| View previous topic :: View next topic |
| Author |
Message |
SherpaDoug
Joined: 07 Sep 2003
Posts: 1640
Location: Cape Cod Mass USA
|
|
 Posted: Thu Apr 10, 2008 7:21 am |
 |
|
I have used a lot of MOSFETs over the years and I don't remember any that had an allowable gate voltage above 20V. That is why I recommended using a voltage divider to cut the 24V gate drive in half.
_________________
The search for better is endless. Instead simply find very good and get the job done. |
|
 |
Matro
Guest
|
|
| Posted: Thu Apr 10, 2008 7:42 am |
|
|
| SherpaDoug wrote: | | I have used a lot of MOSFETs over the years and I don't remember any that had an allowable gate voltage above 20V. That is why I recommended using a voltage divider to cut the 24V gate drive in half. |
I already saw 25V. But here it is still too close to the nominal Vgs.
Personnaly instead of using a N-channel and a P-channel, I would have only one N-channel with source connected to ground and directly controlled with an I/O pin.
Be also aware that if you command inductive loads like motors, the voltage will increase when the circuit will be opened.
Matro |
|
|
newguy
Joined: 24 Jun 2004
Posts: 1909
|
|
| Posted: Thu Apr 10, 2008 8:01 am |
|
|
| SherpaDoug wrote: | | I have used a lot of MOSFETs over the years and I don't remember any that had an allowable gate voltage above 20V. That is why I recommended using a voltage divider to cut the 24V gate drive in half. |
This one tolerates 25V: http://www.fairchildsemi.com/ds/FQ/FQU17P06.pdf
I only spent about 60 seconds searching....I have a feeling there are many others. The voltage divider idea is a very good one too. |
|
|
Matro
Guest
|
|
| Posted: Thu Apr 10, 2008 8:05 am |
|
|
| newguy wrote: | | SherpaDoug wrote: | | I have used a lot of MOSFETs over the years and I don't remember any that had an allowable gate voltage above 20V. That is why I recommended using a voltage divider to cut the 24V gate drive in half. |
This one tolerates 25V:
[...]
I only spent about 60 seconds searching....I have a feeling there are many others. The voltage divider idea is a very good one too. |
More than 25V is very rare because Vgs is limited by the gate insulating dielectric thickness. ;-)
Matro. |
|
|
ckielstra
Joined: 18 Mar 2004
Posts: 3680
Location: The Netherlands
|
|
| Posted: Thu Apr 10, 2008 8:39 am |
|
|
Why is everybody ignoring Don's (ak6dn) post?
I think he made the smart suggestion that switching the low side of the pump is much easier than switching the high side. The BUZ11 will do the job nicely without adding extra parts and there are no problems with a 24V Vgs limitation.
Maybe my electronics knowledge is not on par but could someone explain to me what I'm overlooking? |
|
|
Matro
Guest
|
|
| Posted: Thu Apr 10, 2008 8:54 am |
|
|
| ckielstra wrote: | Why is everybody ignoring Don's (ak6dn) post?
I think he made the smart suggestion that switching the low side of the pump is much easier than switching the high side. The BUZ11 will do the job nicely without adding extra parts and there are no problems with a 24V Vgs limitation.
Maybe my electronics knowledge is not on par but could someone explain to me what I'm overlooking? |
Because Don's suggestion is exactly the initial configuration that seems to be problematic (according to original poster complaining).
I first suggest to use a more appropriate MOSFET (even if BUZ11 should work) and keeping only one N-MOSFET.
But without answer of the original poster, we can't do better. Especially because we don't have any schematic of the system.
Matro. |
|
|
newguy
Joined: 24 Jun 2004
Posts: 1909
|
|
| Posted: Thu Apr 10, 2008 9:00 am |
|
|
| ckielstra wrote: | Why is everybody ignoring Don's (ak6dn) post?
I think he made the smart suggestion that switching the low side of the pump is much easier than switching the high side. The BUZ11 will do the job nicely without adding extra parts and there are no problems with a 24V Vgs limitation.
Maybe my electronics knowledge is not on par but could someone explain to me what I'm overlooking? |
It's not a bad suggestion, it's just that most all switching applications switch the hot lead, not the negative/ground. If the ground is switched, then the device being switched is always "hot." Getting a shock or accidentally grounding the device and having it unintentionally start now become issues.
Granted, for a 24V circuit, this isn't really an issue. But it's just not that common. For 120/240Vac circuits, this type of arrangement is contrary to the electrical code (in North America at least.)
There's also a nice high side switch IC that would work, the BSP772T: http://www.infineon.com/dgdl/BSP772T_20040127.pdf?folderId=db3a304412b407950112b4297bd64108&fileId=db3a304412b407950112b4297c414109 |
|
|
edhaslam
Joined: 15 Jul 2005
Posts: 89
Location: UK
|
|
| Posted: Thu Apr 10, 2008 10:20 am |
|
|
| ckielstra wrote: | Why is everybody ignoring Don's (ak6dn) post?
I think he made the smart suggestion that switching the low side of the pump is much easier than switching the high side. The BUZ11 will do the job nicely without adding extra parts and there are no problems with a 24V Vgs limitation.
|
This is infact how I intend to do it. It's simple and has a low parts count. I've just breadboarded it and it works fine. I'll try and post a schematic for you to take a look at as I'd like some suggestions on how to protect the rest of the circuit from voltage spikes etc.
I understand what you mean about the motor being 'hot' all the time, but hopefully it shouldn't be a problem in this application. |
|
|
Matro
Guest
|
|
| Posted: Thu Apr 10, 2008 10:35 am |
|
|
| edhaslam wrote: | | ckielstra wrote: | Why is everybody ignoring Don's (ak6dn) post?
