Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Biasing criteria for pHEMT switch

Status
Not open for further replies.

cnm

Full Member level 2
Joined
Oct 4, 2010
Messages
144
Helped
9
Reputation
18
Reaction score
9
Trophy points
1,298
Activity points
2,229
For HEMT FET switch, its terminal bias criteria never seems to be clearly stated in any standard text.

For example, is it okay to set the desired Vgs for switch ON and OFF, but leaves Vd open (or floating, or no bias)?

Most people tend to set the desired Vgs, and then leave both drain and source biased with equal potential (so Vds=0). Does it really matter if Vds is not equal to 0V set to any other voltage?
 

jayce3390

Full Member level 6
Joined
May 22, 2007
Messages
352
Helped
29
Reputation
58
Reaction score
21
Trophy points
1,298
Activity points
3,287
Set Vds = 0V and drive the gate with a pulse voltage around the Vgate treshold doesn't make sense.
 

cnm

Full Member level 2
Joined
Oct 4, 2010
Messages
144
Helped
9
Reputation
18
Reaction score
9
Trophy points
1,298
Activity points
2,229
It is a common way to use this device. Vgs is used to switch on/off the device.
 

jayce3390

Full Member level 6
Joined
May 22, 2007
Messages
352
Helped
29
Reputation
58
Reaction score
21
Trophy points
1,298
Activity points
3,287
Please, could you be more explicit...

For example, is it okay to set the desired Vgs for switch ON and OFF, but leaves Vd open (or floating, or no bias)?

And I wrote : Set Vds = 0V and drive the gate with a pulse voltage around the Vgate treshold doesn't make sense.
That means you need a voltage on Vd to switch something. If you have a look on some MOSFET datasheets, the switching time test circuit uses a large resistor (and "bias" voltage) on drain to pull the drain down.
 

cnm

Full Member level 2
Joined
Oct 4, 2010
Messages
144
Helped
9
Reputation
18
Reaction score
9
Trophy points
1,298
Activity points
2,229
MOSFET could be different. But pHEMT is the one we are talking about, bro.
 

mtwieg

Advanced Member level 5
Joined
Jan 20, 2011
Messages
3,645
Helped
1,281
Reputation
2,568
Reaction score
1,284
Trophy points
1,393
Activity points
27,507
All transistors need some output bias voltage and current to operate. Heterojunction devices aren't any different. Where did you hear otherwise?..
 

cnm

Full Member level 2
Joined
Oct 4, 2010
Messages
144
Helped
9
Reputation
18
Reaction score
9
Trophy points
1,298
Activity points
2,229
All transistors need some output bias voltage and current to operate. Heterojunction devices aren't any different. Where did you hear otherwise?..

Where did u hear what u just said above?
This is switch application, dude.
 

jayce3390

Full Member level 6
Joined
May 22, 2007
Messages
352
Helped
29
Reputation
58
Reaction score
21
Trophy points
1,298
Activity points
3,287
Where did u hear what u just said above?
This is switch application, dude.

Everybody knows that, you are not the only one working on pHEMT switch application. You should read some books explaining how transistors work. A pHEMT is a field effect transistor based like MOSFET. Are you a student?
 

BigBoss

Advanced Member level 5
Joined
Nov 17, 2001
Messages
5,183
Helped
1,537
Reputation
3,074
Reaction score
1,390
Trophy points
1,393
Location
Turkey
Activity points
31,158
For HEMT FET switch, its terminal bias criteria never seems to be clearly stated in any standard text.

For example, is it okay to set the desired Vgs for switch ON and OFF, but leaves Vd open (or floating, or no bias)?

Most people tend to set the desired Vgs, and then leave both drain and source biased with equal potential (so Vds=0). Does it really matter if Vds is not equal to 0V set to any other voltage?

Yes..
If you apply Vgs voltage either positive for E-Mode pHEMTs or negative for D-Mode pHEMTs, a depletion layer will be appeared just underneath of the gate and channel resistance will drop very sharply and the FET will act as a switch.Vds voltage is not necessary because FET became a simple resistance and deplation layer touches both Drain and Source.
 

cnm

Full Member level 2
Joined
Oct 4, 2010
Messages
144
Helped
9
Reputation
18
Reaction score
9
Trophy points
1,298
Activity points
2,229
Everybody knows that, you are not the only one working on pHEMT switch application. You should read some books explaining how transistors work. A pHEMT is a field effect transistor based like MOSFET. Are you a student?

Regardless of the answer, I always treat myself a student. :)

---------- Post added at 12:14 ---------- Previous post was at 12:11 ----------

Yes..
If you apply Vgs voltage either positive for E-Mode pHEMTs or negative for D-Mode pHEMTs, a depletion layer will be appeared just underneath of the gate and channel resistance will drop very sharply and the FET will act as a switch.Vds voltage is not necessary because FET became a simple resistance and deplation layer touches both Drain and Source.

Thank you, BigBoss. This is the exact answer I am looking for. I joked around long enough. Now some one finally loses the patience and cannot hold the answer.
 
Last edited:

mtwieg

Advanced Member level 5
Joined
Jan 20, 2011
Messages
3,645
Helped
1,281
Reputation
2,568
Reaction score
1,284
Trophy points
1,393
Activity points
27,507
Yes..
If you apply Vgs voltage either positive for E-Mode pHEMTs or negative for D-Mode pHEMTs, a depletion layer will be appeared just underneath of the gate and channel resistance will drop very sharply and the FET will act as a switch.Vds voltage is not necessary because FET became a simple resistance and deplation layer touches both Drain and Source.
Some finite amount of Vds is necessary, even if it is functioning as a switch and even if it is just a "resistance". If there is no bias voltage, then there is no current, and the device simply does nothing.

Or are you seriously claiming that I could short the drain and source together directly and it would still work somehow?
 

cnm

Full Member level 2
Joined
Oct 4, 2010
Messages
144
Helped
9
Reputation
18
Reaction score
9
Trophy points
1,298
Activity points
2,229
Some finite amount of Vds is necessary, even if it is functioning as a switch and even if it is just a "resistance". If there is no bias voltage, then there is no current, and the device simply does nothing.

For HEMT switch, no operation current is present. It is a voltage control device anyhow. This is a good thing- don't you like it? Having Vd=Vs does not mean they are RF shorted.

Case#1: one can bias up both drain and source terminals, or even set Vd=Vs (so Vds=0), In this case, all 3 terminals have DC bias voltages. But having both drain and source bias voltages is really not a requirement for switch operation.

Case#2, one can bias up only the gate and the source terminals, and leave the drain floating without any DC voltage.

Case #3: I suspect, one can even use the relative voltage difference between gate and drain to switch the device on/off, and leave source floating without any DC bias. In other words, drain and source are interchangeable when used as switch. Either drain or source needs to be biased. Biasing both drain and source is not required. But this needs to be confirmed.

Or are you seriously claiming that I could short the drain and source together directly and it would still work somehow?

BigBoss (and other experts), please comment on the above cases. Thank you all.
 
Last edited:

BigBoss

Advanced Member level 5
Joined
Nov 17, 2001
Messages
5,183
Helped
1,537
Reputation
3,074
Reaction score
1,390
Trophy points
1,393
Location
Turkey
Activity points
31,158
If a pHEMT will be used as switch, there will compulsorily be a potential between drain and source because of signal, RF signal etc. This potential is sufficient to create a Electrical Field between drain source.
 
Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top