Knee Voltage : MOSFET

hi,
I am a newbie at labview.
I wanted to know how to find the knee voltage from the characteristic curve of a mosfet transistor. ( I am working with experimental data which are not IDEAL )
My inputs are 2 arrays of Voltage and Current
My approach :
I was hoping to curve fit the initial part of the values onto a quadratic curve : since my input is an array : i will start a loop(counter i ) from 1 - i - n ( size of array ) and fit the values from 1- i in a quadratic curve and find the value of i which has the least value of error. I used the standard curve fitting tool to do this and generate the mean squared error from that but i am getting a problem from of converting the 2 arrays ( current and voltage ) into one signal which would serve as the input to the curve fitting tool in labview( PLEASE HELP HERE )
I also wanted to know
1) is my approach correct
2) is there any alternative approach
3) my ultimate aim is to find the mobility factor of the mosfet using the Saturation voltage i generate from this approach (Saturation voltage i am assuming is approx equal to knee voltage )
Kindly reply ASAP and for any further information : [email protected]
Regards
Prasad P Iyer
IIT Kanpur
India

Hi Prasad,
I removed those ExpressVIs, the VI should still do the same...
Could you attach some sample data?
Best regards,
GerdW
CLAD, using 2009SP1 + LV2011SP1 + LV2014SP1 on WinXP+Win7+cRIO
Kudos are welcome
Attachments:
Knee Voltage.vi ‏15 KB

Similar Messages

Voltage Regulator problem !

Respect to All Experts !
Last month I had blown up my MSI 790FX GD70 by increasing "CPU VDD Voltage" to 1.45volts
I found this review http://www.legitreviews.com/article/1004/6/ relating to my problem bcoz they killed GD70 too by overvoltaging vCore.
According to them some first launched GD70s had buggy MOSFET/Voltage Regulator (9A2S292) which were not able to live at 1.45v. Unfortunately I bought one of them. My dead mobo had same MOSFET (9A2S292).
And then they successfully overclocked another new GD70 with new MOSFET (xxxx298) http://www.legitreviews.com/article/1103/1/
I got my dead mobo RMAed now
I have new mobo, new MOSFET
But this one has a different MOSFET (xxxx294)
I only wanna know if this one is strong enough to take voltage upto 1.45v or more ? Because I've read in many people are ok with 1.45v with this board.
I am stuck at 4.1GHz at 1.435v and I am also very afraid of increasing voltage more than this.
My Computer:
MSI 790FX GD70 (Latest BIOS)
Phenom II x6 1055T 2.8GHz, 125watt TDP
2x2GB ddr3 1333mhz
2xSeagate 500 in RAID0
Corsair TX850 V2 PSU
XFX HD 5770

Quote
What will happen if I increase voltage to 1.45v in "CPU Voltage" and leave "CPU VDD Voltage (MOSFET)" at default ?
Doubtful anyone can predict the outcome. I suppose it's always possible to ask MSI if they will modify your mainboard or exchange it for one with the xxxx298 & what the cost would be?

Convergence error in mosfet driver NCP5355D circuit, might be bootstrap mistake?

Hi there,
I'm using the Multisim 13.0.1. There's a 8-pin mosfet driver NCP5355D available in the library. I wanted to use it for switching half-bridge MOSFETs, but a simulation error has occured. The report said that 'Transient time point calculation did not converge'.
I've included the circuit with the gate driver, and also the circuit with separate pulsed voltage sources for switching MOSFETs, which could provide a general view of what I intended to achieve with the gate driver.
I also tried to configure the gate driver step by step- I started from the VS, CO, EN, PGND pins side and got no simulation errors then; but when I started connecting the BST and DRN pins, convergence error occurred. So is the convergence error resulting from any mistakes in connecting the BST and DRN pins?
To make it easier, the datasheet of NCP5355D is also attached here, where example application connections are included.
Expecting your insights and thanks in advance!
Solved!
Go to Solution.
Attachments:
cap2-test1_2clk.ms13 ‏183 KB
NCP5355-D.PDF ‏104 KB
cap2-test1_2clk_5355.ms13 ‏172 KB