I think he made the smart suggestion that switching the low side of the pump is much easier than switching the high side. The BUZ11 will do the job nicely without adding extra parts and there are no problems with a 24V Vgs limitation.
|
This is infact how I intend to do it. It's simple and has a low parts count. I've just breadboarded it and it works fine. I'll try and post a schematic for you to take a look at as I'd like some suggestions on how to protect the rest of the circuit from voltage spikes etc.
I understand what you mean about the motor being 'hot' all the time, but hopefully it shouldn't be a problem in this application. |
Motor being 'hot' means that when the circuit is open, the motor is connected to +24V and not to ground.
But it has nothing to see with the temperature. ;-)
The motor will stay cold if it is not powered.
Matro |
|
|
edhaslam
Joined: 15 Jul 2005
Posts: 89
Location: UK
|
|
| Posted: Thu Apr 10, 2008 10:42 am |
|
|
| Matro wrote: |
Motor being 'hot' means that when the circuit is open, the motor is connected to +24V and not to ground.
But it has nothing to see with the temperature. ;-)
The motor will stay cold if it is not powered.
|
No, no, I understand that it's nothing to do with temperature 
Here's the schematic anyway:
 |
|
|
Matro
Guest
|
|
| Posted: Thu Apr 10, 2008 10:45 am |
|
|
I would have add a 4.7nF or 10nF capacitor between gate and ground. ;-)
Matro. |
|
|
Humberto
Joined: 08 Sep 2003
Posts: 1215
Location: Buenos Aires, La Reina del Plata
|
|
| Posted: Thu Apr 10, 2008 1:03 pm |
|
|
I'm agree that switching the low side with an N-channel source connected to ground is
the most simple and easy solution, but.
ckielstra wrote:
| Quote: |
Maybe my electronics knowledge is not on par but could someone explain to me what I'm overlooking?
|
In this case, the only issue that would justify the use of a P-channel MOSFET switching the
hot side, should be the need to know & control the current flowing through the load.
Humberto |
|
|
newguy
Joined: 24 Jun 2004
Posts: 1909
|
|
| Posted: Thu Apr 10, 2008 1:13 pm |
|
|
| Humberto wrote: | I'm agree that switching the low side with an N-channel source connected to ground is
the most simple and easy solution, but.
|
I agree with that too.
| Humberto wrote: |
ckielstra wrote:
| Quote: |
Maybe my electronics knowledge is not on par but could someone explain to me what I'm overlooking?
|
In this case, the only issue that will justify the use of a P-channel MOSFET switching the
hot side, would be the need to know & control the current flowing through the loads.
|
I beg to differ. In an automotive application, employing hot side switching simplifies wiring for the end user. This is even more true when you consider that every automotive factory module and every user add-on device assumes hot side switching. To create a low side switched automotive device means that the end user now has to find a 12V power source and directly wire the load to that, then the other lead of the load goes to the switching unit. Then you aso have to worry about power off/failure mode - when the switching device itself is switched off, you don't want the 'load' to get turned on. It's just way simpler if the switching device has two leads, red & black, and red is energized when the device is turned on. This is particularly important considering that the vehicle is a frame ground system and the end user is probably accustomed to only connecting the red wire and grounding the black to the frame. At least for the automotive market. For the application that started this thread, low side switching is fine. |
|
|
ak6dn
Joined: 08 Jan 2006
Posts: 23
Location: Saratoga, CA USA
|
|
| Posted: Thu Apr 10, 2008 9:11 pm |
|
|
| newguy wrote: | | ckielstra wrote: | Why is everybody ignoring Don's (ak6dn) post?
I think he made the smart suggestion that switching the low side of the pump is much easier than switching the high side. The BUZ11 will do the job nicely without adding extra parts and there are no problems with a 24V Vgs limitation.
Maybe my electronics knowledge is not on par but could someone explain to me what I'm overlooking? |
It's not a bad suggestion, it's just that most all switching applications switch the hot lead, not the negative/ground. If the ground is switched, then the device being switched is always "hot." Getting a shock or accidentally grounding the device and having it unintentionally start now become issues.
Granted, for a 24V circuit, this isn't really an issue. But it's just not that common. For 120/240Vac circuits, this type of arrangement is contrary to the electrical code (in North America at least.)
There's also a nice high side switch IC that would work, the BSP772T: http://www.infineon.com/dgdl/BSP772T_20040127.pdf?folderId=db3a304412b407950112b4297bd64108&fileId=db3a304412b407950112b4297c414109 |
The assumption behind my original post was that that edhaslam was trying to do high-side switching with the NMOS FET. The failure symptoms he described suggested this. To make high-side switching work one would have to pump the gate up to 24V+Vgs(th) or about 30V to turn the FET on. Maintaining the gate within the allowable Vgs range then becomes very tricky. Of course low side switching works very readily, but the end application needs to be able to support this topology.
Switching ANY mains line AC requires an isolated approach, using a rated mechanical relay or an opto-coupled TRIAC. Straight DC coupling from one side of the line (even the neutral) to the PIC controller is just asking for a real problem beyond malfunctioning hardware (ie, a fatal shock to the operator). Do NOT try this at home.
Don |
|
|
edhaslam
Joined: 15 Jul 2005
Posts: 89
Location: UK
|
|
| Posted: Fri Apr 11, 2008 1:57 am |
|
|
Thanks again for your comments guys.
| Matro wrote: | I would have add a 4.7nF or 10nF capacitor between gate and ground. ;-)
|
Thanks, I'll add a cap in between the gate and ground. Anything else to prevent fatal damage to the PIC? Is it worth having a varister as well? |
|
|
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
