https://i.imgur.com/XbT2ghq.jpg splud 250W 0.3Ω resistor mercenary for example, testing an SMPS, a PWM pulsed load can mess with the SMPS's control loop in horrible ways Cowclops has joined (Cowclops@user/cowclops) Cowclops Whewf, that was a fucking ODYSSEY of electrical investigation Cowclops disclaimer: not "asking for help how to fix my obvious problem" because the issue is not a think-out-loud about it one, its a "find whateve rsomebody screwed up in meat space" one splud that's basically an aluminium block with a hole bored through it, and a ceramic core with resistive wire wrapped around it terminated into the contacts on either end, filled with non conductive glass sand and either end capped with cement/mortar Cowclops I also think the following story may involve unrelated issues that coincidentally happened at the same time but having dealt with both in suhc a short time is baffling Cowclops So issue 1. Last night, I lost internet. The link light on my actual computer went out. "Shit, is the server in the basement on?" The GFCI outlet in the basement its plugged into, the GFCI was tripped but the breaker was on splud Cowclops - and when you say "somebody" you mean YOU, right? Cowclops ::unplugs stuff, hits reset on gfci, doesn't reset:: Cowclops No, I didn't wire anything in this house Cowclops Contractors/original builders Cowclops I'm quick to remind people that i am an engineer not an electrician and that gives me insight to contemplate what happened but i don't dig into electrical boxes and wire my own shit N0ZYC https://vftp.net/N0ZYC/hobbies/electronics/test%20equipment/electronic%20loads/variable%20loads/constructing%20my%20own/power%20resistors/index.html Cowclops So anyway N0ZYC spicy https://vftp.net/N0ZYC/hobbies/electronics/test%20equipment/electronic%20loads/variable%20loads/constructing%20my%20own/power%20resistors/full/order0a.jpeg Cowclops GFCI wouldn't reset. "ok somethings either tripping it for good reason or the GFCI has failed, i'll replace the GFCI and see if it keeps happening" then i went to bed. Cowclops i say i don't do my own wiring but installing one outlet isn't a big deal Cowclops and i'd be switching it to a leviton lever edge one that is thoroughly unfuckable N0ZYC each of those is good for about 200 watts at 12vdc N0ZYC (15 amps) Cowclops So I go to bed. When I get up, the lights in the UPSTAIRS bathroom won't turn on. splud N0ZYC - so, imerse in oil... Cowclops ::goes to breaker panel:: Breakers tripped. Huh. DIFFERENT breaker from that GFCI outlet, so seemingly unrelated?:: N0ZYC I'd imagine a good fan would work too Cowclops Turn breaker off. Turn breaker on.. BANG splud Cowclops - missinga 10K pullup, eh Cowclops arc flash ThePendulum oof hope you're okay Cowclops So i'm like well you don't reset a breaker more than once Cowclops Yeah not like, high voltage arc flash, but it arc'ed Cowclops flash is probably overselling it N0ZYC but whatevr the setup, it'd be less convenient and a lot larger than a compact heat sink with built in fans Cowclops And tripped immediately. Cowclops This breaker has a second breaker in the same bus bar slot, its one of those doubled up ones. Same story, i didn't install it mercenary N0ZYC: if you have a bunch of 300W resistors, you only need 300W variable load, plus one mosfet per resistor to add it to that load Cowclops It tripped THAT breaker too. Maybe from the heat or vibration? Cowclops Now the garage lights and door opener are dead, downstairs bathroom dead, upstairs bathroom still dead splud you still have to get a kW of heat out of your box. Cowclops "Well fuck. Time to call an electrician" N0ZYC yep Cowclops If that was the end of the story you'd say "Yeah you have an issue, a profesional that deals with this will trace it out so you don't have to get frustrated dealing with bad original wiring" Cowclops But wait there's one more bit Nerdmind has left IRC (Quit: WeeChat) N0ZYC I don't have a pic of the heat sink I'd like to use, but its about 8" x 8", and about a foot long. its meant to mount a fan on each end, for intake and exhaust. the fins are in the middle of the area, and the sides are flat, for mounting power transistors on Cowclops I'm like "That second breaker in the same breaker slot wasn't tripped when i got up and the screwed up circuit breaker blowing probably tripped its partner" Cowclops So i reset the "top" one. (The bottom one was the one that was tripped and resetting it blew it) Nerdmind has joined (~weechat@user/Nerdmind) Cowclops No arc, no trip Cowclops Garage lights on. Downstairs bathroom on. Upstairs bathroom on. EVERYTHING IS WORKING NOW, EVEN THE THINGS THAT DIDN'T WORK WHEN ONLY ONE BREAKER WAS TRIPPED?!?!? Cowclops i've got PRAHBLEMS Cowclops ...they're comin 8 am monday Cowclops To be clear, the suspect breaker IS still off mercenary you've got a 'creatively wired' place Cowclops I als odon't know what code is, but i have a 150 amp panel with quite a few "two breakers in one slot" Cowclops It might be one of those "Something is wrong AND there's too many 15 amp breakers on your 150 amp panel" splud N0ZYC https://i.imgur.com/cRP0q7M.jpg heatsink. splud though I have much larger ones here. N0ZYC I'd like something compact Cowclops The worst part about the breakres labelled "lights" is that most of them power a random assortment of switched outlets Cowclops like lets say you have one switched outlet for a floor lamp in every room N0ZYC https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcTx9rAfnH8b2t7hhZSjYjL32vJMkK-eZgGRDg2q_0emoi0sCZXBeq-29HdFNzLHthrjKgg&usqp=CAU Cowclops but one breaker is like, 4 different rooms Cowclops Its not overloaded, but