I do not have Multisim so I cannot see your circuits but have some general observations.
1. MOSFET driver models often use switches and these can lead to convergence problems. This is because the switching action introduces a strong non-linearity into the model. I wrote my own model using transmission lines for delay and non-linear dependent sources with tanh functions rather than hard switches.
2. If you do not need to model the detailed behavior of the driver, using pulsed sources with rise/fall times as long as possible without changing the behavior of the MOSFETs might be a better choice.
3. There are books written on how to deal with SPICE convergence problems. This is what I found helpful from one of them:
** From Inside Spice by Kielkowski
** Convergence aids:
** Select appropriate RELTOL
** set VNTOL = RELTOL*Vsmall
** set ABSTOL = RELTOL*Ismall
** set GMIN = 1/Rp where Rp is parasitic resistance which
*** will have minimal effect when added to circuit.
** semiconductor models should have non-zero resistances
** RS (diode), RE, RC (BJT), RD, RS (FET)
** Raise ITL1=500 (DC OP iterations limit)
** use .NODESET
** use source stepping ITL6=500
** Transient convergence aids
** Add capacitance to device models:
** CJO (diode), CJE, CJC, CJS (BJT), CDG, CGS (JFET)
** CGDO, CGSO, CGBO, CBD, CBS, CJ, CJSW
** increase ITL4=40
Good luck.
Lynn

MSI K9A2 Platinum has weak MOSFETs

Well gang I am on my fourth K9A2 Platinum board. The first two I got were just plain unstable. The third one I got and put into my kids computer is very stable. I bought a replacement for my second one and I have to say with 8GB (4x2gb), 9850BE, SB Fatal1ty, 4x 3870 voltmodded to 1.675V at 985/1210 this board was running every test perfectly. I was able to get the FSB up to 230 with DDR-920 5-5-4-10-2T and it ran 10 hours stable on Prime95 (4 instances, affinity) with 1920 MB on each instance.
Well I figured without a working northbridge multiplier this was darned good so I started on the CPU. I bumped the CPU voltage to 1.6V in the BIOS and set the multiplier to 13 for an overclock of 2990MHz.
Started Prime95, came back in a couple hours and the system was OFF.
When I press the power button only a flicker on the mobo LEDs then nothing.
Decided to unplug everything except I forgot to unplug the 4-pin power connector, then I jumpered pins 4 and 6 on the 24-pin power supply. A MOSFET decided to go up in smoke.
Did I mention I am water cooling everything now? NB, SB, CPU, VRM, memory, video cards... I checked and the VRM heat transfer pad was secure on and got burnt from the pop.
All I can say is, gang, after working with this motherboard for almost 6 months now and having nothing but trouble I was ***SO*** encouraged to get everything working so stable. I did not experience a single cold boot loop with this new board either. Then POP... I'm so upset now... this board can't handle the settings in its own BIOS.
I have a DFI Lanparty I bought with a Promise FTTX4650. I am going to switch to that now. It has a digital PWM. This shouldn't happen there. I think it does PCIE slots in a x16/x4/x4 configuration though so I am not entirely encouraged and it only has room for 3 of my 4 video cards. It's either that or suffer with PCIe 1.1 on a Skulltrail and drop $1500 for a CPU.
Well anyway, this board in general has so much potential. It's too bad MSI just didn't manufacture or support it properly (no northbridge multiplier, no PCIe setting in the BIOS, etc).

At work I just put together a Gigabyte X48 based board and the thing has 14 chokes (the square blocks) labeled R80 while the K9A2 Platinum has just 5 I believe labeled R34 or R45 (I forget, I'm too lazy to go look). It does appear this board is a bit weak on the VRM side. I see more MOSFETs on the K9A2 Platinum than on the X48 board I saw, however. I'm sure everyone will be using a Digital PWM/VRM pretty soon.
Honestly if your board is giving you fits return it to where you bought it and ask for a new one. It took MSI 3-4 weeks I think to replace one of mine and it looks like the "repaired" it but I have no idea what they actually did to it... Anyway that's my advice. The newer ones MFG in 2008 seem to be pretty good. The Promise SAS chip onboard is incredibly fast at RAID0 and 4 PCIe 2.0 slots at 8x/8x/8x/8x is really, really sweet. This board is still very unique in the market, I think these two features make it that way. I would have paid $50 more for the same board, just with higher quality caps/chokes/VRM/etc on it.
Looking forward to the 45nm Phenoms, that's all I can say. I seem to have topped out at 2950MHz on the Phenom 9850 I have. I took one core up to 3.3GHz for a while under Prime95 but haven't been able to do it again (BSOD: failed to receive a secondary processor interrupt)...

