• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Question About MOSFET Gate Driver IC

Status
Not open for further replies.

vne147

Member
I have a circuit I'm working on in which I need to switch up to 25A @ 48V to a mostly resistive load that will be in the range of 0.1Ω - 10Ω. Every other time I supply power to the load, I need to switch polarity. For the prototype I have been using 2 SPDT relays to accomplish this. But for reasons of cost and reducing the number of moving parts, I'm looking to change this arrangement to an H-bridge.

I've looked around for MOSFETs that meet the operational requirements listed above and also have a very low RDSON, so they don't get too hot. I've found quite a few N-channel MOSFETs that fit the bill, but the P-channels (at least the ones I've found) seem to have a higher RDSON and have fewer options overall. For that reason, I'm going to try to employ the H-bridge arrangement using only N-channel MOSFETs.

The problem is, that the highest voltage I have in the circuit (48V) is also the voltage I'm switching. So, it won't always be possible for the high side MOSFETs to maintain the necessary VGSTH. For this reason, I started looking at using a gate driver IC for the high side MOSFETs.

Now finally to my question:

I've found a few gate driver ICs that have very high output currents. The LTC4441 for example, can output up to 6A. Why would this much current every be necessary to drive the gate of a MOSFET? I can't for the life of me imagine an instance where you'd need 6A to turn on a MOSFET, or even a bunch of MOSFETs. Knowing that the people who design these parts aren't dummies, and knowing that I don't understand why I would ever need to use the full 6A this part can source, makes me think I don't understand something about the proper use for this IC, or worse yet something fundamental about how MOSFETs work.

So, can someone please explain this to me? You can talk to me like I'm 5. You won't hurt my feelings.

Also, if you think I'm going about this H-bridge arrangement the wrong way by trying to do it with all N-channel MOSFETs, please don't be shy.

Thanks in advance for your help.
 

alec_t

Well-Known Member
Most Helpful Member
I can't for the life of me imagine an instance where you'd need 6A to turn on a MOSFET, or even a bunch of MOSFETs.
The gate of a power MOSFET has an inherent high capacitance (think 2nF) which needs to be charged/discharged when the FET is turned on/off. If the FET is being switched at high frequency (seemingly not in your case), charging a high capacitance in a short time requires a high current.
Note that the high-side FET driver ICs use a bootstrap arrangement to create a voltage above the drain voltage. This bootstrapping requires the lower end of a bootstrap capacitor to be periodically toggled high/low. If your FET polarity switching is infrequent then you may need some other way to achieve this toggling for the bootstrap arrangement to work correctly.
 
Last edited:

vne147

Member
The gate of a power MOSFET has an inherent high capacitance (think 2nF) which needs to be charged/discharged when the FET is turned on/off. If the FET is being switched at high frequency (seemingly not in your case), charging a high capacitance in a short time requires a high current.
Note that the high-side FET driver ICs use a bootstrap arrangement to create a voltage above the drain voltage. This bootstrapping requires the lower end of a bootstrap capacitor to be periodically toggled high/low. If your FET polarity switching is infrequent then you may need some other way to achieve this toggling for the bootstrap arrangement to work correctly.
This makes a lot of sense. Thanks for the explanation. The datasheet for the part I referred to above does say that the peak current is up to 6A. That implies it can do that for only a short time, or for a small percentage of the time.

I guess that since in my application I won't be switching the load on and off fast enough for a bootstrapping arrangement to work, that I'll need to seek alternate solutions. One other idea ran across was using a voltage doubler to attain the necessary voltage above the drain voltage. Any thoughts on that? Also, have any ideas about where I should look for a suitable P-channel MOSFET, or am I simply constrained by what's available/possible?

Thanks.
 

alec_t

Well-Known Member
Most Helpful Member
What switching rate do you anticipate using?
Apart from the 48V, is another supply used to control the switching?
 

vne147

Member
What switching rate do you anticipate using?
Apart from the 48V, is another supply used to control the switching?
The switching will be very slow. Ultimately, it will be determined by a control algorithm on the μC, but it will likely never be switched on and off faster than maybe once every couple of minutes. I do have a PWM source on the μC I could use for some sort of voltage boost if need be though.

