Castle - Make A Special Edition Monster

[lutach]

Originally Posted by Pdelcast

I've found that LFPAKs are inferior to the SO8-FL package (flat lead, leadless SO-8 -- or PowerPAK SO-8 depending on who makes it) because they still have wire bonds on the source leads, where the SO8-FL has copper clips. The SO8-FL package also dissipates heat much better than the LFPAK, which is kind of a half SO-8, half DPAK package.

The DPAK doesn't take as large a die as either the LFPAK or the SO8-FL, and it has about 2 milliohms of package resistance -- making it not very well suited for 30V applications.

Will the LFPAK work better in a higher voltage controller? I do own a controller that uses 30V LFPAK MOSFETs, rated for 6S lipos and it works pretty well. It didn't get hot at all for the 5 minutes I used it. I was impressed by that as the heat sink on it looks really crappy. If you made a HV car controller that looks like the HV and SHV controllers using the LFPAK MOSFETs would that work? I like the way the HV controller looks and I have my HV110 that I still need to send in if you need a guinea pig. I know the LFPAK have some 75V MOSFET and that could even make a possible 16S controller.

Now for my other question I posted before.

Being able to drive all the MOSFETs (keep this in mind), would a controller with 240 MOSFETs total or 80 per phase, but 40 in the H bridge config. work less to make the same amount of power as one with less MOSFETs?

Now, this controller don't need to be all that small. Something that is smaller then the Schulze 40.160 will do . I know more FETs and other components will add to the cost, but that's only if parts are being bought from Digi Key, Mouser and any other catalog distributors so lets keep that on the side for now. I'm just trying to get a rock solid controller. I even went as far as contacting and sending 1500 MOSFETs to a company in China to make me a 15S car controller and they never did one so you know the problems they are facing. All they make is Airplane and Helicopter controllers. I know they are using the Atmel's ATMEGA8L-16AU MCU and some car controllers that I have uses the same MCU. My guess is they are still having trouble with the software.

[Pdelcast]

Originally Posted by lutach

I even went as far as contacting and sending 1500 MOSFETs to a company in China to make me a 15S car controller and they never did one so you know the problems they are facing. All they make is Airplane and Helicopter controllers. I know they are using the Atmel's ATMEGA8L-16AU MCU and some car controllers that I have uses the same MCU. My guess is they are still having trouble with the software.

If they are using the Atmel part, they are likely using either stolen Jeti software, or stolen Schultze software. Both companies had their software compromised and copied in China. Both companies have tried unsuccessfully to stop imports of controllers with their stolen software. (I can't imagine how much money it would take to go after these companies... they setup shop and only do business out of Hong Kong where they are shielded by the Chinese government.)

So it's unlikely that they have any resources to modify the software for RC car use.

It's funny, most of the controller coming out of China identify themselves as a Jeti-18 (regardless of the type/size of the controller) when connected to Jeti's field programmer. Several others ID themselves as Schultze. I've yet to see a Chinese or Taiwanese controller that has decent software that was not stolen from another company.

[lutach]

Originally Posted by Pdelcast

If they are using the Atmel part, they are likely using either stolen Jeti software, or stolen Schultze software. Both companies had their software compromised and copied in China. Both companies have tried unsuccessfully to stop imports of controllers with their stolen software. (I can't imagine how much money it would take to go after these companies... they setup shop and only do business out of Hong Kong where they are shielded by the Chinese government.)

So it's unlikely that they have any resources to modify the software for RC car use.

It's funny, most of the controller coming out of China identify themselves as a Jeti-18 (regardless of the type/size of the controller) when connected to Jeti's field programmer. Several others ID themselves as Schultze. I've yet to see a Chinese or Taiwanese controller that has decent software that was not stolen from another company.

That sucks. So did Jeti and Schulze forgot the protect their software ? This company promised me they were going to do it and now it is basically tha, a promise. I might just get the controllers done with what ever software they have and use them is a car with mechanical brakes .

[Pdelcast]

Originally Posted by lutach

That sucks. So did Jeti and Schulze forgot the protect their software ? This company promised me they were going to do it and now it is basically tha, a promise. I might just get the controllers done with what ever software they have and use them is a car with mechanical brakes .

No, they didn't forget to protect their software. The ATMega software protection can be compromised.

[lutach]

Originally Posted by Pdelcast

No, they didn't forget to protect their software. The ATMega software protection can be compromised.

Good thing the quality of Jeti and Schulze are top notch compared to the Chinese version.

[Pdelcast]

Originally Posted by lutach

Will the LFPAK work better in a higher voltage controller?

I don't mean to say that the LFPAK isn't a good package - - it is one of the better ones on the market. It's just that all of the testing we have done has showed that it is inferior to the SO8-FL package, and the LFPAK doesn't have a price advantage (in fact it is usually more expensive) over the SO8-FL.

