BJT (NPN) Pull-Down

General Electronics
1

@quozl / @ToyBuilder I’m at it again :smiley:

This is a follow on from my last post MOSFET Pull-Down but this time on BJTs (NPN)

I was looking at the schematic for the IoT 6 Click from Mikroe (https://download.mikroe.com/documents/add-on-boards/click/lte_Iot_6/lte-iot-6-click-schematic-v100.pdf)* specifically this section

image

looking at the BJT attached to the PWR label, my understanding is that the purpose here is to be able to use the AN pin on the header to pull PWR_ON to ground and there by powering on/off the module (the same effect as the pushbutton would have). Similarly the BJT on the RESET pin allows the module to be reset.

As is my way, I started to think about the role R1/R2 is having - especially R2 (R1 being to limit the base current of course). Digging my Art of Electronics out, page 63 as the Transistor switch

image

neglecting the pull-down (and indeed this is the most common variant I have seen and used). Googling turns out some useful material - this StackExchange post was a good read: Why pull base of BJT switch? - Electrical Engineering Stack Exchange

image

image
image719×256 43.7 KB

(thankfully we’re all using R2 designator!) - the main explanation is all fine and aligns with the common usage of an NPN as a switch. Also ensuring the base is low rather than floating. the last statement I reckon I’m ok with - if you have leakage current from collector to base, you have a path for it, and as long as the voltage across R2 is lower than the Vbe threshold, the transistor will remain off.

so far so good! But I found the following video - https://www.youtube.com/watch?v=6sXCqDel6oA (@ToyBuilder, I watched it at 0.25x first this time!); looking at this screen grab has got me scratching my head somewhat and wondering if I missed something or this might be a mistake…

image

“pull down resistor prevents stray signals” - experimentally he touches the base resistor (with no pulldown) coupling external AC into the base and switching on the transistor; he adds the pull-down and the current into the base is reduced and the transistor remains off (companion video - https://www.youtube.com/watch?v=HKnGkQJR-1Y).

What I see is that the 10K Pull down is before the base resistor, which I can’t see as having any useful effect. Any stray EMF induced (“stray signals”) would still induce a current into the base of the transistor right? and leakage current from the collector is more likely to be injected back into the base. If the pull down was connected to the transistor base (as in all the other diagrams above) then I wouldn’t have an issue, but its not… have I missed something?

* small aside - am I the only one who finds this style of schematic annoying to read? :wink:

2

Have stumbled on this pulldown - Which configuration is better for pulling down an NPN transistor's base? - Electrical Engineering Stack Exchange post on StackExchange which is a pretty good discussion

i need to give it a couple of reads I think when I have a bit more peace and quiet in the house! :smiley:

3

Conceptually, it’s the same problem as you’ve already observed with the FET. Sufficient charge on the base/gate will allow the transistor to turn on.

For this situation, a first-order look at the behavior is enough. While the BJT will self-discharge the base, the base current will be ~ 100x smaller than the collector current. If your BJT is driving, say, a 10K ohm impedance load, you base is effectively a pull-down on the order of 1Meg ohm. A 10K ohm pull-down (or 20K ohm to ground as seen from base) is going to be two magnitudes faster. In effect, the base is practically shorted to ground as long as there is not an actual short to the +V rail.

4

@ToyBuilder yep, agree :slight_smile: - I grabbed my noise cancelling headphones and read through the link i added in the second post (pulldown - Which configuration is better for pulling down an NPN transistor's base? - Electrical Engineering Stack Exchange) and I can agree with everything in that post.

what is not covered in either case is the notion of external EMF coupled in - as per this video: Pull down resistor for NPN bipolar junction transistor switch circuit using electronics 2N3904 BJT - YouTube and in that case I think i would prefer to have a pull down closer to the base… but actually it probably wouldn’t make any difference in reality - the differences would actually be minimal

this was another rubber ducky I think; Just needed the process of explaining to align my thoughts… if i keep this up CE will need a couch for me to lie on whilst i rubber ducky… :wink: