2N3055 resistance question / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

2N3055 resistance question

previous 1 2 3 next

Move Thread

LAN_403

radiotech

Sat May 05 2012, 10:58PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

With a junction transistor, there is a intrinsic resistance in the emitter, that is calculated
as 0.028 Volts per Amp of current in the emitter terminal.

So if the current was 1 amp Re would be 28 milliohms in series with RE,
RE, being the external emitter resistor to ground.

In studying Klugesmiths reply, and recalling some things, the constant Re, always in series
with emitter, is a voltage drop you cannot measure, but depends on what current exists
in your circuit at saturation.

A good question is, why is it a universal constant of bipolar transistors.

A very bright fellow told me it involves Planck's constant. Over to the physicist crowd.

Mattski

Sun May 06 2012, 05:58PM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

Ash, that's right although if you're just trying to characterize the BJT I'd put the resistor on the collector instead of the emitter, as your output current will otherwise put some negative feedback into your circuit: as the emitter voltage rises it decreases the base-emitter bias and reduces output current. The simple way to describe bias bias current of a BJT is ~0.7V Vbe drop, so your emitter current is Ve=Vb-Vbe, Ie=Ve/RE=(Vb-Vbe)/RE which is limited by RE. And since you have some resistance in series with the base too you also need to account for that.

Radiotech, I think you're thinking of the small-signal emitter input resistance r_e which is often used in AC modeling where the voltage swings are small compared to the bias point. It results from the fact that the small-signal transconductance (gm, where Ie=gm*Vbe) of a BJT is equal to Ie/(kT/q) where kT/q is equal to about 26mV at 300K temperature.

If the base of a BJT is held at a constant voltage then as the emitter voltage rises by deltaV then the current leaving the emitter changes from (Vb-Ve)*gm to (Vb-Ve-deltaV)*gm. So the current change deltaI was deltaV*gm. The effective input resistance is deltaV/deltaI=1/gm=(kT/q)/Ie= 26 milli-ohms per inverse amp of drain current.

This is not an actual resistor and it does not generate a voltage drop inside the BJT, rather it describes how the emitter current responds to small voltage fluctuations at the emitter because these fluctuations reduce the Vbe bias and thus reduce current. Depending on how the transistor is used in a circuit this term may or may not apply.

radiotech

Sun May 06 2012, 08:08PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

Thank you Mattski, that was exactly where we used the RE and Re. I have copied your
explanation, and saved it on paper and put it with the other papers from then.

Ash Small

Mon May 07 2012, 12:26PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I picked this up in a pack of assorted power resistors from Maplins yesterday.

Datasheet here:

]eng_ds_1773035_c.pdf[/file]

It's 150 Ohm, 50 Watt (20 Watts without heatsink), 5%, wirewound, and is standard as opposed to low inductance.

Is this suitable for this application at 25kHz?

Mattski

Mon May 07 2012, 08:25PM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

I reckon it will work because 25kHz is not terribly high, so even if we guess 100uH inductance which is pretty big then the reactance is reasonably small compared to the resistance. I have no idea what a good guess is for the inductance of that resistor though.

If you want to be sure then measure the inductance and make sure the reactance (2*pi*freq*L) is much smaller the resistance, or you can do a frequency sweep to see if it behaves differently at 25kHz than at, say, 1kHz.

klugesmith

Mon May 07 2012, 11:38PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Not sure about 150 ohm wirewound power resistors.
But I once measured an assortment with R values around 1 ohm.
For most, the corner frequency associated with L/R time constant was roughly 1 MHz.

Ash Small

Tue May 08 2012, 12:34AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I usually measure inductance by connecting the inductor in parallel with a capacitor of known value to form a tank circuit, then connect this to a sig. gen. via a series resistor. I usually use a resistor around 60 Ohms. I then measure the amplitude of the waveform in the tank with a 'scope, while doing a frequency sweep. The peak amplitude occurs at the resonant frequency of the tank circuit, and I can then determine the inductance.

If the inductor I'm measuring has a resistance of 150 Ohms, will I need a much higher value series resistor between the tank circuit, as this is currently only around a third the resistance of the inductor in the tank circuit that I'm measuring?

I'm planning on using a small (3/4 inch long) wirewound resistor as the series resistor, as I assume the inductance of this will be negligible.

(I can provide a schematic if required)

radiotech

Tue May 08 2012, 01:16AM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

The capacity of the resistor itself to the metal heatsink may be an issue if it couples
the very high frequency components of the 25 kHz switching frequency to the case or
other metalwork enclosing your projects. Might cause unexpected "snivets" or glitches.

Ash Small

Tue May 08 2012, 07:16AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I did suspect I had some issues with stray capacitance/inductance when I was last playing with this circuit.

The heatsink for the 2N3055 is huge, and there are fairly long wires connecting everything. Also, the two cooling fans for the heatsink are powered by the same 12V supply I'm using for the rest of the circuit, so are in effect in parallel with the transistor and flyback primary that I was using.

I'll connect it all up later today and see what happens.

Ash Small

Tue May 08 2012, 08:46PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Well, I tried everything I could think of to measure inductance, but apart from a 'glitch' at 660kHz I didn't get anywhere.

I then connected the transistor collector to the resistor, connected the base to the sig. gen., and connected the resistor and transistor emitter to a 12V car battery.

sig. gen. output was +-1V and I obtained the following:

Do I need to use, say, a 1 Ohm resistor, which will will allow 12 Amps to flow, and then a 0.5 Ohm resistor, which will allow 24 Amps to flow in order to find out if the transistor is saturated, or does the fact that the voltage across the resistor is 12V as near as makes no difference tell me that the transistor is saturated?

'Scope was set to 2 V per division. The distortion of the waveform (ringing, etc.) isn't as bad as when I was driving a flyback with the same setup. I'd be very surprised if I don't need to amplify the sig. gen. output in order to drive the 2N3055 into saturation.

previous 1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob