Battery Disconnect method.

Sorry dude.
Click to expand...


I was just highlighting that the switches you are recommending aren't fit for purpose. It wasn't a personal attack on you. But you're response does make you seem like a bit of a fuckwit, tbh.
Click to expand...


True, and I am an absolute fuckwit, honestly. Not a bit of one. No offense taken.



84VDC can make a spectacular arc
Click to expand...
Good thing I have a secondary protection in place , and a therd Tord? Treez? Thray?? Tertiary? Is that the word im lookin for? Lol yea. I think it is.

Hey Im cheap. Whatta shame. Zap.

Highlight what you wish, I'll be using 10$ USA made 75A 30v rated switches from fire trucks. I derate my shit for fun.... cause I do what I want sometimes. The purpose for a rate on the switch is for the engineering. True that.

No personal fuckwit attacks taken note.
 
I use a 40A AC single pole breaker as an on/off switch for my battery. Since it's AC, it won't actuate based on current, when used in a DC circuit, so I'm only using it because of the contact ratings. Still working fine after a thousand plus actuations. $6 Home Depot.
 
E-HP said:
I use a 40A AC single pole breaker as an on/off switch for my battery. Since it's AC, it won't actuate based on current, when used in a DC circuit, so I'm only using it because of the contact ratings. Still working fine after a thousand plus actuations.
Click to expand...

Lol Ive used those to break DC. lol. Works.

Also I got about a ton of DIn mounted 30a therabouts DC 600v shit. Magnetic contactor from machinery..
 
PlanetDad said:
I suppose I do not 'need' to disconnect the battery. But wanted a key option for theft and convenience, as well as the aesthetically satisfying manner of a key switch that would allow power to battery disconnect switch and DC-DC, and then lighted led switch's for all the other stuff.

I've nearly given up on the idea,
Click to expand...
well, as i noted, you can use the keyswitch to turn off the controller* so it can't respond to any input, and at the same time turn off the dc-dc.

this effectively cuts power from the battery, if you don't have anythng else connected to it directly.



* depending on the type, almost all "generic" ones and quite a few others based on the style will have a keyswitch/ignition/doorlock/etc wire that has to be connected to the battery positive for them to work, separate from the actual battery positive main power wire. those that dont' usually have this wire as an internal jumper wire you could cut and then run out to your keyswitch the same way.
 
There are a few BMS out there with an on off switch; they use the mosfets to switch the battery current on and off, with no need of a large switch.
 
PlanetDad said:
Issue is a proper battery disconnect, the new packs do not have there own disconnect switch. And finding a switch rated for 60 volts up to 50 amps is difficult ( in an aesthetic size ),
Click to expand...

Move into the digital age. Get a bluetooth BMS. You can turn the power off in the app. Discharge FETs will cut off, still allowing charge. No sparks plugging in. Simple and quick.
 
In spite of the caution that was advised by builders who have not tried the Solid State Relay / SSR that I linked to earlier, it was designed with the highest-spec internal components, and was designed by a US engineer who is also a member of ES (not me, I am not an electronics engineer).

The designer has requested anonymity because the only retail outlet that agreed to carry them was Luna Cycle, and various ES members have had issues with the owner of Luna.

I have one, and it is a good product. I paid full retail for it. That being said, I'm sure the other linked switches work just as the user stated.
 
spinningmagnets said:
The designer has requested anonymity.... only retail outlet that agreed to carry them was Luna....

I have one, and it is a good product. I paid full retail for it.
Click to expand...

Lol. Sure.. Invented by... Flipsky.. Or at least here is the Flip sky one that has been around forever and can handle like 280/800A... not the wimpandcheapasf Luna offering.

Amain has them. Always had. Alien power systems have them.
https://flipsky.net/products/antisp...-xQfe8W8SKxveemf04Ph7Nv-sUYGDtSAaAomoEALw_wcB

Here is the Alien one.. SAME EXACT PCB but BETTER RATINGS withGOOD COmponants... Rated for 2x the current and power. Luna is only ( lol ) 120A PeAk.... Plus the Lunashit is always unavailable. Lol. What a surprise, they will come out wit a "better one" and talk up them "ratings"...

See a Luna pattern ? Emerging? Old tech talked up and sold for more than it is worth? lol.... Lemme pay 99.95$ for that INFINEON clone 60a controller again...

https://alienpowersystem.com/shop/r...lien-antispark-power-switch-3-16s-300amp-max/

Its an old switch, from my experiences. I have seen one with the exact same sticker. Same PCB, low ratings... on the luna.

Another. https://www.amazon.co.uk/Trivent-Electric-Skateboard-Anti-Spark-Longboard/dp/B07CMYPZ6Z
Same exact look.