NB voltage control not working

I'm on a 790XT-G45 (2.5 BIOS although I have tried other BIOS') which supposedly supports a 2600 MHz NB. There is nothing I can do to even get the board to post with a 2600 NB and raising the NB voltage doesn't seem to do anything either. If I can get into Windows (only sometimes) @ 2400 MHz NB, and then increase voltage through overdrive, it stays stable.
So the question is how can I raise the NB voltage on this board? The BIOS setting that raises NB voltage in conjunction with HT link doesn't seem to really do anything.
Otherwise does anyone know a way to get Overdrive to allow you to change the NB multiplier so I can just overclock it in Windows? I'm not happy with a 2400 NB and even worse it hardly will ever get into Windows with stock voltage and this frequency so really I'm stuck below 2200 MHz.
Should I RMA the board?

Quote from: wlahdone on 13-February-11, 21:31:36
I have opened up AMD Overdrive ,Enabled Advanced ,Clock voltage andi see CMOS Items-> CPU NB Multi and NB Vid present ,but i presume you need to enable ACC to change them.I did not play with this values though.I have never gone to more than 2400 M with my CPU on any motherboard with mild voltages for the NB.But either the BIOS menu would freeze either Windows freezes.There is an increase in the speed games are loaded but thats it .I doubt you can achieve more on another motherboard.You are beating a dead horse in my opinion.Do some CPU clock overclocking better with a proper cooler.Try a 1T timing for RAM and you will get more on the memory performance side.
I have all those enabled and I still cannot access the cmos items section. Can you tell me specifically what you have set? I can adjust NB VID under the voltages section but I cannot adjust the mutiplier in the CMOS items section.
MANY, MANY users are reporting having ABSOLUTELY NO PROBLEMS getting a 2600 MHz NB-CPU frequency, and many are pushing that easily to 2800 and even 3000 MHz. According to AMD themselves, (http://sites.amd.com/us/Documents/AMD_Dragon_AM3_AM2_Performance_Tuning_Guide.pdf, scroll to the bottom), you can expect 2800-3000 MHz on the NB link, with 1.35-1.5v - basically you can run your NB voltage as high as your CPU.
As far as overclocking, I am pushing 4.2GHz, on high end water (CPU & GPU) and I have added MOSFET coolers to the board so I'm pretty sure I'm good. Regardless I would like to push the board as far as it will go, and even at stock frequencies, and two different CPU's, I cannot get the NB frequency stable above 2300 MHz or so. Look it up, people are reporting better performance with a higher NB frequency, especially with higher clocked modules and overclocking.
All I'm expecting this board to do is what all the other boards can do - to be able to increase the NB frequency to at least 2600 MHz. If I am beating a dead horse, it's because MSI users have given up and accepted lack of support as a status quo. It would probably take no time at all to add a NB-CPU voltage adjustment to the BIOS. And yes I have contacted them so we'll see what happens. If this board is supposed to support up to 5200 M/T, then I would like to see it.

Chipset voltage adjustment (K8n neo2 plat)

I posted this in overclocking, with no luck. Sorry to be repetitive . . .
Where in the BIOS do I adjust the chipset voltage?

My ram happily does 290 at 2.5-4-3-10. I haven't yet tested how it handles higher timings. But, it seems to perform quite nicely.
What I want to do is 333X7.5 with 5:6 mem divider, just to see how the increased FSB goes. Anyone know if there are hard voltmods (you know with sawter, where you just set your warranty card on fire) for the chipset voltage?
My setup:
MSI K8N Neo2 Platinum
A64 3000+ 90nm
1 GB OCZ PC3200 EL platinum Rev2
BFG 6800GT OC at 425/[email protected]
Stuff, things, junk, wires, and some random-ass bits of copper glued to all sorts of things. About to break out the mini-heatsinks for every flat, non-conductive thing on my mobo, starting with MosFets, and working up from there.

MOSFETs don't work

I have been trying to get a N-channel MOSFET to work for a project. I'm using a standard n-channel mosfet part number BUZ10. I''m running 13V and 250mA through the source, 13V and some odd pA to the gate, and the output is 13V and around 1-10 pA. I need to keep the same number of amps through the FET. I've changed the circuit to give more current and less voltage to the gate (about 3v and 1mA).
I've tried other FETs with different specs and can't get more then maybe 100nA which is way to low for what the project requires. I'm I doing something wrong?

i to have a problem with the mosfets not working. 32khz pulse at plus and minus 4 volts doesn't appear on collector. just a 15 volt level.
Attachments:
VCRCAM2 TRANSMITTER.ms10 ‏484 KB

Lowest DRam voltage settings on modern AM3 boards?