describing what exact outlets are gonna shut off if you kill it is impossible Cowclops Its not "Main bedroom switched outlet." N0ZYC https://in.element14.com/productimages/standard/en_GB/462216906-40.jpg N0ZYC something like that N0ZYC breadbox with fans on both ends, fins inside, flats outside for mounting. nice and compact. good heat dissipation. splud Cowclops - if you have a tripped breaker, should be pretty easy to identify which rooms and outlets are inpacted by it. So draw a map of your house and mark outlets and fixtures by what is OFF while that breaker is inop. Cowclops Yep, thats the right answer, its just a lotta work and runnin up and down the stairs splud and at some other time, update it when another is toggled. Cowclops The issue is if you need a paragraph to describe what goes off when you flip a breaker N0ZYC my heat sink is bigger and older, but same idea Cowclops thats still annoying even if you have it documented splud er, no friends, family, housemate? Cowclops well one time i asked my wife to tell me when the light went out when i flipped a breaker splud you mean bigger and dirtier? Cowclops and i'm flipping a breaker ::silence::: Cowclops "Did it go out" Cowclops no Cowclops ok i'm gonna flip another breaker Cowclops .....i flipped it Cowclops ::no answer:: ChanServ has changed mode: +o litharge litharge has changed mode: +q $a:Cowclops splud well, sounds like you need to change your approach. litharge has changed mode: -o litharge N0ZYC maybe I need to come back and ask again when cow is done with his breakers splud also, you could throw em all, then clip a tone gen to the output side of various breakers and go inside to check outlets. splud or vice-versa - inject tone at the outlets, splud These are tools professional electricians would have invested in. djph I went and bought one just to deal with the wiring here at the house splud N0ZYC - ask _what_ specifically. Seems your spitballing on a project which needs to be scaled up from the one you're basing it on. ChanServ has changed mode: +o litharge litharge has changed mode: -qo $a:Cowclops litharge Skalar backflips breakers Birbasaurus has left IRC (Quit: Konversation terminated!) Skalar I'm talented like that. N0ZYC I just need to know WHAT needs to be scaled up. can I just scale up the mosfets? Cowclops https://www.reddit.com/r/homeowners/comments/12uby10/new_home_and_outlets_connected_to_wrong_locations/ not my post but this sounds like the situation i'm in. Birbasaurus has joined (~birb@user/Birbasaurus) splud Cowclops - you should be starting the process with a reasonable understanding of which outlets are controlled by wallswitches (not uncommonly one or two outlets in a room are wired to the wallswitch, if not simply a ceiling fixture, depends on age of house). N0ZYC is that one TC4420 going to be enough to drive multiple mosfets? mercenary Skalar: can you come work for our utility? took them 4 days to send someone out to backflip a breaker Cowclops Yeah, this is from 1993 and there are almost no fixed lights N0ZYC is it even CAPABLE of driving one of my big mosfets, or is the voltage too low? djph Cowclops: there's nothing inherently "wrong" with outlets being on the "wrong(tm)" breaker -- old houses was super common to have like "all bedrooms" on 1x 20 amp breaker... Cowclops every room was intended you to have a floor lamp plugged into an unganged outlet where one of the two plugs is always on and one is switched Cowclops Yep N0ZYC can I just replace it with a higher voltage op-amp? splud N0ZYC - If it were me, I'd be more prone to either run the digital output from the µC to inputs on multiple gate drivers, or even run separate digital outs to separate drivers. p0indexter has left IRC (Ping timeout: 260 seconds) splud the latter allowing you to do staged ouput djph Cowclops: when I get around to it here, I'm gonna label the wallplates N0ZYC I also have some old (20v iirc) "line drivers" that were originally designed to run the pins on a dot matrix printer. they're high impedence input, high drive output. (great for driving relays) splud djph - you can do that on the inside of the plate. splud (cover) p0indexter has joined (~alpha@user/p0indexter) djph splud: yeah, "when I get around to it" ;) Cowclops Yeah i think the correct answer is, don't get too heated about illogical outlet-to-breaker pairings if everything is working and safe Cowclops because there's a breaker that won't reset without arc'ing, "everything isn't working and safe" joerg N0ZYC: >>its meant to mount a fan on each end, for intake and exhaust.<< sounds like a receope for disaster, is the fan at exhaust side rated for 120°C? splud Cowclops - call an electrician, Cowclops Still need a pro with enough experience to sort this out quickly and safely and to code. And separately, make a lucid chart of every outlet in the house labelled with what breaker number it goes to Cowclops Yeah they're coming 8 on monday lol N0ZYC looking at some implementations, they specifically say DO NOT mount a fan at the exhaust, only at the intake, for that reason joerg :nod: splud afk N0ZYC the heat sink is a single block of extruded aluminum, so the two ends look identical, good for fan mounting. so yes I will probably only put a fan on the intake N0ZYC so the mosfets don't "mind" being used for the purpose of dumping power to heat? I don't see a "watts dissipation" rating on the datasheet like I normally see on big transistors joerg there's probably a thermal resistance rating djph Cowclops: sort "what" out? breaker replacement? mercenary wonders if wrapping one of those extruded heatsinks in kapton tape to force the