As far as other voltage sources aside from the 48V, on the board I have 3.3, 5, and 12 VDC sources any of which I could use for switching.

Also, I'm not sure if it's pertinent but I'll say it anyway, the 48 VDC source won't always be 48V. It's variable up to 48V.

Thanks!
 

spec

Well-Known Member
Most Helpful Member
I have a circuit I'm working on in which I need to switch up to 25A @ 48V to a mostly resistive load that will be in the range of 0.1Ω - 10Ω. Every other time I supply power to the load, I need to switch polarity. For the prototype I have been using 2 SPDT relays to accomplish this. But for reasons of cost and reducing the number of moving parts, I'm looking to change this arrangement to an H-bridge.

I've looked around for MOSFETs that meet the operational requirements listed above and also have a very low RDSON, so they don't get too hot. I've found quite a few N-channel MOSFETs that fit the bill, but the P-channels (at least the ones I've found) seem to have a higher RDSON and have fewer options overall. For that reason, I'm going to try to employ the H-bridge arrangement using only N-channel MOSFETs.

The problem is, that the highest voltage I have in the circuit (48V) is also the voltage I'm switching. So, it won't always be possible for the high side MOSFETs to maintain the necessary VGSTH. For this reason, I started looking at using a gate driver IC for the high side MOSFETs.

Now finally to my question:

I've found a few gate driver ICs that have very high output currents. The LTC4441 for example, can output up to 6A. Why would this much current every be necessary to drive the gate of a MOSFET? I can't for the life of me imagine an instance where you'd need 6A to turn on a MOSFET, or even a bunch of MOSFETs. Knowing that the people who design these parts aren't dummies, and knowing that I don't understand why I would ever need to use the full 6A this part can source, makes me think I don't understand something about the proper use for this IC, or worse yet something fundamental about how MOSFETs work.

So, can someone please explain this to me? You can talk to me like I'm 5. You won't hurt my feelings.

Also, if you think I'm going about this H-bridge arrangement the wrong way by trying to do it with all N-channel MOSFETs, please don't be shy.
This makes a lot of sense. Thanks for the explanation. The datasheet for the part I referred to above does say that the peak current is up to 6A. That implies it can do that for only a short time, or for a small percentage of the time.

I guess that since in my application I won't be switching the load on and off fast enough for a bootstrapping arrangement to work, that I'll need to seek alternate solutions. One other idea ran across was using a voltage doubler to attain the necessary voltage above the drain voltage. Any thoughts on that? Also, have any ideas about where I should look for a suitable P-channel MOSFET, or am I simply constrained by what's available/possible?
Hi VNE,

My advice is not to mess about and use a gate driver for the reasons that Alec says in post #2. Especially as the effective gate capacitance for the type of MOSFETs you will need will likely have an order higher effective gate capacitance (gate charge) of around 20nF.

Even if you do not need a gate driver it makes no sense not to use one. Gate drivers are cheap, small, and freely available.

It is quite true that the performance of PMOSFETs lags NMOSFTs, but all the same there are plenty of PMOSFETs that would do your job. There is a list of PMOSFETs in the image below for you to consider. You can also search the DigiKey data base for suitable PMOSFETs- just enter the parameters you want (80V minimum and 40A minimum).

So I would advise:
(1) Use one dual gate driver for each half of the bridge (one channel non inverting and one channel inverting)
(2) Use gate stopping/shaping resistors
(3) Use an NMOSFET and PMOSFET in each leg of the bridge UPDATE 2017_01_30 (but see later posts about using all NMOSFETS)

spec

2017_01_29_Iss1_PMOS_POWER_TRANSTOR.png
 
Last edited:

vne147

Member
Hi VNE,

My advice is not to mess about and use a gate driver for the reasons that Alec says in post #2. Especially as the effective gate capacitor for the type of MOSFETs you will need will likely have an order higher effective gate capacitance (gate charge) of around 20nF.

Even if you do not need a gate driver it makes no sense not to use one. They are cheap, small and freely available.

