By changing mosfets this switch can go up to 400V if anyone needs
Thatās awesome! Do you have a final price for them? With heatsink
Not yet
10char
All the parts assembled firmware written and everything is tested. Just left to design a heatsink for it.
Here is small showcase video of it turning on by pushing board. Doesnāt require any modifications
So first batch of switches assembled now just left to make heatsinks.
Final Feature List:
- Operational from 8S (25V) up to 19S with PTS or 22S without PTS (must be disabled) (80V with PTS or 100V without PTS)
- Turn on by pressing button (or if PTS enabled pushing your board)
- Turn off by holding a button for 2 sec
- When switch powered up you will hear beep pattern (in my opinion most important feature when having push to start)
- Enable/Disable Push-To-Start
- Pre Charge circuit (Hard limit inrush current limiter to pre-charge any size capacitor bank available on ESC side)
- High Side Power Switching (Positive wire is switched not negative)
- Current capabilities in 4 stages 50A, 100A, 150A, 200A (I know that mosfets are rated higher but its up to you mosfets will be heatsinked so you can try push more)
- Available in DIY style just the switch or fully assembled with push button and XT90 connectors.
Pricing:
Current Rating | Price for DIY style | Price for Kit |
---|---|---|
50A | 35ā¬ | 55ā¬ |
100A | 45ā¬ | 65ā¬ |
150A | 55ā¬ | 75ā¬ |
200A | 65ā¬ | 85ā¬ |
* Prices donāt include VAT or Shipping
Specifications about size and weight will be announced tomorrow as heatsink is yet to be made. Should be available to purchase starting from next week.
P.S. The higher current rating the less heat will be produced so if you concerned about heating issues grab the biggest one
Heatsink will look something like this:
The heatsink will be around 53x35x8mm and total height should be less than 18mm or something like that
So the switch is up online in the store
Awesome as usual. Is the amp spec the continuous, momentary burst, or āa few secondsā number?
With PTS can we run switch-less (like perhaps a timer to turn itself off)?
Well amp settings are the same as esc At lower ratings it will warm quite a bit when running at the limits so I always suggest getting bigger as when less heat will produced and you will have some overhead Running continues for more than 5-10 min you will definitely will want attach it to something bigger metal or grab higher rating one
Regarding PTS there are no auto turn off timer, I donāt like to have accidentally turn on board and wait till it turns off or anything happening automatically so for this reason both button and PTS work at the same time
Ordered one.
Excited to have one of your products on one of my boards
Few more 100A and 150A added to the stock. More stock will be available with new parts arriving soon
Tease us please
Iām just curious, how does your switch generate the 12V bootstrapped rail for driving the high side N-Channel MOSFETS?
Also, couldnāt a precharge transistor be avoided by adding a RC network between the mosfet driver, mosfet gate, and ground? Thatās what Linear Technologies recommends with their LTC700x line of high side mosfet drivers.
It uses magic Well mosfets donāt like to work as linear regulators while you make them do pre-charge They work out of their comfort zone So I better dump everything through separate resistor
Please donāt tell me you are using a 555 Timer as a charge pump; I was considering that design, but I opted to spend the extra money for the LTC7004 (and PCBs to solder the pesky thing onto) instead.
I would highly recommend the LTC7000-1, you get 135V operation, current limiting feature, and no need for additional components aside from the transistors themselves and some passives. I think I saw someone else on this forum using that chip with some chassis mount MOSFET blocks, but he was charging well over $100 for it.
You are right that it is best to avoid operating MOSFETs in their linear region, but sometimes you can cut costs for both yourself and the end user with minimal impact on performance, or use a smaller transistor for pre-charging since it only needs to pass like 10% of the rated load current for a short time. As long as you keep the slew rate reasonable with regards to the load capacitance, MOSFETs will have no problem operating in their linear region for a short period of time, just make sure to have a sufficiently long rise time (10-500ms) or to not exceed avalanche energy if the rise time is under 10ms, which shouldnāt be a problem with International Rectifier MOSFETs since the actual MOSFET die is so big (this is also why their gate charge characteristics suck, but since this is a static application, that characteristic is irrelevant). If this method didnāt work, Linear Technologies wouldnāt recommend it in their datasheets and application notes.
Yeah, I use 555 timer and some other logic chips to implement 4 bit MCU design for all the logic I am doing
It was me who made that switch but it wasnāt using that chip and its solid switch. This switch if not MOSFETs and some other components can do up to 1000V
Well, I am not a Chinese manufacturer to cut corners to increase profits, I want my device to be solid and robust and not just āworkingā.
P.S. Not all application notes and datasheets are good examples to follow, from my experiences and some other well know engineers and some youtube guys have discussed this topic.
So you are using a 555 TImer.
All manufacturers, not just Chinese, cut corners (civil engineering failures in the U.S. are clear indications of that), but there is a difference between cutting corners and adding unnecessary cost and complexity. Many times, it is best to keep it simple because it reduces the number of required components and allows you to provide a lower cost to your customers. LTC7000 -1 only needs like five extra passive components and the transistor. I literally plugged my switch in backwards (by mistake) and after correcting that, it still works no problem.
But, I guess using a 555 Timer instead of a MOSFET Driver is already being cost efficient (50 cents for a CMOS 555 vs 4-5 dollars for the LTC7000). I have no idea whether a 555 or the LTC7000 would be more rugged for this application . I did blow up quite a few 555 Timers when I was prototyping my charge pump idea, but Iāve also blown up many bootstrapped MOSFET drivers (none from LT though, I did some crazy stuff to other LT chips and they didnāt die, haha.) If you want really rugged parts, try Allegro Microsystems, all their parts are automotive rated, and despite being shorted, overvolted, overheated, the chips I got from them still work perfectly fine.
Most, if not all, datasheets from large semiconductor manufacturers contain good and reliable information. The issue is not with the datasheets or ANs, but that many engineers donāt read them or misinterpret their information, and as result, end up selecting more expensive and inferior parts, or straight up select the wrong part altogether (there is this story of someone who was designing a triangle arduino board, and he ended up selecting the wrong 3.3V regulator or designed the PCB for it incorrectly because he neglected to read the datasheet, the enable pin ended up having inverted logic).
It is impressive that you designed such a device to work off a 555 timer and implemented pre-charge functionality in such a small form factor and for only ~$40 despite the complexity. I was only able to figure out how to implement pre-charge on a lowside configuration . High side pre-charge is much more complicated; Iām guessing this is why Linear Technologies opted for the RC network approach.
Yes, I have implement pre-charge, push to turn on, push and hold to turn off, push to start, beeper melody using 555 and some other logic elements Using only 11 components
Wow! Very cool. I guess the next step would be to add a coulomb counter with bluetooth .