What happens with bipolar Rbb as a function of collector current?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Member
Joined 2007
Paid Member
Hi,

I have a quick question that I hope you can help with:

It is "What happens to a bipolar transistors Rbb (base spreading resistance) as a function of collector current? Does it increase or decrease and how much?

I'd appreciate your insights ;-)

Best regards,

Jesper

P.S.: Just learned on the internet that the Rbb drops with increasing collector current. But would one of you know how much?
 
Last edited:
Hi,

I have a quick question that I hope you can help with:

It is "What happens to a bipolar transistors Rbb (base spreading resistance) as a function of collector current? Does it increase or decrease and how much?

I believe that Rbb is dominated by ohmic resistance and therefore should not be very sensitive to collector current at typical values.
It is an excellent question, where is your reference that it decreases with collector current?

Best wishes
David
 
From Analysis and Design of Analog Integrated Circuits by Gray and Meyer. "The value of rb varies significantly with collector current because of current crowding." Have to run, will try to add more later.

I don't have a copy of G & M handy to check their use of the symbols but Rb is perhaps not the same as Rbb, which is what you asked about in the first post.

Best wishes
David

Edit. It may be that they do use Rb for base spread R.
Google books shows only a short excerpt.
Edit2 Amazon has more. Typically 50% variation for 100X alteration of collector current.
Not very sensitive.
 
Last edited:
Hi

I think it just used as the change in base current for a given change in base emitter voltage or B*1/gm

Hope this helps
-Antonio

There is a significant difference between Rbb' and rb'e. They are entirely different quantities.

Yes;) The OP was about Rbb. Antonio's comment is about rb'e, I think.
I just wasn't sure what G & M meant by Rb. Apparently it is Rbb.
If so then Rbb is not very sensitive but varies a bit more than I expected.
I like questions where I learn:)

Best wishes
David
 
Yes;) The OP was about Rbb. Antonio's comment is about rb'e, I think.
I just wasn't sure what G & M meant by Rb. Apparently it is Rbb.
If so then Rbb is not very sensitive but varies a bit more than I expected.
I like questions where I learn:)

Yep I was wrong, just thought of it as the standard diode equation even though I saw the words (base spreading resistance).

One more dog-ear to my G & M

Thanks
-Antonio
 
Correct, to my knowledge it is modeling the (gold wire) connection from the chip to the pad.

The bond wire resistance is a fraction of an ohm. It is trivial compared to the ohmic resistance of the base silicon. Rbb varies from about 2R for an ultra-low Rbb (low noise) transistor like the 2SB737 to say 30R for a low noise transistor and G & M use 100R to 300R for nondescript transistors in examples.
I should have remembered this immediately. The fall in Rbb is modelled in SPICE. Rbb is set by parameter RB and the decline of Rbb to parameter RBM is controlled by the current parameter IRB.
Best wishes
David
 
Last edited:
Yes the base resistance is due to the conductivity of the base region. I presume that at high current levels the current is unevenly distributed (SOA hot spots developing) and the active parts of the base become isolated from the bond wire by now inert base material, raising the Rbb
 
Yes the base resistance is due to the conductivity of the base region. I presume that at high current levels the current is unevenly distributed (SOA hot spots developing) and the active parts of the base become isolated from the bond wire by now inert base material, raising the Rbb

The resistance decreases as current rises. Read the earlier posts.
It is a bit counter intuitive but the effect is discussed in G & M.
Basically the emitter current moves towards the edges of the emitter, this shortens the distance that the base current must flow.

Best wishes
David
 
Last edited:
The resistance decreases as current rises. Read the earlier posts.
It is a bit counter intuitive but the effect is discussed in G & M.
Basically the emitter current moves towards the edges of the emitter, this shortens the distance that the base current must flow.

Best wishes
David
Sorry, yes this does make sense, the current is being pushed to regions with thicker base and lower resistance per square (and for the same reasons lower hfe)
 
The resistance decreases as current rises. Basically the emitter current moves towards the edges of the emitter, this shortens the distance that the base current must flow.

Best wishes
David

Dave, Really well said, thanks
Seems to tie well with Bonsia's expectant correlation of beta decrease with increasing Ic, as both effects would be dependant on the current crowding.

Thanks All
-Antonio
 
Member
Joined 2007
Paid Member
Hi all,

& thanks for your replies :) It has been a very interesting read this morning to read through your posts and considerations about rbb.

@David: In your first post you ask me where I had found the information on the internet that rbb' decreases with increasing collector current. I can post this information again (it was a website related to testing the simulation algorithms of one of the major simulation softwares, I think) but given what you (all) have written here I doubt it is correct. In my memory the drops in rbb' were much larger than what have been discussed here. But let me know & I'll post a link.

So - just to sum up in a general way what you have written here:

rbb' does drop as a function of collector current (for various reasons), yet the drops are not expected to be e.g. 10* for a 10* increase in collector current but probably/more likely to be on the magnitude of some percent per doubling of the current (also depending on the level of the collector current)....?

Thanks again for replying - this is very informative ;-)

Jesper
 
So - just to sum up in a general way what you have written here:

rbb' does drop as a function of collector current (for various reasons), yet the drops are not expected to be e.g. 10* for a 10* increase in collector current but probably/more likely to be on the magnitude of some percent per doubling of the current (also depending on the level of the collector current)....?

Jesper

Yes, that's an excellent summary. The exact variation depends on the details of the transistor internals but the data sheet referenced by Samuel is probably typical. As you say it depends on the collector current, in particular at high currents the drop is a bit more, but simultaneously the Hfe also drops so it is not usually an area where you would run the transistor anyway.

Best wishes
David
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.