air all the way past the fins would make matters better or worse splud before I go - perhaps rather than one ungodly beefy "portable" load, you make one 250W one, get it all sorted, and then make three more. Any one of them could be used on a bench on its own, or you can set up multiple. Extra points if you set up the micro to be able to manage subordinate units. splud now afk. N0ZYC the outside perimiter of the heat sink is flat solid aluminum, with a few holes in it for its previous use for mounting big mosfets to N0ZYC though it was probably being used for actual switching, not heat dump joerg mercenary: I assumed all 4 sides that are not "fan" are closed, so... kapton? N0ZYC when pulling potentially 100+ amps at 12v, do I need to have any sort of "filtering" for the PWM switching of the big fets? aren't they basically "shorting" the load to ground? mercenary depends on what it looks like. I saw a dummy-load the other day that was resistors bolted onto old CPU coolers, with a fan on each CPU cooler N0ZYC these STE Rds is listed as < 6 milli ohm mercenary N0ZYC: that's why I asked, can you still consider it a 'purely resistive' load N0ZYC (that'd be 2,000 amps at 12v0 N0ZYC I've looked at several schematics for electronic loads but I don't think I've ever seen any sort of "smoothing" network on the load joerg you certainly do NOT want to drive those monster FETs with the contact screwblocks in PWM when you want the FETs to dissipate all the energy N0ZYC yet this seems to be what great scott is doing? https://vftp.net/N0ZYC/hobbies/electronics/test%20equipment/electronic%20loads/variable%20loads/constructing%20my%20own/Great%20Scott%20adjustable%20load%20project/full/schematic.png joerg what you want to do is have a PWM smoothed to feed a clean DC bias to the FET gates N0ZYC just plunk that fet right between the supply and ground joerg yes, exactly N0ZYC ok that's probably what R1/C3 are for then joerg and operate the FET(s) linearly N0ZYC smoothing the PWM out a bit Cowclops djph: Well different things. I definitely want them to sort out why the breaker is tripping and arcing Cowclops I don't think i need them to sort out why breakres go to weird combinations of rooms Cowclops on different floors joerg note your monster fets themselves got quite some gate capacitance p0indexter has left IRC (Quit: Reconnecting) p0indexter has joined (~alpha@user/p0indexter) N0ZYC would it make more sense to tie in a D-to-A converter on the arduino then, to supply an actual DC voltage directly to the fets? joerg ((schematics)) oh yes, nit 470uF, they got 28nF or somesuch joerg if you got some, sure mercenary 470uF would make it very-low-pass N0ZYC I haven't worked much with large FETs, my last project with them was to make a poor-mans charge guard N0ZYC I ended up slowly degrading the fet, I suspect because I was driving it in its linear range. I don't want to fry the fets here for the same reason joerg LOL, he's using a mosfet gate driver (TC4420) to drive a 10R-470uF lowpass filter N0ZYC why is that bad? joerg it's not exactly bad, just a tad weird mercenary it'll do the job N0ZYC I wasn't sure if he was just using it to amplify the current from the mcu to the fet mercenary Same as pwm through low-pass into an opamp buffer to driver the fet joerg it'l do, yes. Nevertheless I probably had used an OPAMP LPF that does both the filtering/integrating and the driving joerg the FET gate is voltage controlled, no huge current needed for slow variations of the voltage N0ZYC is it harmful to run a big fet like that in its linear region? joerg not per se mercenary As long as you can drop the control voltage quickly enough when current goes up joerg maybe not the usual way to use them, maybe not very comprehensively specified in datasheet, but not "intrinsically bad" VanUnamed chatgpt says lifepo4 cells are 3.2v nominal and 3.65v charged VanUnamed and so to make 48v i need 16, not 14 like in the case of 18650s VanUnamed accurate? joerg mercenary: he's defeating exactly this by using a MONSTER 10R-470uF LPF mercenary joerg: that is what I would be afraid of yes joerg which is just doing one thing: burn power from the 12V voltage supply mercenary that's what, 34Hz if my math maths out N0ZYC arduino can supply pwm in a variety of frequencies N0ZYC though they start in the 2khz or so range? mercenary yeah. but to drain that 470uF cap through 10R takes a finite, non-zero amount of time VanUnamed and it also says i need a bms. Jesterboxboy has joined (~Thunderbi@195.140.136.207) mercenary VanUnamed: you need a bms N0ZYC so that RC us just a cheap way of converting the PWM into a fairly stable DC level mercenary yes N0ZYC and with such a large c it'll be slow joerg https://www.microchip.com/en-us/product/tc4420 *cough* N0ZYC slow to react to changes from the mcu VanUnamed even for those big 100Ah cells, i need a bms? mercenary Unless the cells have a built-in bms N0ZYC in anticipation of this project I picked up some similar tc4451 https://vftp.net/N0ZYC/hobbies/electronics/test%20equipment/electronic%20loads/variable%20loads/constructing%20my%20own/TC4451%20gate%20driver.pdf N0ZYC I was hoping they'd be a suitable replacement in that circuit N0ZYC just a bit higher current, not that this should matter here? joerg don't get me wrong, the whole circuit will work fine, just some design decisions in it that I don't understand and would do differently N0ZYC I don't see why they're rated for such high current when driving the gate of a mosfet... current really isn't important for gate drive? VanUnamed my apc3000 works though. hjf_ cleaned up the computer today hjf_ it was clogged with lint