It is quite true that the performance of PMOSFETs lags NMOSFTs, but all the same there are plenty of PMOSFETs that would do your job. There is a list of PMOSFETs in the image below for you to consider.

spec

View attachment 103909
Spec,

As you recommend, I'm leaning towards using a gate driver regardless. I'll start going through the list you posted for the PMOSFETs.

Thanks!
 

spec

Well-Known Member
Most Helpful Member
Spec,

As you recommend, I'm leaning towards using a gate driver regardless. I'll start going through the list you posted for the PMOSFETs.

Thanks!
No problem VNE.

spec
 
Last edited:

crutschow

Well-Known Member
Most Helpful Member
If you want to use all N-MOSFETs, then you could likely use a high-side driver with a charge pump, such as one of those listed here to drive the top FETs.
 

hyedenny

Active Member
I have a circuit I'm working on in which I need to switch up to 25A @ 48V to a mostly resistive load that will be in the range of 0.1Ω - 10Ω. Every other time I supply power to the load, I need to switch polarity. For the prototype I have been using 2 SPDT relays to accomplish this. But for reasons of cost and reducing the number of moving parts, I'm looking to change this arrangement to an H-bridge.
You're not using this in your (Pipistrel? Tecnam?) airplane, are you?!?
 

vne147

Member
If you want to use all N-MOSFETs, then you could likely use a high-side driver with a charge pump, such as one of those listed here to drive the top FETs.
I'm still trying to figure out if I can get away with using PMOSFETs for the high side. From the list Spec sent most of the PMOSFETs have a relatively high RDSON compared to their NMOSFET brothers, which at the currents I need would make them get a little too hot. I know there are heat sinks and other shenanigans I can try but finding a solution that simply doesn't dissipate too much power would be the best way. There is one from the list though (SMP3003) that may fit the bill. I'll keep your suggestion in mind though if I jump the PMOSFET ship and go back to NMOSFET town.

Thanks!

You could use a h-bridge gate driver such as the MC33883.

Mike.
At first glance this seemed like exactly what I needed. The only problem though is that VCC needs to be between 5.5 - 55V, and the voltage I'm looking to switch won't always be above 5.5V. The simplified application diagram on the first page of the datasheet shows VCC of the IC connected to the drains of the high side MOSFETs. I'm not sure if that's a requirement. I'm still looking into it.

If it isn't, I could connect VCC of the IC to a constant 12V source, and the drain of the high side MOSFETs to the variable voltage that would be anywhere between 0 and 48V. Any ideas is that would be a problem?

Thanks!


You're not using this in your (Pipistrel? Tecnam?) airplane, are you?!?
No sir. This is for a strictly ground based application.
 

spec

Well-Known Member
Most Helpful Member
POST #13 Issue 3 of 2017_01_30

Hi VNE,

At first glance this seemed like exactly what I needed. The only problem though is that VCC needs to be between 5.5 - 55V, and the voltage I'm looking to switch won't always be above 5.5V. The simplified application diagram on the first page of the datasheet shows VCC of the IC connected to the drains of the high side MOSFETs. I'm not sure if that's a requirement. I'm still looking into it.

If it isn't, I could connect VCC of the IC to a constant 12V source, and the drain of the high side MOSFETs to the variable voltage that would be anywhere between 0 and 48V. Any ideas is that would be a problem?
I have had a further look at some circuits to meet your requirements. With the benefit of the suggestions that the other members have posted, below is what seem to be viable options. I don't think there are any problems that you have mentioned, unless I have miss-interpreted what you said.

(1) Discrete approach:
2 * power NMOSFET
2 * power PMOSFET
3 * small signal NMOSFET
6 * resistor
(2) Half bridge driver approach #1:
2 * power NMOSFET
2 * power PMOSFET
2 * half bridge chip UCC21520
1 * LM7912 negative voltage regulator
(3) Half bridge driver approach #2:
4 * power NMOSFET
2 * half bridge chip UCC21520
1 * 12V @ 500mA min isolated power supply (wall wart)
(4) Full bridge driver approach #1:
2 * power NMOSFET
2 * power PMOSFET
1 * full bridge chip MC33883
1 * LM7912 negative voltage regulator
(5) Full bridge driver approach #2:
4 * power NMOSFET
1 * full bridge chip MC33883

spec

DATASHEETS
 

Attachments

Last edited:

alec_t

Well-Known Member
Most Helpful Member
:confused: LM7812 is a +ve reg.
 

vne147

Member
Spec,

I'm a little confused about how exactly I'd use the 7912 negative voltage regulator. Can you elaborate please?

