FET gate waveform
in [Design]
|
Prev: Rebranded transistors?
Next: DC converter
From: Raveninghorde on 19 Feb 2010 16:12 On Fri, 19 Feb 2010 12:18:47 -0500, Hammy <spam(a)spam.com> wrote: >On Fri, 19 Feb 2010 15:20:51 +0000, Raveninghorde ><raveninghorde(a)invalid> wrote: > > >>I may have missed something but the Rdson is only specified at 10V. >>And it has Qgd of 43nC which given the drive capabiltiy of the LM3150 >>looks too high. > > >TH leaves you with pretty slim pickings but here are a couple speced >at 5 or 6 Vgs. > >http://www.fairchildsemi.com/ds/FD%2FFDP5800.pdf > > >http://canada.newark.com/fairchild-semiconductor/fdp5800/transistor/dp/30M0749 > > > >http://canada.newark.com/pdfs/datasheets/Fairchild/FDB14AN06LA0.pdf That looks good. Obsolete but Farnell have stock and if it works well enough it get me out of a corner. Rdson of 14.6 milliohms max at 5V and only 7.9nC Qgd. If the LM3150 calculations are correct it should save about a third of the FET dissipation. > >http://canada.newark.com/fairchild-semiconductor/fdp14an06la0/mosfet/dp/84H4572 > > >http://canada.newark.com/pdfs/datasheets/Fairchild/FDP5645.pdf > >http://canada.newark.com/fairchild-semiconductor/fdp5645/mosfet/dp/58K8848
From: Raveninghorde on 19 Feb 2010 16:50 On Fri, 19 Feb 2010 11:58:35 -0800, Joerg <invalid(a)invalid.invalid> wrote: >Raveninghorde wrote: >> I'm back looking at an on going project. >> >> http://www.zen88234.zen.co.uk/design/Output-2.png >> >> I am investigating the dissipation of Q2 the high side FET. The >> designers prototype (20 miles away) has an IRLZ44Z. My prototype gas >> an 2SK3704 fitted. >> >> The temperature rise is about the same for both FETs though I expected >> about a 10% improvement with the 2SK3704. >> >> The heatsink is an SW50-4 with a theta of 8.6C per watt. I calculate a >> FET dissipation of under 2.5W using the methodology in the LM3150 >> datasheet. >> >> The heatsink rises by about 56C after about 20 minutes at 12V 8A >> running at 200kHz. This T rise drops to 47C at 140kHz. >> >> So I scoped the gate drive with the scope grounded on the source of >> Q2. I got this: >> >> http://www.zen88234.zen.co.uk/photos/IMG_0195.JPG >> >> I repeated the measurement with my old Iwatsu analog scope and got the >> short negative pulse after 100ns but it only goes down to -1V instead >> of -3V. >> >> As a check I scoped the probe's ground point and there is very little >> signal there so the probe isn't picking much radiated noise. >> >> The designer (bro in law) has tried the same measurement using his >> Tektronix scope and gets a pulse down to +1V. >> >> Clearly there is a gate drive issue of some sort. >> >> The IRLZ44Z has a max Rdson of 0.02 at 25C and 5V. The typical gate >> drain charge is 12nC. >> >> The 2SK3704 has a max Rdson of 0.021 at 25C and 4V. The typical gate >> drain charge is 10nC. >> > >Since you are in dire need to get a demo prototype out, some more ideas: > >You could try a snubber from Q2 source to GND. Maybe at least some of >the ring-out can be muffled. It may not really boost efficiency much but >at least move some dissipation away from the FETs. The snubber resistor >needs to be able to stomach that. Ok, I know that is sort of a last >resort sledge hammer method in this case. This paper might help a bit: > >http://etd.uwaterloo.ca/etd/y6chen2005.pdf Thanks. I should have the FET Hammy suggested tomorrow. If that doesn't work well enough I'll try the snubber idea. > >Other: Does this board have a full ground plane? If not can you kludge >at least a local one around the LM3150? Is the line to the SW pin nice >and fat and of low inductance? If needs to be so the bootstrap cap has >something to "lean onto". Full ground plane for this area. I'll boost the track to the SW and boost pins.