lol Jesterboxboy has left IRC (Ping timeout: 252 seconds) Matt|home has left IRC (Quit: Matt|home) hjf_ last time i cleaned it was like 1 year ago N0ZYC I'm not "formally educated" in electronics, so for a lot of things I prefer to just find someone else's successful project and go with that, I'm not so good at designing my own things, and often when I see something like this I don't fully understand why things are done the way they were, I just "go with it" joerg the TC4420 (at 5V Vcc) will burn an average of 5V/2 / 10R = 250mA, for absolutely nothing mercenary N0ZYC: high current is needed to drive the gate _fast_, when you are doing 1MHz PWM N0ZYC but not through THAT rc network? N0ZYC so it looks like his use of a gate driver, with a fat RC network, was kind of a waste of the gate driver? joerg waste of gate driver and waste of power mercenary somewhat. you can't drive the gate directly from the mcu, so needs some form of buffer, and if that is a gate-driver, an opamp, or a bjt, doesn't matter for this application N0ZYC I'm not very good with op-amps, but I assume one could be used as a "buffer" (in zero-gain configuration) to connect the micro to that rc network instead? joerg yes joerg unity-gain N0ZYC but the voltage from the MCU is probably going to be too low to fully drive my mosfets mercenary also means you can put the lowpass before the opamp, without monster capacitors joerg and no, you'd use a 470kR-470nF LPF *before* that unity ganin buffer N0ZYC but that would remove the filtering, and cause the mosfets to try to follow the fast transitions of the pwm? joerg use a 470kR-470nF LPF N0ZYC oh ok filter before the op amp joerg sorry use a 470R-470nF LPF N0ZYC higher resistance same same cap so slower response Skalar yeah, you dont want particles getting into the op amp N0ZYC o_O mercenary with enough gain, 1 particle goes in, 100 particles come out, and the roomba overloads N0ZYC I think I have some higher voltage op amps here. (creating the 20v for them will be slightly annoying) I assume I can't wire the op amp in "unity" gain to get a 5v micro to drive a 20v op amp? joerg no, higher resistanc lower capacitor, you could go much further to the PWM frequency and thus gain speed in control loop, without wasting 200mA. You even could use a decent s&h instead of a LPF and get speed in the kHz range N0ZYC this is just a load tester, it doesn't need uber fast response time joerg exactly, it does some implicit current regulation in the way the FET works in linear mode N0ZYC btw, thank you all for the help here, I realize a lot of this is very elementary for you but I'm getting out of my comfort zone and your ideas and feedback are really helpful joerg but e.g. for impedance sp4ectrum tests of batteries, you might appreciate a fast adjusting loads joerg yw siak has joined (~quassel@user/siak) splud (haven't read 25m of scrollback yet) Blondie101010 what a slow reader! N0ZYC what's a "s&h" ? a different design of op amp filter? splud joerg - yea, if the fets themselves are supposed to do the dissipation, there's not a lot to gain from the PWM. µC should be monitoring current from the hall effect, and pwming through an RC into an opamp that drives the FET gate. N0ZYC so the pwm is already DC before it gets to the op amp kabe has left IRC (Quit: WeeChat 4.6.3) ftg has joined (~ftg@fwa-access-567380-203.dhcp.inet.fi) splud filtered PWM into opamp gives the opamp basically a DC signal, and the opamp is high impedance input, so virtually no load on the µC output. EconoDillo has joined (~Kurt@2605:a601:aacf:2800:98f0:a2bb:6996:70c9) N0ZYC iirc the arduino is limited to 50mA per dpin N0ZYC I have a few A-to-D (SPI controlled) converters with fairly good resolution, I may use those instead of PWM splud and skimmed backscroll up to current, and I see that RC before opamp has again been suggested. splud Got a link to the GreatScott video from which you nabbed the schematic? N0ZYC is it more sensible to just RC and buffer the pwm, or to use a D2A chip before the buffer? splud PWM is a little code setup and ONE pin from your µC N0ZYC https://www.youtube.com/watch?v=VwCHtwskzLA electrobot N0ZYC just linked to DIY Adjustable Constant Load (Current & Power) - GreatScott! akibageek2 has joined (~akibageek@103.155.232.216) splud RC values should be matched to your frequency, but that and a quad opamp would allow you to drive 4 fets separately (with four digital IO). Lord_of_Life has left IRC (Ping timeout: 272 seconds) Lord_of_Life has joined (~Lord@user/lord-of-life/x-2819915) N0ZYC I've seen in radio where you can't drive multiple BJTs in parallel at frequency due to capacitance adding up, but I've seen mosfets commonly driven in parallel without problem N0ZYC is there a reason I would want to drive each mosfet separately? password man ultrawide monitors is some of the best marketing ever joerg N0ZYC: Sample&Hold joerg https://en.wikipedia.org/wiki/Sample_and_hold N0ZYC oh I just noticed on the spec sheet for these, total dispipation is 360w N0ZYC so I will be using several splud thx for the link, going to check it. Oh, yea, I've seen this. splud https://youtu.be/VwCHtwskzLA?t=185 he talks about ohmic vs. saturation. electrobot splud just linked to DIY Adjustable Constant Load (Current & Power) - GreatScott! N0ZYC I like great scott's builds, they're easy to follow and well-explained Skalar in ze german accent, ja N0ZYC ja N0ZYC I suppose if I budget for 250w per device, four for 1kw akibageek2 has left IRC (Quit: WeeChat 4.6.2) joerg I *think* you could replace/augment the low-pass filter by a s&h stage driven from either the rising or the falling edge of the PWM N0ZYC