Help me build a quality, datalogging, charging/discharge bank @ home, DIY.

Please? High power preferably?
Thankyou

you need to give us more. right now you are sounding like "i need a new laptop, which one should i get?" and expecting us to magically know what you want to do exactly.

Well, I know friend. Just tr5ying to drum up interest in common questions pertaining to my build. Thankyou sir.

So the new : "labtop:" ideas are as follows...: Brainstorming, with the help of yall for saftey and knowlege, experience.

Goal: 24-240v charge and discharge capable. That is 6s - 56s capable. On Datalog. Resistive heating load bank. AH, wH, A, V, temp, pressure, datalog over time. Variable current, either through individual setup resistance or some (other? ) way to throttle current. Lots of current. Goal would be 200A continious, 500A + to begin.

I have:

(14) 50Ohm 240v 1250w resisters. To parallel and or series for certain current load bank.
(6) 33 Ohm 240v 1700w resisters.

A Cycle Analyst, V3, and an Analogger. This gives me all data on a .csv file for easy perusal later.

I also have a Chineese 300A hall sensor battery testing panel, the one with all the options on the right. Dont know how to record data from it. I will use this primarily for secondary data, like a cross check to make sure everything is accurate while testing. Would like to some how intergrate this into the piggyback on the CA data.

A 3000w power supply capable of 240v Ac input and 240 V DC output. Over 60A @ 5v or.... you know. Like 12A max @ 240V DC out.
Relays. Transistors. Ect. A 4-600A emergency disconnect. Fuses everywhere, using DIN rail and common rated propper fuses and breakers for secondary.

Shunt with the CA, probally a 200A or bigger. Maybe much bigger.


Questions: Should I go for 125v max as a beginner goal ( to build and test the load bank?) and forgo the 200+ HighV stuff?

Question: How can I vary current through the Cycle analyst to throttle the amperage draw? I dont know of a rheostat big enough? My engineer Father told me to use a big big Knife switch. Lol.

Anyone wire up a stress strain gauge to the CA3 or Analogger? Got data? I got a few, have not figured which one I want to use yet.


So what am I capable of, when it comes strictly to the design criteria? What would you like as a DIY battery builder? Thanks for the humoring me and the critique.

Primary goals, recreational EV in the 3-20Kw + range, and the current testing capability of each.

So it sounds like you are not asking about a battery bank (synonym for "pack").

But for advice about setting up a testing bench, related "lab" equipment?

Asking specific, precisely worded questions will help you get better answers. . .

Maybe similar to this?

https://endless-sphere.com/forums/viewtopic.php?t=103013

Hi john. I know that thread. Just brainstorming here.

This is a Load "Bank" I want to design.

I mean, Grin Ebikes sells a wimpy one, I figure you guys would want to have one with a little more power. Grins is for like 36v lol. Like 1Kw? Mebbe? Here let me find a link:

https://www.ebikes.ca/shop/electric-bicycle-parts/test-equipment/dual-resistor-battery-load-bank.html

https://www.ebikes.ca/product-info/grinspector.html#loads

35A max ... Nah. I want to see 20Kw flowing through my ( voltage and power rated ) load BANK.
yeah not battery bank ( lol) but a load bank. Bank of high power resisters. I mean, at 240v

i want a machine that I can put any size battery on, and do a discharge and charge, and know exactly Wh, temp rise, Ah, V drop at certain loads, IR, ect.

Then move on to the next battery and doit again. Save it to a .csv file, date it, and go.

just feeling the water and looking if anyone has done it before ( that I know you have.. ) and not using hair dryers. I will be using large ceramic resistance bank on a server rack.

So im thinking 24v control voltage for relay, contactores, fans, switches, ect.. and 240v max battery datalog capable. HV system isolated from the LV system. 240v-110v ACinput ( probably 220Ac.. ) . Things to stop the current if it goes to high, and the like.

i got a powerlab for cell level tests. This s for cells in series ( ie battery) and monitoring the empirical heat created during a certain discharge of a certain amount. BMS controlled battery or not.

Aha

so current flowing up to 240Vdc is that right?

and 200-500A "to start"

is 50-120kW

That's a whole lotta big-ass resistors.

Active coolant circulation, cooling towers out back?

I got about 30Kw of resistance. Yes, rated for the power. I can get more.


Dunnon about 200A, or a peak design of 500A, those were just numbers Im feeling around. I am feeling I should build it big enough to be expandable and Higher V later. Hence the thinking calibrate it with a 500A or 800A or 1000A shunt first.