Thanks.
 

spec

Well-Known Member
Most Helpful Member
Spec,

I'm a little confused about how exactly I'd use the 7912 negative voltage regulator. Can you elaborate please?

Thanks.
Sure,

Assume that the high side of the bridge are PMOSETs and also assume that the PMOSFETs require -12V between the PMOSFET's gate and drain to turn the PMOSFETs fully on.

With these assumptions, you need a gate drive signal of 48V (off) to 36V (on), so the negative voltage regulator generates 36V (input = 0V, ground = 48V, output = 36V).

The 36V line is then used to define the negative input line of the top side driver.

Of course, the 12V gate drive is only nominal and the negative regulator voltage of a practical design would match the gate drive requirements of the PMOSFETs actually used. For the SMP3003 power PMOSFETs -9V maximum gate drive would probably be optimum, so in the case of the SPM3003 the negative regulator would be -9V instead of -12V.

I hope that explains what I am on about- things get a bit convoluted with bridge drivers.:)

spec
 
Last edited:

tomizett

Active Member
I'm sure I've heard somewhere about high-side driver ICs that use a photo-electric cell to generate voltage on the isolated side, and so can stay high indefinitely (unlike the bootstrap types we've been discussing here). They're slow as a wet weekend, of course, because they can deliver next -to-no current, but in this application they might just work.

Lunchtime's over now, so I've not got time to try to look up what the parts where called, but it's worth a Google, surely?
 

spec

Well-Known Member
Most Helpful Member
I'm sure I've heard somewhere about high-side driver ICs that use a photo-electric cell to generate voltage on the isolated side, and so can stay high indefinitely (unlike the bootstrap types we've been discussing here). They're slow as a wet weekend, of course, because they can deliver next -to-no current, but in this application they might just work.

Lunchtime's over now, so I've not got time to try to look up what the parts where called, but it's worth a Google, surely?
Unless I am mistaken, the MC3383, that Pommie suggested in post #10, generates a bootstrapped DC supply for the top MOSFETs by using a self oscillating charge pump. It is thus suitable for 'DC' conditions.

Also, option (1) in post #13 requires no top MOSFET auxiliary supply.

spec
 

vne147

Member
Sorry I've been a little slow to respond but I'm trying to fully digest all this information.

Spec,

Thanks for your explanation in post #17. I think I now understand why the 7912 would be necessary, but I'm still a little fuzzy on specifically how I would use it. The data sheet for the UCC21520 only has an example schematic using NMOSFETs which I believe is what your option 3 from post #13 describes. If I swapped out the top side NMOSFET for a PMOSFET, as in option 2, I'd have to do a little thinking about what else I'd need to change, but specifically with respect to the 7912 you said:

input = 0V, ground = 48V, output = 36V
But the 48V connected to the top side drain will be variable. I'll call it VTD. In reality, I'd want the output of the 7912 to always be 12V less than VTD, correct? If that is indeed the case, would I hook up the 7912 as follows?

input = VSSA, ground = VTD, output = VTD -12V

Similarly, if I went with option 3 from post #13, would I hook up the V- output of the wall wart to VSSA?

Tomizett,

Thanks for the suggestion. This option is intriguing. It seems to simplify the problem a great deal. I did some Googling and quickly found the VOM1271. It looks pretty damn perfect for my application. I have no need for fast switching. In fact if it took 5 or even 10 seconds for the MOSFET to turn on, it would matter one bit.

Does anyone have any thoughts on changing direction to use the VOM1271 in lieu of the other driver ICs previously discussed?

Thanks!
 
Status
Not open for further replies.

EE World Online Articles

Loading
Top