I'm not familiar with "Sample&Hold" filters ID-10-T has left IRC (Remote host closed the connection) akibageek2 has joined (~akibageek@103.155.232.216) mnrmnaugh has joined (~mnrmnaugh@static-100-0-45-11.washdc.fios.verizon.net) N0ZYC (random though) I wonder if anyone has just converted a hair dryer to an electronic load joerg N0ZYC: it's sort of over-engineering for this particular case. Sorry, nevermind joerg a simple R-C LPF PWM from MCU GPIO and buffering with a unity-gain amp will do joerg Scott did it the other way around: "amp" (aka gate driver TC4420) and *then* R-C LPfilter. A tad weird roxfan has left IRC (Ping timeout: 268 seconds) akibageek has left IRC () splud N0ZYC - potentially different FET parts, or variations, granted if you're constantly monitoring and adjusting, variations might not be a big deal. Phantom "I wonder how humid it is outside after this rain" . . . 100% "oh..." nullcone has joined (uid11626@id-11626.helmsley.irccloud.com) N0ZYC I'm thinking I will need a higher voltage op amp, something that can run at 20v or a little more, and set the gain on the op amps to 4:1, so I can get the 5v pwm max on the micro up to the 20v the STE mosfets appear to wanr? joerg absolutely correct approach N0ZYC I assume that's how the op amps work splud As I mentioned earlier, you could have the FETS switching separate LOADS, so the FETs themselves aren't doing the dissipation. Then your programmable loads become a programmable current limiter. N0ZYC one of these days I'm going to read that op amp book I have here :P splud 20V isn't really a "high voltage" opamp. N0ZYC using loads is an option, but it brings up two issues N0ZYC first, it means things get physically larger. room for big resistors, larger mounting hardware, different cooling N0ZYC (I want this to be portable) splud bog stock LM358 runs 3-32V on single supply config. N0ZYC and second, I'm a bit limited with the resistors I have. 0.73 ohms at 300w, and only six of them. on the other hand, I have a handful of those STE mosfets N0ZYC is that 358 rail to rail? will that matter? mnrmnaugh has left IRC (Quit: Client closed) joerg I N0ZYC your STE180BE10 needs only 10V Vgs. 20V is ABSMAX where it breaks N0ZYC ah mercenary 358 isn't rail to rail iirc, it will only matter for 'how close to 0V can you get to keep the fet closed' N0ZYC what should I consider the "high end", for full switching? N0ZYC full-on splud the drive circuit has feedback - the current sensor. If you have a set of banana plugs for + and - on high side of fet (still mounted to heatsink), you could have an external load plugged in (current regulated), or you could have a beefy jumper wire (= FET is the load) joerg 10V N0ZYC ah ok splud mercenary - sorrry, wasn't specifically suggesting the 358 as the device to use, just pointing out, the voltage range isn't expecially "high" for an opamp Phantom LMV321IDBVR splud also, the opamp is driving the mosfet gate, and is not itself in series with the load.. Phantom rail to rail 5V N0ZYC arduino analog input are pretty poor. better to use a current measuring accessory that has SPI output N0ZYC I have some that can handle this N0ZYC re: rail to rail I just need to make sure the mosfets can turn ALL the way off mercenary briefly considers LM386 as a fet driver, and rejects the idea joerg N0ZYC: for the current sensor you'll need sth other than a ACS712 :-) joerg if you want to do a 100s of Ampere, not <10A splud insofar as rail to rail - the low range of the mosfet Vgs(th) is all you need to be able to drive down to. if your mosfet isn' splud isn't even starting to switch on at 0.2V, does your Opamp need to go lower? akibageek2 has left IRC (Quit: WeeChat 4.6.2) splud and fucking AGAIN, my reference to the 358 is "even that part runs to 32V" N0ZYC I could get away with a ACS712 by simply peeling off a smaller wire in parallel with the main delivery wire, to get a small percentage of the current joerg yes, you could, when you calibrate that contraption afterwards splud ALL the desired specs should be laid out. If you're doing 1kW from 12.0V (let's not call it "12V"), that is 83.333 A. akibageek has joined (~akibageek@103.155.232.216) splud do you need 1kW very often? You're making the basic build much more costly. joerg splud: I gether N0ZYC got shy of a dozen of those https://vftp.net/N0ZYC/hobbies/electronics/raw%20parts/mosfet/high%20current%20mosfets/STE180NE10%20180A%20100v%20360w%20power%20mosfet.pdf and plans doing a pimped up version of https://vftp.net/N0ZYC/hobbies/electronics/test%20equipment/electronic%20loads/variable%20loads/constructing%20my%20own/Great%20Scott%20adjustable%20load%20project/full/schematic.png with them Friithian ⸘1kW@83A‽ Famine N0ZYC, you shouldn't be using standard fets as loads anyways they tend to run away. you need linear fets splud the 712 isn't the appropriate sensor for this project not just because it won't satisfy the current (without annoying calibration), but it is for bidirectional current (thus the near 2.5V output when at 0A) joerg ooooh another not-so-great scott decision? ;-D dbkb heavy joerg Famine: you got some link to a source with more details about that run-away? joerg so far thought FETs were generally inherently run-away safe Famine joerg, https://www.nexperia.com/applications/interactive-app-notes/IAN50006_Power_MOSFETs_in_linear_mode N0ZYC their gain is fairly stable at temp? N0ZYC and yeah I don't like that 712 since its not a digital output, and the arduino analog inputs are crap splud Allegro has ACS724LLCTR-40AU-T which is a +40A part. (vs ±10 or whatever the specific version of the ACS712 in the Great Scott project is - there are ±5 ±20 and ±30A versions, depending on full PN) N0ZYC I've got SPI interfacing devices here I can use Jesterboxboy has joined (~Thunderbi@195.140.136.207) Famine joerg, also https://www.infineon.com/dgdl/Infineon-ApplicationNote_Linear_Mode_Operation_Safe_Operation_Diagram_MOSFETs-ApplicationNotes-v01_00-EN.pdf N0ZYC and branching a something like a single #10 wire off in parallel with the load line to the fets will produce a scaled down amount of current that will allow me to use lower current measuring devices splud also 722 series. As you get parts with increasing current measurement capability, your mV-per-A drops, and with it, your resolution. N0ZYC I don't need 1mA res. I have smaller portable load testers. this one i will be fine with 1 amp res N0ZYC what I need is high power splud just remember, when you use something "I have here", someone will look as your resulting design and ask "why'd they use THAT?", just as with the gate driver and post-driver RC here... N0ZYC :) Elw3 has joined (~Elw3@user/Elw3) N0ZYC I know, "when all you have is a hammer, everything looks like a nail". But that's why I've accumulated suitable parts for this project splud how long do you figure you'd be running the load? N0ZYC I'd like to expect up to a 5 minute runtime, but design it for as long as 10 Jesterboxboy has left IRC (Ping timeout: 252 seconds) splud that has a potential impact on your cooling requirements. N0ZYC typical ceiling of 1kw. I like to over-engineer though so 1.5 to 2kw max if possible and practical N0ZYC I only want to make this once N0ZYC I figure the heat sink I have, with the fan on full, should be able to keep up with 1kw N0ZYC probably 2 for awhile akibageek has left IRC (Quit: WeeChat 4.6.2) N0ZYC it'll be a hair dryer tho ;) joerg Famine: many thanks! TIL :-) >>However, this phenomenon can be avoided. In fact, for a given VDS, there is a critical current above which there is a negative feedback and thus thermal stability. This is known as the Zero Temperature Coefficient (ZTC) point.< akibageek has joined (~akibageek@103.155.232.216) N0ZYC so will I need a rail to rail op amp to insure the mosfets can turn fully off? akibageek has left IRC (Client Quit) akibageek has joined (~akibageek@103.155.232.216) joerg so for small currents there might be a positivve feedback of temperature to Ids, but it will stop at ZCP N0ZYC zcp? splud hrm, the infineon URL redirects to their homepage. No "not found" or anysuch error. joerg sorry ZTC Famine bleh stupid infineon https://www.infineon.com/dgdl/Infineon-ApplicationNote_Linear_Mode_Operation_Safe_Operation_Diagram_MOSFETs-ApplicationNotes-v01_00-EN.pdf?fileId=db3a30433e30e4bf013e3646e9381200 N0ZYC I don't see that in the datasheet for the STE joerg N0ZYC: nevermind, it's pretty much irrelevant to you N0ZYC k akibageek has left IRC (Client Quit) splud wget to the rescue.. mercenary looking at that datasheet, fig.9 vs 10, whichever effect wins akibageek has joined (~akibageek@103.155.232.216) N0ZYC as long as the op amp can get below cutoff I don't mind joerg N0ZYC: for absolute good measure, you *could* add a 50 or 100 milliΩ resistor to the source of each of your paralleled STE N0ZYC I've seen that before in pass transistors in linear power supplies joerg for BJTs it's normal N0ZYC those are going to be chunky joerg they are much more prone to fatal thermal runaway splud wonders why we don't see more 1kW programmable loads N0ZYC ballparking things, if I aim for 1kw nominal at 12v, thats around 20 amps per STE if I have four. N0ZYC the 400 I have was a cheap model and was still rather pricey. I can't imagine what the cost would be for a 1-2kw electronic load. they're unreasonably expensive N0ZYC I think that's why there are so many DIY projects on youtube joerg I'd *think* your STE180Ne10 will bever get so much out of sync by different temperature that one would draw so much more current than the others that it burns out joerg never, even Famine joerg, lol just for fun https://ntrs.nasa.gov/api/citations/20100014777/downloads/20100014777.pdf infisc has left IRC (Quit: Connection closed for inactivity) N0ZYC for the upper limit of 40 amps per STE, a 50 milli ohm resistor would produce 80w nki has left IRC (Ping timeout: 276 seconds) splud DISSIPATE 80W joerg Famine: you really make my weekend, many thanks :-) splud your PV module might be producing 80W splud I'd consider using an ESP32 for the programmable load - then you could have an API, graphing, logging... N0ZYC I can do that with arduino splud there are models, yes. N0ZYC I'm used to getting high performance out of low end hardware. I'm in my element with arduino splud not the one in the GreatScott project. N0ZYC yep, even that one splud Nano doesn't have wifi. N0ZYC tho if I was going to go ham on it I'd use a mega2560 with a 320x240 touch panel N0ZYC for this I will probably use a button touch shield on an uno. 16x2 with full menu navigation keypad N0ZYC button KEYPAD shield that is N0ZYC >> your PV module might be producing 80W System_Error has left IRC (Ping timeout: 264 seconds) N0ZYC I think you might have missed https://vftp.net/N0ZYC/power/solar%20power/my%20portable%20solar%20grid/evolution%20of%20solar%20power/full/2023.06.24%20Summer%20Field%20Day.JPG N0ZYC each of those four racks is rated 400w, though typical max output in iowa is more like 300w N0ZYC I don't mean to test panels or USB power banks. I mean to test solar arrays and 100AH batteries r3trograde has joined (~cumbayah@user/cumbayah) mercenary Now there is an application note I will take seriously. "This means that MOSFETs of previous technology mercenary generations