Another. Same exact PCB... I guess your friend who invented it is getting rich from the royalties?
https://fliermodel.en.alibaba.com/p...Flier_Power_Switch_2_14S_120A_for_RC_ESC.html

No offense to you, SM.... Just your place to purchase is always the most jazzed up with ratings and prices that can be easily beat. I believe you have some proprietary design info like a choice of populated components 0r something, but I dont think it was made specifically for the Luna, when Flipsky has been selling the same thing but with a 300A/800A rating. They are cheap, and they fail straight closed. Dont hate me for being the Devils Advocate. I believe there is a list of options the ( chinese) manufacturer offers to the ( business that sells to) consumers with options for population and certain design aspects, but the contracting company still just picks the stuff off a list, chooses how many to buy, marks up, and then caters to the individual buyers..

Another: Same exact PCB, but this one is doubled pcb for more power and Voltage handling... Emcotech is good stuff.... You know, for when you are flying your 3-5000$ RC jet that only allows ONE mistake... Double redundancy... http://www.rcjetshobby.com/en/inter...urity-emcotec-power-switch-4047991720117.html

Another.
Betterrating. https://www.mboards.co/products/power-switch-on-off-button?_pos=1&_sid=6b3ea2bfb&_ss=r

Not trying to knock your idea, S. Magnets. Only and just the idea that the LUna switch is the best or the one that was first, or proprietary in some way ( I would love to see concrete evidence of this)... Show me the concrete and I will stand on it as you do. It is a RC switch from what I have experienced. Commonly availible from the world of RC for ages now. My best heli cost 3000$ and that is real brand new money. I have been doing the RC thing for many years now, as long as you and the Ebikes. We all strive. My heli is reliable as i could make it. They kill in an instant .. so they have to be.

You would think they ( Luna) would have designed their own sticker to go under the clear heatshrink... Not to much more ( cost in) setup in a production run ( of a number of products) to implement such a thing? A sticker, printed from a graphic, on a computor, that you will be printing anyway?... The Flipsky one is even in a hardcase. Lol.

You said above that I have not used them ( that design.. that pcb, with that red and black silicone wire, and that shiny remote mounted switch, same length of wire and same look: carried by many other online retailers... ). I have. They failed.

People, please knock my ideas freely. Honestly, using a 30v rates switch for a 48v system should not be to much a problem, a 120/240v AC breaker will still break at rated current on DC of half the voltage at load, and a precharge will stop eroding of contacts if implemented correctly. You can double up two DC DIN power breakers for 2x rated current. ( 2x30A DC DIN) = 1x60A DC DIN). Understand zero crossing, amplitude, magnetic flux, power and load as you will.

If you want to buy cheap common Solidstate RC switches, do as you wish.

Lotta Wh in our application vs hobby rc application ( where the batteries are put away after flyig for the day.. NOT left hooked up FOREVER and wheeled INSIDE to charge .. permanently mounted... on vehicles... that shake and go through rain and mud and snow... ( unattended) (in your home, where the children play)( and sleep).

Any Ev system should have a fuse, a breaker, and an emergency/maintenance disconnect. All three, IMo, if it has any significant power or energy stored. I bet you will knock me for using "12v rated" audio fuses, on 90v.. too. Even when some have sworn by them on the forum. Sound like a pistol when the go and they go fast. I have alot of (laymans) experience with power in industry and what is commonly used and the redundancies surrounding them.
 
serious_sam said:
highlighting that the switches you are recommending aren't fit for purpose.

Says you and what engineer.. or are you one? Could you define "derating " for me (us) then?

They aren't rated to break an arc above that voltage.

They are.. there is such a thing a "derating" and whatnot... you only see what was embossed on the face of the thing... Engineers derate things from the faced value all the time...

I don't think you understand that.

I certainty do.

I dont think you understand what I understand.





It would be a shame Word flawed advice and reasoning ( or your flawwed interperatation of my reasoning and experience and knowledge), and got hurt because of it. (OOCH!)... too bad they wont. Another shame. Lol.

bit of a fuckwit, tbh. Takes one to know one.
Click to expand...

Nothing personal, Sam, dont think you understand what I understand. Understand that I like durability, reliability, cost efficiency, design aspects, service life, availability and the like. I have experience and experts to ask when I do things Tthat I do, and dont just put anything out there for no reason. Dont call me a wrong F-Wit and say I dont understand, or are negligent, before you ask me why I would think this or that first please. Peace dood.

Its not rocket science.
 
amberwolf said:
PlanetDad said:
I suppose I do not 'need' to disconnect the battery. But wanted a key option for theft and convenience, as well as the aesthetically satisfying manner of a key switch that would allow power to battery disconnect switch and DC-DC, and then lighted led switch's for all the other stuff.