Look at Infineon's "Super-SO8", On Semi's "SO8-FlatLead", Fairchild's "Power 56", or Siliconix' "PowerPAK SO-8" package. They are all essentially the same package, are reasonably priced, and perform very very well.

[lutach]

Originally Posted by Pdelcast

I don't mean to say that the LFPAK isn't a good package - - it is one of the better ones on the market. It's just that all of the testing we have done has showed that it is inferior to the SO8-FL package, and the LFPAK doesn't have a price advantage (in fact it is usually more expensive) over the SO8-FL.

Look at Infineon's "Super-SO8", On Semi's "SO8-FlatLead", Fairchild's "Power 56", or Siliconix' "PowerPAK SO-8" package. They are all essentially the same package, are reasonably priced, and perform very very well.

How well would a bunch of "Siliconix's PowerPAK 1212-8 Package" MOSFET (Keep in mind, being able to drive a lot of them) do in a ESC designed for car that can do 12S lipos? I'll bet you can fit a lot of them in a controller similar to the HV. Is there anyway you can treat us RC-Monster followers to a mother of all controllers "RC-Monster Edition"? I know you play with a lot of designs. I can't imagine you just build the products we see in you website. You probably have a 12S controller now that can handle spikes of 200A that you are playing with and not telling us.

[Pdelcast]

Originally Posted by lutach

How well would a bunch of "Siliconix's PowerPAK 1212-8 Package" MOSFET (Keep in mind, being able to drive a lot of them) do in a ESC designed for car that can do 12S lipos? I'll bet you can fit a lot of them in a controller similar to the HV. Is there anyway you can treat us RC-Monster followers to a mother of all controllers "RC-Monster Edition"? I know you play with a lot of designs. I can't imagine you just build the products we see in you website. You probably have a 12S controller now that can handle spikes of 200A that you are playing with and not telling us.

Why the 1212-8 package? The footprint to silicon ratio is better on the larger Powerpak FETs...

I am playing with a version of the Monster that is stackable, and can use 30, 40 or 60V FETs. I already said that earlier...

[BrianG]

Originally Posted by Pdelcast

...I am playing with a version of the Monster that is stackable, and can use 30, 40 or 60V FETs. I already said that earlier...

I want a double stacked version using 60v FETs please. And leave out the UBEC.

[lutach]

Originally Posted by Pdelcast

Why the 1212-8 package? The footprint to silicon ratio is better on the larger Powerpak FETs...

I am playing with a version of the Monster that is stackable, and can use 30, 40 or 60V FETs. I already said that earlier...

I just had an idea that the more FETs would be better? The 1212-8 is roughly half the size of the larger PowerPak and I keep thinking the more the better as each FET will work less to produce the same power as less FETs. I know I'm not an engineer and what we think off might be possible in our heads, but not in true life application. But since you are working on the stackable Monster, when will the HV one be available? Will you be having other guys put the controller to the test or will it just be team drivers?

[Pdelcast]

Originally Posted by lutach

I just had an idea that the more FETs would be better? The 1212-8 is roughly half the size of the larger PowerPak and I keep thinking the more the better as each FET will work less to produce the same power as less FETs. I know I'm not an engineer and what we think off might be possible in our heads, but not in true life application. But since you are working on the stackable Monster, when will the HV one be available? Will you be having other guys put the controller to the test or will it just be team drivers?

I gotcha -- what we really look for during design is the silicon (or RDSon) to footprint ratio. So if you calculate milliohms / mm^2, you can compare silicon ratio pretty directly from FET to FET. Right now it's pretty tough to beat the 5x6mm leadless packages for up to 100V.

I haven't completed the layout of the control board yet -- I'm using a 72V maximum input switching BEC for the HV version of the board -- But I'll make it so that an external BEC can be used (Brian!)

[lutach]

Originally Posted by Pdelcast

I gotcha -- what we really look for during design is the silicon (or RDSon) to footprint ratio. So if you calculate milliohms / mm^2, you can compare silicon ratio pretty directly from FET to FET. Right now it's pretty tough to beat the 5x6mm leadless packages for up to 100V.

I haven't completed the layout of the control board yet -- I'm using a 72V maximum input switching BEC for the HV version of the board -- But I'll make it so that an external BEC can be used (Brian!)

Ok, so basically being able to switch all the FETs, it would be possible to get a lot of FETs in a PCB that is a little larger then what the Monster Max uses. So a ESC that can go as high as 400A is possible to keep things of the safe side if one wants to use a Neu 1527/1D, 2215/1Y, 2230/1Y or a Lehner 3080-5. Patrick, I would put a high priority on such a controller . We are hungry for it.