Yeah, heat up the garage pretty good. The finned 1700w turn almost red hot ( dull) on current.. and well, 6 of those and 14 of the 1.25 Kw, thats...

......................... about 27 Kw.... ( but only maxxed out @ 240v.... )

That I have, ready to be mounted on copper standoffs and have forced air blown past them.... The 14" strip heaters are designed for continuous operation. They have datasheets.

I have considered hooking it right up to a electrical warehouse/ industrial heater, not big enough...

https://www.tempco.com/Tempco/Resources/08-Strip-Resources/ChannelStripCatalogPages.pdf
The CSF00019 I got retail at $174.25. I have 6.
https://www.tempco.com/Tempco/Resources/Part-Numbers/Finned-Channel-Strip-Heater-Preconfigured-Part-Numbers.htm

The CSH00331 I got ( 14 at the time being, cost is scrap price retail msrp prolly round 50$ each.

Spare CA3 and analogger....


Retail they are expensive but I can get alot of them pretty cheap. I also sell copper, so you dont have to school me on the gauge of the wire too much in this whole idea generating thread. I got a pretty good handle on how to engineer things, just want to know how you guys think.

In the end, we gonna see something that I build datalog something, this is for sure... but how would you want to do it? TY I mean, its not an overnight. Just a plan, a pile ( 30Kw lool) of resisters, and an spare CA3..... that can datalog.

water will sink your heat much more effectively than pushing air around

Your "grinspector" is limited to ~72V and 35A I believe, talking about their stacked resistor bank

"With a 36V pack it provides 2.5A, 5A, and 10A test currents, while a 48V/52V battery can be discharged at 3.5A, 7A, and 14A"

Yes, but the CA3 can datalog the HV ( well above 84v, or 20s, common ebike pack) of an actual EV battery and use many confiigurations of shunts for a custom bank power burning system... that can datalog the stuff.

I know how much the average house ( Kwh) uses per day, and I know how much power I will be using ( .. )
I am not using a "grinspector", buddy. Lol. I am building an electrical load that will be monitored by a (HV) (HC) datalogging CYCLE ANALYST with a sharable file that anybody can peruse.

I have researched water based systems and have two Water heater coils for PointOfUse systems that can pump water through for thermal heating and datalogging temp rise / volume.

Thats plan 'B". I have an 8 Kw and a 13 Kw. In hand. Plus pumps. And drums, both in plastic and metal. I also have wter. I also have a thermodynamic expert steam engineer. I know the density of H2o and the volume calculations. I understand entropy quite well.

Who said it was " my" Grinspector? That was just an example of a resistance load that uses Grin Technology to see teh battery specs. Last time I checcked.. It was a Grin Tech product... ?


I want to do the same but with much greater capability.


You wanna see the pic of my " big ass pile " of resisters? 30Kw? Certainly soon, and mounted, and ready to wire.. how I go about the switching and the design load, IDk. Maybe 100A contin is good enough, but I dont think so. I think I need at least 200A.

The Grinspector product uses a CA for exactly that purpose, just custom firmware

Cool where can learn more about this firmware?

Water heaters are alot less common for me and more expensive, i literally have a pile of these air finned strip heaters of high wattage... and a cold garage.... I figure I dont want to waste the heat. Yes, there are many ways to do this, but have to pick one for a first draft of the plan.. then revision...

I mean, I have 1225Ah Lead acid batteries, to test, and I dont want to take a month to test all of them:!!?!?!

So everything ( versatility vs performance specs . vs reason ability) ( in the design is what I need) from a 24v battery to a 120v or 134v or maybe even into the higher voltages? Im thinking? Is what I am kinda looking for in this brainstorm. Should I keep it so High? What current is good enough to test the typical EV pack? ..: what would be cheapest, most accurate, or fastest..easiest... ect....

The standard capacity test for lead is at the 20-hour rate, and of course you should do each unit on its own, not get average for a big composite bank at a time.

Google Grinspector for Justin's info page.

I did not mean "water heaters" as such, but lots just toss the big aluminum wound resistors in a barrel of water, or as you say, DIY some short paralleled nichrome wires.

That part is easier I think, than finding / building the dataloggesr capable of handling both the high amps and those higher voltages.

I have the dataloggger, capable of handling real HV EV amps, and volts, in hand, ready to go, with datalogging capability.

I need a reliable simple circuit to discharge big batteries fast. While logging. With the datalogger in hand, that i have and is built for this sort of thinking, tinkering,... and thing....



i got a few spools of Nicrome too.,


0.020" ( inch ) ( 1.66 O.P.F)

...and the spreadsheet that will output the temperature vs resistance curve of the metal vs lengt of the Nicrome...