I've nearly given up on the idea,
Click to expand...
well, as i noted, you can use the keyswitch to turn off the controller* so it can't respond to any input, and at the same time turn off the dc-dc.

this effectively cuts power from the battery, if you don't have anythng else connected to it directly.



* depending on the type, almost all "generic" ones and quite a few others based on the style will have a keyswitch/ignition/doorlock/etc wire that has to be connected to the battery positive for them to work, separate from the actual battery positive main power wire. those that dont' usually have this wire as an internal jumper wire you could cut and then run out to your keyswitch the same way.
Click to expand...

Yes, ATM they are "generic" ish controllers. That up grade also comes with changing out all wiring connections to motor/throttle/breaks ect, and not sure im willing to do that ATM, however will be considering it.

Have one controller already, ill be taking it apart to find that jumper wire.

Also just to clarify on the Solid state skate board option, When they fail how do they fail? Just stuck closed?
 
Tommm said:
PlanetDad said:
Issue is a proper battery disconnect, the new packs do not have there own disconnect switch. And finding a switch rated for 60 volts up to 50 amps is difficult ( in an aesthetic size ),
Click to expand...

Move into the digital age. Get a bluetooth BMS. You can turn the power off in the app. Discharge FETs will cut off, still allowing charge. No sparks plugging in. Simple and quick.
Click to expand...

Opening my new packs and replacing the BMS makes me cringe.. But with my skill/experience level could probably do it, would need some guidance.
 
They're decent mosfets. Is there a particular reason you stopped using this setup?
Click to expand...

When I made the board, I accidentally left a small air pocket where the control switches for the on/off circuit was. Water managed to seep in through the XT60 after about 6 months and shorted out the control transistors (a PNPN self-latching configuration). By the time I got to it, the board was a write-off and I ended up using the MOSFETs for other projects.

As for why I didn't just rebuild the board, just 3 FETs in parallel got pretty hot when I was running 80+ amps through the thing. After giving XT90s a try, I decided to stick with them. They are a lot smaller than a MOSFET array of the appropriate size, are way cheaper, and are significantly less complicated.

If I had more FETs (I'd recommend 1 FET per 20 peak amps w/o a heatsink) and epoxy encapsulant I wouldn't hesitate to do the MOSFET approach over again.
 
DogDipstick said:
there is such a thing a "derating" and whatnot... you only see what was embossed on the face of the thing... Engineers derate things from the faced value all the time
Click to expand...
Derating is the opposite of what your talking about. You're trying to use a component over its rated voltage. Derating is reducing the rating of a component. Not increasing. SMH.
https://en.wikipedia.org/wiki/Derating

You know what, I've been sucked into arguing with an idiot, and I feel just a little bit dumber for it. Shame on me for not knowing better.
 
thorlancaster328 said:
If I had more FETs (I'd recommend 1 FET per 20 peak amps w/o a heatsink) and epoxy encapsulant I wouldn't hesitate to do the MOSFET approach over again.
Click to expand...
Yeah, 20A per mosfet is a good starting point. It will depend on the Rds and heatsink. I'm running just under 60A peak per mosfet without any significant heat. But they're 1.5mOhm, and soldered to a copper bar bolted to an aluminium enclosure.

I like being able to disconnect power with a switch on the handlebars, and sparkless connectors can't provide that functionality for me.
 
I would use a silicone controlled rectifier.
It could be located in close proximity to controller and battery with high voltage/current line and then use a small low current control line to your handlebar or other desirable place and use a low voltage current switch to control. This should work nicely:

150 amp rated and 100v to 1600v
Voila.

https://www.ebay.com/itm/Powerex-T6001215-T600-Phase-Control-SCR-150-175-AMPs-100-1600-V-FNFP/183771947038?hash=item2ac9a9601e:g:BB8AAOSwvlRcsNL~
 
epictetus said:
I would use a silicone controlled rectifier.
https://www.ebay.com/itm/Powerex-T6001215-T600-Phase-Control-SCR-150-175-AMPs-100-1600-V-FNFP/183771947038?hash=item2ac9a9601e:g:BB8AAOSwvlRcsNL~
Click to expand...
The problem with SCRs is that they are latching (can't be turned off by the gate, in the way you would turn off a mosfet). The current flowing through an SCR needs to be reduced to below the holding current (say, below ~10mA) to turn it off. Otherwise it stays on.

Also, power dissipation from the SCR will be very high (30W @ 30A).

For example, using T6001215-T600:
- Vtm (voltage drop across SCR) @ 30A (OPs requirement of 1500W@48V) = 1.0V (approximate value from graph in datasheet)
- Power dissipation = 1V x 30A = 30W (soldering iron hot)

Compared to a mosfet (for example, IPB017N10N5LF):
- Rds = 1.5mOhm
- Vds @ 30A = 1.5mOhm x 30A = 45mV
- Power dissipation = 45mV x 30A = 1.35W (basically useable without a heatsink)
 
serious_sam said:
The problem with SCRs is that they are latching (can't be turned off by the gate, in the way you would turn off a mosfet). The current flowing through an SCR needs to be reduced to below the holding current (say, below ~10mA) to turn it off. Otherwise it stays on.