Greetings all,
while I am considering an Intel socket 1156 motherboard for the next machine mainly because I want to reduce idle power consumption/wastage, I'd like to know whether MSI socket AM3 motherboards support low voltage memory. Can it be lowered to 1.25 volt? I'm not averse to specifying DRam parameters in the bios if need be. The memory kit I had in mind is the 1.25v Kingston HyperX KHX1333C9D3UK2/4GX.
Considering the native SATA3 support on the 850 southbridge, an AM3 would also be a great excuse to get a Crucial RealSSD, wouldn't it?

While I can't answer the Dram voltage question I can speak to the MOB question in general. I was forced to buy a new MOB this summer due to a liquid cooling accident that killed my liquid cooled ASUS M4A-79T deluxe. I tried a Gigabyte 890FXA UD5 MOB and was not happy with some of it's features so I returned it for an MSI 890FXA-GD70. As a systems engineer with decades of experience I consider this board to be the best laid out of all the current AMD 890/850 supporting MOBs that also support USB 3.0 and SATA 3. I am currently liquid cooling this MOB - North and South bridges, MOSFET voltage regulators, and Kingston HyperX performance 2133MHz memory with built-in liquid cooling. AMD 965 BE CPU has been running liquid cooling since day one using Swiftech Apogee GTZ liquid cooling block. After working with this board with liquid cooling blocks I am even more convinced it is the best laid out of this series of AMD 890/850 MOBs out there! If you are still waiting to pull the trigger you may want to consider this MOB. It is neither the cheapest nor the most expensive of the MOBs by the big 3, ASUS, Gigabyte, MSI, but I have found it to be a good board so far. To your question of power if you look on Tom's HW web site you'll find comparisons of like Intel vs AMD CPUs. While the Intel's are a bit faster in most area's, not all- check the bench marks, the AMD CPUs do well against the Intel's power consumption wise and if you OC your systems as I do, I can beat the Intel machines standard operating CPU clock speeds in speed and for 1/4 of the price. I got to 4.2 GHz without bothering to stretch this system at all(I don't stretch a system until I've finished cooling the MOB's North+ South bridges and MOSFET areas. You mentioned you like to keep your systems for a long time, well so do I, baring accidents I generally keep a system for at least 5 years and during that time they still perform reasonably well. I still have a 1999 Slot A MSI MOB system that will work using XP OS, does run a bit slow because of 3/4 GB memory limitation.
One last final comment about the Intel MOBs that support USB 3.0: they have a tendency to not do the USB 3.0 well due to PCI E lane limitations on the Intel controllers.

7830R digital output voltage

Hi,
I am using a 7830R DAQ card to work with MOSFETs as switches.
We have a 25-MOSFET array. Each MOSFET is controlled by a DIO line of the 7830R card.
When the DIO line is at 0V, the MOSFET is off, and when the DIO line is at 5V, the MOSFET is on.
However, the output voltage automatically drifts to 5V and gradually
turns on the MOSFET if I do not keep updating the voltage of the line.
Can I set the default output voltage at 0V once for all?
Thanks.
Guang

Gaung-
I believe some clearification is in order. Are your DIO lines floating to 5V while your FPGA VI is running or are they floating after the FPGA is reset (usually the cloase FPGA reference resets teh device)? The DIO lines are expected to float when not being explicitly written to. You have a few options:
1) Set the default power up states via the instruction in this post:
http://forums.ni.com/ni/board/message?board.id=170&message.id=221423#M221423
2) Dont reset your FPGA VI when you close it. This will continue to drive the digital lines in thier last state.
3) Use pull down resistors on the DIO lines in question to force lines to 0V when not being driven
Xaq
Applications Engineer
National Instruments

975x PowerUP Edition : Voltage measurement points with DMM - Where?