In hand also.....Ready to go.

I know most common standards like the 20hr rate you mentioned.

Link to the data logger please?

ITS AN ANALOGGER! Lol A Grin Analogger and a CA3 . I thought I said that a few times already. ( two times lol)

https://www.ebikes.ca/product-info/analogger.html

So, I have the Analogger, the CA3 ( backup for my bike,) switches, dissconects, relays, shunt trips, ect.. Wire, lugs, Resistance load ( to heat the garage in its spare time, ) a computer to read the data ( .csv), all this stuff I have.
.. here is a screenshot of a 1000W pull @ 80v on a datalog that show voltage droop per ~1000w current draw: :::::::


... So..... How many amps should I design for , do you think?

if you're after something that is automatic, i don't know how to build the hardware (and perhaps software?) you need to control the load for discharging, and the charger for recharging. so i can't offer much advice on that part.

if you're just going to manually control it, and will be there to shut off in case of problems on either charge or discharge, etc., then it's even easier, basically using high-current high-votlage contactors (like gigavac, etc) to actually pass the currents, and manual switches to engage or disengage the contactors. since this is presumably a stationary unit, size of those contactors / etc doesn't matter, so you can probably get cheaper ones as they don't have to be portable like those in an ev.


buuuut...as has already been mentioned, grin tech, makers of the cycle analyst, already have an existing project that basically does what you appear to be wanting to do, in a different way:
https://www.ebikes.ca/product-info/grinspector.html
it is a control box (that can use your existing cycle analyst) and windows software on a computer that logs (afaict) all the stuff you're after, though it doesn't presently switch as high a load as you want to. that should be easy enough to fix, by adding the contactors you'd probably be using anyway, and your load resistors, and some drive circuitry between the output of the grinspector and the contactors so that instead of the grinspector directly controlling the load resistor(s), it would control the contactors taht then control the load resistors. you'd just need to setup the load resistor bank for the proper current you expect, and then tell the software what that is. (I think--you'd want to ask grin about that part to be sure it would work).


anyway, just trying to help you simplify the project, if it is useful to you.

if not, apologies for the interruption.

regarding the load bank:

if you haven't already considered this, in order to draw a certain current from a certain voltage pack, you have to have the right resistance to do it.

so let's say you have a 100v pack, and you want to test it at 100a. you need a 1ohm resistance to do that (100v/100a=1ohm. so if your resistors are 20ohm, you have to parallel 20 of them to get the 1ohm you need (20/1=20).

now let's say you want to test a 24v pack at the same 100a. for that you need a 0.24ohm resistance (24v/100a=0.24ohm). using the same 20ohm resistors, you have to parallel 83.34 of them to get 0.24ohm resistance (20/0.24=83.333333...).

and so on, so for each different voltage, you'd have to reconfigure the resistance of the load bank for the same current.


if the resistors you have are all low resistance, like half an ohm, you'd instead series them to get specific currents.

you might havet o use parallel/series networks to get a specific current, but likly you can just live with a current tht is close to what you are after, and use just the closest loadbank resistance to whaty ou're after to do it with.

is that useful?

as far as how much current to design for, that depends on your needs.

if you're testing packs to make sure they'll work on a bike or vehicle, without doing a road test (or to simulate a road test but while logging all teh info under controlled conditions), you'd need to be able to draw enough current to equal whatever the bike or vehicle, etc., would ever be able to draw.

let's say it's a 2kw ebike using a 52v battery pack. that's 2000w / 52v = 38.46a needed to simulate the max draw of that system.

however, it's common for such systems to have momentary peaks under startup from a stop that are much higher, so if you also want to simulate those peaks, you'd need a load bank that's able to draw that much more current. how much more depends on how hard you want to push the packs being tested.


if you want to be able to test large-ev packs, you're probably going to need to be able to draw hundreds of amps at hundreds of volts. depends on the pack.


basically you have to define what you're going to test, and why you're testing it, to determine how much of a load the tester needs to be able to put on the dut (device under test).

That is exactly what I am doing. Making a big big "grinspector" like rig. With a CA3, shunt, and analogger.

I suppose I am capable of driving an igbt easy enough, or like my dad suggested, a knife switch.

Yes, calculated Ohm load for certain currents specific to the "DUT". Arrangement of wires bolted to terminals with the 14 (1250w) resisters in parallel and the 6 (1700w)... also. I get so low, and trying to get the ball park figures common packs are made of.