and/or higher voltage classes will be more suitable for this kind of application". If it admits latest is not always greatest, it hasn't been 'redacted' by sales siak has left IRC (Ping timeout: 265 seconds) siak has joined (~quassel@user/siak) hentai What are your "generic" connectors of choice? I soldered XT30s to so many things... LibertyDillo has joined (~Kurt@136.49.186.179) N0ZYC running some numbers now to see what sort of max values I'll have N0ZYC hentai: depends on several things. max current, ease of connection, interoperability with others hentai Almost everything where I had the opportunity to change the power port got XT30 now N0ZYC I use Anderson Powerpoles. they are unisex, so no need to keep male and female sets. N0ZYC they are also the ARES standard for emergency situations System_Error has joined (~SystemErr@user/systemerror) N0ZYC for higher current, powerpoles are available in up to 350 amp versions hentai Interesting N0ZYC SB50 style are very common in automotive use for things like winches and blades. larger powerpoles are used for things like forklift batteries N0ZYC XT have the advantage of being very compact, they're well-suited for use in RC toys N0ZYC cars and now quadracopters are using them m00se what's the link to that circuit diagraming site you guys like? I'm blanking mnrmnaugh has joined (~mnrmnaugh@static-100-0-45-11.washdc.fios.verizon.net) N0ZYC powerpoles (15-45 amp) can also be color coded and also keyed, to prevent accidental mismatch N0ZYC and can be bulked up by built in paralleling for increased current if needed m00se comon people help me out here N0ZYC lush projects thing? the simulator? m00se I got a suicidal nerd on discord that's gonna off himself if he can't figure out how to wire these leds m00se no m00se omg you all use it like 500 times a day N0ZYC the lush diagrams are good for showing off stuff m00se jaggz m00se joerg? N0ZYC lush is hosted by A LOT of places. you're probably thinking of one of them m00se mnrmnaugh you m00se you're smart what's the name of that circuit diagraming site joerg huh? N0ZYC falstad m00se thank you N0ZYC https://www.falstad.com/circuit/ m00se falstad - thanks m00se :) N0ZYC np m00se you get a cookie! N0ZYC mmmmm coooooookie m00se (one you can't see that gives me your CC information) N0ZYC C is for Cookie, that's good enough for me! joerg http://www.falstad.com/circuit/circuitjs.html edgar-rft CC is Crunchy Cookies :-) N0ZYC crypto cookies joerg https://www.falstad.com/circuit/circuitjs.html?startCircuit=tlterm.txt&hideMenu=true Matt|home has joined (~Matt@2601:580:c005:5a10:a077:eacc:43a9:bac9) joerg m00se: https://tinyurl.com/y8tokwmv wiring LEDs Krock spice is better for simulations electrobot joerg just linked to www.falstad.com Krock runs darsie What's the purpose of the 'PUSH' button? https://dl.imgdrop.io/file/aed8b140-8472-4813-922b-7ce35ef93c9e/2025/05/17/2025-05-17-202429_1680x1050_scrotbd60a653f712de6d.png dbkb its for pushing darsie thx dbkb looks like ai image darsie, so ask the ai darsie ic darsie It's a screenshot of a short video in a video. mercenary digital coin toss. Either it stops the countdown, or it fast-forwards it to 0. 50/50 chance. dbkb ai can make videos Krock darsie: instant explosion darsie mercenary: I guess it only stop the countdown when pushed within the last 250 ms. Krock that's also an interesting arrangement of semi-ICs splud N0ZYC - a resistor doesn't _produce_ power was my point, not that your PV modules are rated at 80W. They convert light into electrical energy. mercenary very AI-looking indeed. Also, which bomb manual started the tradition that wires must be coiled? N0ZYC I was just looking at what size (wattage) resistor was going to be required mercenary the resistor produces 80W heat otr has joined (~b@user/abbacis) splud mercenary - I stated as _dissipates 80W_ m00se thsat's the Dildo200 rev b mercenary splud: that is production of heat, out of current raw material ;) splud mnrmnaugh - it's plumbing, part of a faucet assembly. Nothing STANDARD about anything like that. N0ZYC it does seem a bit silly to have the resistor producing a significant fraction of the load Elw3 has left IRC (Remote host closed the connection) splud every manufacturer will have their own take on it. splud N0ZYC - not arguing that. N0ZYC is the purpose of that drain resistor to prevent too high of a current spike? N0ZYC / to provide an upper limit to the max instantaneous current on the fet splud 1. that's not steel braid, it's nylon. 2. that connector likely RETAINS the bottom portion of the faucet which is on the topside of the sink. mercenary to prevent one fet taking more load than another one N0ZYC for balancing ok N0ZYC if I use four of them it will probably be a bit overkill splud N0ZYC - your code could "soft start" and/or step to a new setting. mbiscuit has joined (~mart@user/mbiscuit) N0ZYC I will be implementing that, with inertia mercenary HexaCube-level over-engineering: 4 parallel FETs, with individual current monitor, and balance them in software ;) BillyZane2 has joined (~BillyZane@user/BillyZane) BillyZane has left IRC (Read error: Connection reset by peer) splud HexaCube-level is "wait until there's parts on the ground, then investigate the source of that squealing noise. Skalar https://gallery.skalar.org/upload/2024/01/13/20240113120112-6871d05a.png mercenary no, that is my method. hexacube watches for smoke signals Skalar hexacube's bmw N0ZYC you're assuming it even *starts* mercenary rule one of vehicles: if it makes a noise, that's good, it means it hasn't fallen off the car yet splud If the engine bucks, that means your harmonic balancer has sheared. N0ZYC if I have four STE's with 100mΩ on their drain, at 12v and dead short STEs thats only 360w B-| has left IRC (Ping timeout: 252 seconds) N0ZYC will definitely need to use smaller R value on those resistors splud switches on a bunch of automotive lamps. N0ZYC 2 amps per filament, and cheap! they served me well for yers B-| has joined (~b-pipe@user/B-/x-9249737) N0ZYC it's a bit surreal dealing with resistors with such low values. reminds me of the opposite with large resistors on tubes. 