Does anybody know where on this mobo I can measure various voltages, in particular vdd (mem) and vcore?
I'd like to be able to see how inaccurate most of the software monitoring apps I use are

Quote from: Oratiek on 23-August-06, 04:14:34
Thanks for the input so far guys.
I just want to double check...
Click to enlarge
For vdimm...
RacerX says top leg of mosfet (number 1 in the pic) = 2.27v
Maesus says north pin of coil (number 2 in the pic) = 5.01v
Even EDO Ram only needed 3.3v so I guess number is the correct measuring point since I've set 2.2v in the BIOS.
As for the vcore, I'll have to figure out a way to get the mobo to stand upright on the floor so I can access the back of the board with both hands
You need to measure the coil under the PCI slot(tough with video card) but it measures the same as the mosfet I mentioned. Still don't understand the vcore measurement but whats important is where to measure this so call vMCH or PCI xpress voltage. To be honest it does nothing in terms of OCing compared to other boards with this option. Any ideas?

Need to a voltage converter to run US-bought 110v HP Printers in 220v Pakistan any recommendations?

I Purchased Three Printers
1.
HP LaserJet Enterprise 500 MFP M525dn(CF116A)
2.
HP Color LaserJet Enterprise CP4025n Printer(CC489A)
3.
HP LaserJet P2055d Printer (CE457A) -
All of them three operates on 110v USA. but i need them to use in 220V . Can anyone recommend me any good Voltage converter ?
This question was solved.
View Solution.

Hi,
Before go out to buy a converter/transformer (you need over 2KW for all 3 of them), please check the switches at the back, they may have switches to switch from 110V to 220V. I don't know your market, my suggestion: talk with an electrician who knows the real world much better.
Regards.
BH
**Click the KUDOS thumb up on the left to say 'Thanks'**
Make it easier for other people to find solutions by marking a Reply 'Accept as Solution' if it solves your problem.

Voltage question for Inspire 5200 5.1 Speakers

Hi, I'm new to the forum, so I'm not sure if this question has been answered in the past, but...
I just moved back States-side after living in Pakistan for about 6 months, and I brought my Pakistan-bought Inspire 5200 speakers. I don't know if I should buy a voltage converter or whatever.
On the AC adaptor, it says "PRL. :230V~50Hz 320 mA... SEC. : 2V ~ 4.2A". I read up a little on this stuff, and, unless I'm mistaken, in Pakistan, the voltage is 220V and 50Hz, while in the US, its 110V and 60Hz.
Is it safe to use the speakers here WITHOUT a voltage converter or a step-down transformer?
If it isn't safe, then what specific things do I need to pick up from Radioshack to make sure my speakers work and don't short-circuit or fry?
Thanks.

If you use the right step down transformer than it's perfectly safe. The problem is, step down transformer worsen the sound a lot unless you use the very expensi've one. But if you can't hear the different than it's good then, trust your own ear anyway. I think the best way is to change the adaptor.

In Brazil. . . Do I need an adapter to plug in my HP Mini 110 into a 220 Voltage plug?

Hi! I'm living in Brazil for a while and just got here. I have an HP Mini 110 and the plugs here say 220V. Is it safe to plug in my computer or do I need some kind of converter? If so what kind?
My power is about to run out!!!!!
Thanks in advance,
Sasha

No. The power brick for the iPod senses the voltage and adjusts for the different ones.
You will need an adapter to switch form flat us pins tot the round pins used on their plugs.
Allan

Guys, I'll be receiving a MacBook Pro bought in the US as a gift. I'm from UK, will the voltage difference (110v to 220v) and warranty/insurance coverage be an issue? Any other issues to face?

Thanks guys...

Normally the voltage difference is not a problem because the charger can handle both.
I am not sure about the adapter plug: the wall sockets in UK and USA may be different, use a simple (earthed!) adapter for that if necessary.
The rest should be the same, although two keys on the keyboard might be different (@ is shift-2, try to find the other...).
gratulations with your gift,
Lex

I have a Macbook Air purchased in 2011 in the US. I am traveling in South Africa and am unsure if I need a step down voltage converter before I plug my laptop into the wall to charge it (using an outlet adapter). Am I ok to just plug in?

So standard electrical output from wall outlets in the US is 110V - 120V, and standard output from wall outlets in South Africa is 220V - 240V. In SA I know I can plug in an iPhone using the iPhone charging cable and it will automatically step down the voltage, thus preventing my phone from overheating. I cannot plug in a hairdryer or flat iron without step down voltage converter in SA. Can I use the charger that came with my US 2011 Macbook Air and plug into a wall outlet in South Africa or do I need to purchase a step down voltage converter in order to charge my laptop while in SA?

You only need a plug adaptor as long as the voltage is between 100 and 240 volts.
(123957)

Knee Voltage : MOSFET

Similar Messages

Maybe you are looking for