The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> low voltage bjt operation
https://designers-guide.org/forum/YaBB.pl?num=1439695358

Message started by Mohit_Kumar on Aug 15^th, 2015, 8:22pm

Title: low voltage bjt operation
Post by Mohit_Kumar on Aug 15^th, 2015, 8:22pm

Is there any bjt design working below 500 mV. If not why?
I was wondering because when they operate Mosfet in sub-threshold it is supposed to behave like a bjt with capacitive divider at the input. Is there any disadvantage of using bjt (excluding input impedance) at lower voltages?

Mohit

Title: Re: low voltage bjt operation
Post by RobG on Aug 16^th, 2015, 7:26am

The emitter-base voltage is too large: on the order of 600mV at room temperature and 800mV at -40C biased at 1uA (the numbers are very rough and depend on size f transistor). You can get some improvement as you lower the current, but each time you half the current you only get a Veb reduction of 17mV at 30C, 13mV at -40C. You'd have to decrease current by 7 orders of magnitude (0.1pA) to get 500mV at -40C, and that wouldn't even give you enough headroom to bias the thing.

Also, at low currents you get low level injection which lowers the current gain if the transistor (beta or Hfe). This may not matter if you are just using it as a diode.

You can use Schotkey diodes for less base-emitter, but there is some issue or they would be used all the time.

Title: Re: low voltage bjt operation
Post by loose-electron on Aug 18^th, 2015, 12:59am

RobG wrote on Aug 16^th, 2015, 7:26am:

Agreed Vbe does not get adjusted like Vth does. The PN junction on silicon is pretty fixed in characteristics.

Title: Re: low voltage bjt operation
Post by AnilReddy on Sep 12^th, 2015, 11:07pm

Hi Mohit,

What is the intent of trying to use a bjt at low Vbe levels?

Bye
Anil

Title: Re: low voltage bjt operation
Post by Mos on Sep 13^th, 2015, 4:53am

Anil,
Whenever MOS is used for sub-threshold it is mentioned that it behaves as BJT. So I was curious that why BJT is not used in first place, is it only due to its low impedance.
Regards,
Mohit

Title: Re: low voltage bjt operation
Post by AnilReddy on Sep 15^th, 2015, 10:38pm

Hi,

May be i should rephrase my question a bit. If the intent of using an MOS device in sub-vt mode is to make it work like a bjt, then you already have a bjt.

or, let's say you can directly use a bjt device. Then, if you have a bjt then why again operate it with low vbe?

Bye.

Title: Re: low voltage bjt operation
Post by boe on Sep 16^th, 2015, 8:04am

AniReddy,
a sub-vt MOS behaves like a BJT, i.e. has exponential relationship between vgs and id. Depending on your process, the Vbe equivalent (Vgs) can be significantly lower with a MOS.
Also, "real" BJTs (with reasonable beta) often require additional masks (and therefore costs).
- B O E

Title: Re: low voltage bjt operation
Post by Mos on Sep 22^nd, 2015, 7:03pm

Anil,
I was wondering about the effectiveness of BJT for low-power operation (equivalent to sub-threshold of MOSFET). BJT has more gm for the same current. This was the reason why I asked this question.
Regards

Title: Re: low voltage bjt operation
Post by DanielLam on Oct 1^st, 2015, 11:42pm

BJTs require a Vbe voltage to turn on. The Vbe is usually 600-800 mV. So you can't use BJTs for low voltage operation. I suppose you can look for a low Vbe BJT, but I doubt it will be lower than the Vt of a MOS.

BJTs are used for high performance applications.

Sub-threshold MOSFETs are used for low power and typically low speed applications.