10k 100w resistors, what are you ...oh... yeah, high voltage, right. N0ZYC I suppose at this stage the wires leading to the gate may as well be the balancing resistors mercenary That's nvidia-style engineering mercenary It will work though, if you don't use ridiculously heavy wire N0ZYC and maybe opt for the asbestos jacket ;) infornography has joined (~infornogr@mobile-166-177-121-182.mycingular.net) darsie has left IRC (Quit: Avoid fossil fuels and animal products. Have no/fewer children. Protest, elect sane politicians. Invest ecologically.) EconoDillo has left IRC (Quit: ZZZzzz…) darsie has joined (~darsie@84-113-82-174.cable.dynamic.surfer.at) N0ZYC looking through a box of 10 ohm 100 watt resistors.... what's this? oddball. 100 watts, 470k splud all the cool guys had asbestos jackets in the 50's Now, there's endless Mesothelioma lawyer adverts on teevee. roxfan has joined (~roxfan@2a02:a03f:a2be:4f00:956f:25ed:f22d:48cd) N0ZYC that'd require 6.8kV to max out siak has left IRC (Remote host closed the connection) siak has joined (~quassel@user/siak) xx has left IRC (Ping timeout: 264 seconds) Turn_Left has joined (~Matthew@host86-136-206-155.range86-136.btcentralplus.com) splud Rando caramic Ohmmite from my shelf: 500Ω, 225W. It's about a foot long, and 1.5" or so in diameter. muffindrake What, did they use the asbestos jackets as bulletproof vests? Was the asbestos not contained inside other fabric? splud correction, 10.5" and 1.125" xx has joined (~xx@user/xx) splud asbestos was a popular insulation material some decades back. splud also "popcorn" ceiling, splud insulating texture spray. What fun. splud but not applied in such a thickness tyhat there was really much insulative quality. Left_Turn has left IRC (Ping timeout: 248 seconds) N0ZYC asbesdos was used most commonly for fireproofing N0ZYC and insulation N0ZYC (thermal) N0ZYC also used in brake pads for quite some time N0ZYC friend of mine died awhile back from lung problems, from his days working as a brake mechanic in a car shop. brake dust gets everywhere, they'd just grab the compressor hose and blast it off the drum with air muffindrake Is PPE not in the car mechanics culture? N0ZYC breathed in loads of asbesdos brake dust daily for decades N0ZYC notback then it wasn't muffindrake Like, they are dealing with machines covered in tire debris and god knows what else, eight hours a day N0ZYC they washed their hands in that what is it... TPFE or whatever, to get the grease off. now you can't get near that without a space suit Skalar yep. Skalar https://pbs.twimg.com/media/GHccDqpWoAAdfbl?format=jpg N0ZYC it's quite amazing how crazy things were back in the 50s Skalar this happened in the 50s edgar-rft everything was much better in the old times ... muffindrake excuse me what the fuck N0ZYC my mom said when she was in high school the science teacher got out a big glass bottle of mercury and poured it out into an open tub. the kids circled around it and scooped it up into their hands and poured it from one hand to the other Skalar yup N0ZYC today somebody drops a thermemiter and the EPA seals off the building Skalar my old thermostat for the A/C used mercury bulb switches Skalar still have it N0ZYC I've got a few mercury switches around here somewhere N0ZYC 99 cents at radio shack N0ZYC loved them N0ZYC though I think today's culture goes a bit overboard N0ZYC you want dangerous, try *organic* mercury N0ZYC I don't even want to be in the same ROOM as that stuff N0ZYC a few drops of that just roll off your hand, give it 6 months, you'll have a room temp IQ. give it another year and you'll have the mental powers of a 3 yr old N0ZYC by the time anyone realizes there's a problem, there's no saving throw Skalar horrifying way to go.. N0ZYC indeed hackkitten waddles about with some methylmercury~ Skalar metallic mercury isn't a huge disaster, open a window & clean it up without creating a lot of dust or touching it, hackkitten trips (o/ hackkitten kills half the channel ^-^;; hackkitten oops~ splud N0ZYC - Celcius Room Temp ? N0ZYC F N0ZYC that stuff just *wrecks* the brain. and it's SLOW N0ZYC at least you have time to get your affairs in order Skalar Home Affairs, Financial Affairs, Extramaritial Affairs splud an Affair to Remember Birbasaurus2 has joined (~Birbasaur@user/Birbasaurus) m00se state fairs m00se don't forget state fairs!!!! password rip Jesterboxboy has joined (~Thunderbi@195.140.136.207) ziper3 has joined (~ziper2@host-216-182-254-251.wlb.terranova.net) siak has left IRC (Remote host closed the connection) siak has joined (~quassel@user/siak) N0ZYC love affairs Jesterboxboy has left IRC (Ping timeout: 244 seconds) ziper2 has left IRC (Ping timeout: 252 seconds) ring0_starr hackkitten: methylmercury isn't *that* bad. there are standardized procedures for cleaning it and removing it from the body. Birbasaurus2 has left IRC (Remote host closed the connection) ring0_starr dimethylmercury, on the other hand... cannot bind via chelation ring0_starr so it accumulates in fatty tissues. like the brain. splud Section 80 of the Coroner's Handbook. 14:44 bleb has left IRC (Ping timeout: 245 seconds)