Yes one setup for certain V and A, and rewire the terminals for the certain load with the next battery. So I can test a 24v battery one day, an 80v another, ect... . Based on V=IR.

I know the Grinspector exists, but I am absolutely not paying 150$ for a (14A 48 volt) load ( that Grin sells, ) , when I pay 15$ for 10 (1250w) ceramic coil heaters. I cant afford 150$. Esp not for a load that is only like 1000w power ( less). I need datalogs where reader can see the initial voltage, the amp draw, and the resulting "droop" or " sag". The Ah and the V @ rated discharge capacity, ect. The pack IR. CA3 will tell me all this. On datalog.

The goal IS hundreds of amps and hundreds of volts. For EV packs.

I'm not suggesting you should buy commercial dummy loads from Grin or anyone.

But saying you can't afford $150 is a big red flag IMO, you're dealing with very lethal voltages in some cases crazy heat production,

bad batteries even lead can "explode".

Do not compromise on safety!

Comon, Jonh, Budget is not always about what you can spend the most money on.

Here: For simplicity:

I CAN afford more than a high school-er, generally, but not as much as a millionaire or billionaire. Ok? Understand my buying price range now? Feel better about my safety yet? Thankyou for the concern based on my budget (choices)...

I meant I cannot FATHOM spending 150$ for something I can build 10x more powerful for .. 15$.

It would be like buying bottled water to fill the bathtub with, @ 8.34 lbs / gallon and 7.35 Gal/CuFt.. ridiculous. Yes, It is safer to pour water out of a bottle for 1.50 each, but when a hose can output 3 gal/min, I feel I can "Afford" the danger of getting splashed .... so I dont have to pay 1.50 for a pint of water that is almost free from the hose....

Affording things is not always about money, friend. I have a budget of reliability I will be spending here too: Among other "budgets"... ...

I mean, so far I have

125$ in the CaV3
10$ in temp senders ($$$$)
about 100$ in random thermocuples, meters, ect ($$$)
100$ inthe analogger ($$)
A 16 gb Core I5 Win10 3.4Ghz 1250 Gb computer (1000$)
3Kw AABailey 240v power supply ( 5000$+)
About 2000$ WORTH of strip heaters... purchased smartly for 15$. The Grin setup is wimpy comparable. The Grin setup is wimp when it comes to power: YOu could buy 10 of those for 10Kw power handling, and spend 1500$... when I just spent 15$ and purchased a bank of resisters that is 10x more powerful... for 20Kw, or 2x the Grin resistance at 4x Voltage.
A 300A hall sensor and current meter/relay/joule counter for backup data, calibration, Wifi... (40$)
Overload shunts, underload shuts, alarms, disconnects, contactors, double and matching.. ($$$$$$$$)

So there. So far I am about 9000$ into it, retail. Just for the maching e parts. Not including the Engineer or the land.

I have 150$ I coulds spend: i have 1500$ I could spend: I have 15,000$ I could spend...:

I could wast the whole 15K$ on a new air conditioner for the building this machine is going in, and call it part of the machine.. but nah...

I will consider some things a waste of money, and not do them. Unneeded or wasted expense.

I have fire protection in place. I have a registered certified practicing engineer ( that the average man cannot cannot cannot afford) at my biding inspecting the installation and a phone call away. The building is insured. I do not need to worry about safety for I have plans in place.

Feel a litte insulted.. that you think since I have less than "150$" I cannot do this project.

I assure you I have much more than that already in the machine. There will be consessions for " battery explosion" that which is a very real possibility. A dedicated building will be used. Buildings are expensive, and cost a little more than 150$, last time i checked. YOu might think the "150$ " way is best, but I come from a heritage of engineers, who sell things to their clients, for a good price, and the Grin resistance bank is NOT a good price in my opinion.

Nothing in the world shuts down because of budget, when the budget is exuberant.. they find a better budget.

Red flag my ass. I am the most capable and interested person you probally know when it comes to hogbby testing of OEM lithium battery and individual cells. Right now I have 5, 15, 25, 35, 50, 60, 63, and 105Ah cells to play with. And batteries to be built with each, that I need the datalog from before I sell them.

Red flag my ass. You a red flag. Lol.

So onto the control relay ladders and output sections.

Anybody know the average current of a :
Power wheels

Ebike

GoCart

Golfcart

Forklift/Scissorlift

Lithium car audio bank

and finally

An EV ( like a Tessla or a Volt, Bolt or Spark)?

How about average battery V?