Also, power dissipation from the SCR will be very high (30W @
Click to expand...
Wouldn't a commutation circuit, where the anode current is diverted momentarily via a transistor, thereby reducing the anode current to below the holding threshold be applicable?

The heat dissipation would be a factor and a concern, however, if a SCR is designed, built, and rated for 100 amps at 100volts the heat generated has had to be taken into consideration and designed for continuous operation so as to be manageable. How would the amount of heat generated be dissipated, especially given that the device is as big as ones thumb? A heatsink is one way of course. Wouldn't that suffice under the parameters of these loads?
 
Would you actually use a silicon controlled rectifier or are you just hypothesizing that it would maybe be a good idea? They are designed to switch AC, not DC. If you are using them to switch DC you'd need a super low resistance MOSFET to turn them off, defeating the point. You could just heatsink the FET, take out the SCR, and be way more efficient to boot.

As for the heatsink, this thing looks like it is meant to be screwed into a heavy bus bar or enclosure to take the heat away. Dissipating 30 watts on a bike without a fan would require a heatsink the size of a large smartphone and quite a bit thicker. Also, every watt that gets wasted as heat doesn't go towards turning the wheels.
 
epictetus said:
Wouldn't that suffice?
Click to expand...
Well, technically you could make it work. Yes you are correct.

But there's really zero reasons to use an SCR in this application.

The commutation transistor (let's call it a mosfet) should be rated big enough to carry full load current (otherwise it'll pop if you activate it under load). Which means you're using a mosfet (capable of switching the load) to trigger a (now redundant) SCR, 20x less efficiently.
 
An AC switch for a given amperage rating will be smaller than a DC switch for that same rated amperage. DC switches can be surprisingly large. The rotating "fire truck" switch shown earlier is intended and designed to be use when the vehicle is off. It's purpose is prevent the battery from slowly discharging if the firetruck is not used for weeks at a time. It will be fine if you use it for that same purpose.

The reason DC switches are much larger for a given amperage than AC switches is to be able to move the contacts as far away from each other when current is flowing through the switch. A "kill switch" has a different purpose and design than a "battery isolation switch". AC passes through the zero-voltage point 60-times a second (60-cycle), and this fact allows a relatively light duty switch to break the arc. The higher the voltage, the longer the DC arc will be as the switch is opened to break the flow.

The purpose of an ebike kill switch is for the very rare occasion when the throttle or controller fails in the "on" position, and you must break the current flow while the system is running. Owners of off-road Jeeps and boats have made improvised repairs in remote locations by using two car batteries to weld (24V 200A dead short), and that would be an extreme example of the physics of a DC arc. https://youtu.be/m95cWlSPjtU

The MOSFET is a semiconductor that was first demonstrated in 1960. It can be used as a diode, a signal amplifier, or an on/off-switch. In a configuration where it is used as a Solid State Relay / SSR, a small signal from a light-duty switch can activate the FETs and turn a large DC current on and off with no arc. Several FETs can be paralleled to increase the amount of current a device can flow. Two examples of this are the Malectrics and the kWeld FET-based welding "switches".

The Malectrics uses eight FETs, and the kWeld uses only six, but...the kWeld can flow a significantly larger amount of current. The Malectrics works well for the task it was designed to do, but this is one example where the Malectrics uses much more affordable FETs, but they are smaller, less efficient, and have a higher "on" resistance value. That's not a problem, they work fine. The lower resistance of the kWeld FETs costs more, but they not only flow more current per FET, they run cooler while they do that.

The SSR style of "on off" switch that uses FETs can be used as a battery isolation switch when the system is parked, and it can also be used as a kill switch when the system is running. I want to thank everyone who posted a link to a switch, so readers can compare features and find a switch that matches their needs...
 
It's simple as you make it, limiting the inrush current by using resistor over time will not weld any switches.

Any switch inbetween the battery and controller will need to be capable of large current and usually a relay be it solid state or mechanical is used for that reason but it's also just as easy to use a 2 stage switch that brings in a resistor load first to safely charge the circuit to equal potential then flip to the direct path through the chosen relay.

There's the more expensive route of using fets but if you have a look how it done on EV then you will see mechanical contactor, I rest my case save cash do what's simple and works don't over complicate simple.
 
You must log in or register to reply here.

Similar threads

P
Sanity check of proposed battery solution
Replies
9
Views
736
powerlord
P
Back
Top