From: Joerg on 19 Feb 2010 16:59 Raveninghorde wrote: > On Fri, 19 Feb 2010 11:58:35 -0800, Joerg <invalid(a)invalid.invalid> > wrote: > >> Raveninghorde wrote: >>> I'm back looking at an on going project. >>> >>> http://www.zen88234.zen.co.uk/design/Output-2.png >>> >>> I am investigating the dissipation of Q2 the high side FET. The >>> designers prototype (20 miles away) has an IRLZ44Z. My prototype gas >>> an 2SK3704 fitted. >>> >>> The temperature rise is about the same for both FETs though I expected >>> about a 10% improvement with the 2SK3704. >>> >>> The heatsink is an SW50-4 with a theta of 8.6C per watt. I calculate a >>> FET dissipation of under 2.5W using the methodology in the LM3150 >>> datasheet. >>> >>> The heatsink rises by about 56C after about 20 minutes at 12V 8A >>> running at 200kHz. This T rise drops to 47C at 140kHz. >>> >>> So I scoped the gate drive with the scope grounded on the source of >>> Q2. I got this: >>> >>> http://www.zen88234.zen.co.uk/photos/IMG_0195.JPG >>> >>> I repeated the measurement with my old Iwatsu analog scope and got the >>> short negative pulse after 100ns but it only goes down to -1V instead >>> of -3V. >>> >>> As a check I scoped the probe's ground point and there is very little >>> signal there so the probe isn't picking much radiated noise. >>> >>> The designer (bro in law) has tried the same measurement using his >>> Tektronix scope and gets a pulse down to +1V. >>> >>> Clearly there is a gate drive issue of some sort. >>> >>> The IRLZ44Z has a max Rdson of 0.02 at 25C and 5V. The typical gate >>> drain charge is 12nC. >>> >>> The 2SK3704 has a max Rdson of 0.021 at 25C and 4V. The typical gate >>> drain charge is 10nC. >>> >> Since you are in dire need to get a demo prototype out, some more ideas: >> >> You could try a snubber from Q2 source to GND. Maybe at least some of >> the ring-out can be muffled. It may not really boost efficiency much but >> at least move some dissipation away from the FETs. The snubber resistor >> needs to be able to stomach that. Ok, I know that is sort of a last >> resort sledge hammer method in this case. This paper might help a bit: >> >> http://etd.uwaterloo.ca/etd/y6chen2005.pdf > > Thanks. I should have the FET Hammy suggested tomorrow. If that > doesn't work well enough I'll try the snubber idea. > Just don't hold your breath with that FET. I don't know why they state such long rise/fall and delay times under "Switching Characteristics". Maybe too much gate path resistance, doesn't look too promising but definitely worth a shot anyhow. > >> Other: Does this board have a full ground plane? If not can you kludge >> at least a local one around the LM3150? Is the line to the SW pin nice >> and fat and of low inductance? If needs to be so the bootstrap cap has >> something to "lean onto". > > Full ground plane for this area. I'll boost the track to the SW and > boost pins. Main concern is between boost cap and pins, the rest is not that critical. Same for the VCC bypass caps. -- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
From: Hammy on 19 Feb 2010 17:49 On Fri, 19 Feb 2010 13:59:38 -0800, Joerg <invalid(a)invalid.invalid> wrote: > >Just don't hold your breath with that FET. I don't know why they state >such long rise/fall and delay times under "Switching Characteristics". >Maybe too much gate path resistance, doesn't look too promising but >definitely worth a shot anyhow. Those are some pretty beefy FETS. If you want low switching times your going to have to pay. Even The HUF device you suggested has similar switching characteristics for almost twice the price. I'm going test that EL cheapo STP55NF06L and see just how fast it can switch vs the data sheet. >>> Other: Does this board have a full ground plane? If not can you kludge >>> at least a local one around the LM3150? Is the line to the SW pin nice >>> and fat and of low inductance? If needs to be so the bootstrap cap has >>> something to "lean onto". >> >> Full ground plane for this area. I'll boost the track to the SW and >> boost pins. > > >Main concern is between boost cap and pins, the rest is not that >critical. Same for the VCC bypass caps.
From: Joerg on 19 Feb 2010 17:59
Hammy wrote: > On Fri, 19 Feb 2010 13:59:38 -0800, Joerg <invalid(a)invalid.invalid> > wrote: > > >> Just don't hold your breath with that FET. I don't know why they state >> such long rise/fall and delay times under "Switching Characteristics". >> Maybe too much gate path resistance, doesn't look too promising but >> definitely worth a shot anyhow. > > Those are some pretty beefy FETS. If you want low switching times > your going to have to pay. Even The HUF device you suggested has > similar switching characteristics for almost twice the price. > I didn't suggest those, I suggested to re-do the thing with a better PWM chip, meaning one that can pounce onto the gates with 10V or more :-) > I'm going test that EL cheapo STP55NF06L and see just how fast it can > switch vs the data sheet. > Good. I found that switching times in FET datasheets are often not very dependable. I've had some that I could swing a lot faster. So it's always best to try, maybe your FETs will indeed do the trick. I guess we'll know soon if the UK has parcel service on Saturdays. [...] -- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM. |