How is RF BJT bias current value calculated

Could some electronics guru please shed some light on this ? For a non-RF BJT amplifier a simple design rule is to set the maximum base or biasing current to 1/10 of the calculated collector current. Does tha same rule apply to RF BJTs as well ? Thanks in advance.

Reply to
dakupoto
Loading thread data ...

Most linear common emitter amp stages would never work if you did that.

NT

Reply to
tabbypurr

The OP appears to be talking about the technician-school rule of thumb for a common-emitter amp with emitter degeneration and voltage-divider base bias.

If beta is at least 100, pushing I_C/10 through the voltage divider will keep the bias current constant to within about 10%.

RF transistors often have low betas, so it's generally better to use another approach.

Cheers

Phil Hobbs

Reply to
pcdhobbs

It seems the OP confused bias current with resistor current. So it might be worth pointing out there are other ways to do the job too.

NT

Reply to
tabbypurr

May I draw your attention to the simple design example at:

formatting link

I understand that the collector resistor value can be computed from the value of the maximum collector current(obtained from BJT datasheet), but how is the resistor value on the BJT base bias leg of the circuit calculated ? Is it from the knowledge of Ibe ? Also, I am fully aware of the vaious NJT base biasing schemes, both simple ones and temperature compensated ones. Please shed some light on this.

Reply to
dakupoto

May I draw your attention to the simple design example at:

formatting link

I understand that the collector resistor value can be computed from the value of the maximum collector current(obtained from BJT datasheet), but how is the resistor value on the BJT base bias leg of the circuit calculated ? Is it from the knowledge of Ibe ? Also, I am fully aware of the vaious BJT base biasing schemes, both simple ones and temperature compensated ones. Please shed some light on this.

Reply to
dakupoto

Depending on the particular device and frequency RF small signal amp transistors often need to have their emitters tied directly to the ground plane (or nearly so through a small DC impedance) so a simple voltage divider won't work. A resistive divider + the BJT input impedance also likely won't be well-matched to the output impedance of whatever the signal source is and transducer gain will be poor.

Don't know how OP defines "RF" though; with thought one could probably build a discrete transistor amp not much different than you would an audio amp and with the proper design it would work OK well into the AM band, perhaps out to 10s of MHz. Above that base biasing techniques like constant current bias or voltage feedback from the collector would be more appropriate than resistors

Reply to
bitrex

can be computed from

You've got that back to front. You pick the collector current using the dat asheet curves for power output, gain, and distortion vs. I_C, V_CE, and fre quency. Then you check to see that you aren't going to cook the transistor.

bias leg of the circuit

simple ones and

Well, the voltage divider equation is pretty simple. You put the base at so me low voltage like 1.5 or 2V, enough so that the emitter degeneration resi stor can stabilize the bias over beta and temperature but nor so much that the output swing is reduced much. Then you pick I_C from the distortion/gai n curves, which gives you R_E, and then you centre the output voltage. Last , you pick the impedance level of the base biasing scheme so that the outpu t voltage is OK over beta.

Of course if you want impedance matching, there are a few more steps.

Cheers

Phil Hobbs

Reply to
pcdhobbs

That used to be true back in the days of Allen Bradley 1/2 W resistors and dogbone capacitors, but not any more. A few 0402 caps in parallel will get you inductances way under a nanohenry. I've measured laser drivers whose to tal loop inductance (laser, GaN FET, 0402 cap) is under 400 pH.

Cheers

Phil Hobbs

Reply to
pcdhobbs

Wait, so it would work, or it wouldn't?

The best scope I have has a bandwidth of 60 MHz so my boots-on-the-ground experience (outside a sim) with the really fast stuff is pretty limited

Reply to
bitrex

and dogbone capacitors, but not any more. A few 0402 caps in parallel will get you inductances way under a nanohenry. I've measured laser drivers whos e total loop inductance (laser, GaN FET, 0402 cap) is under 400 pH.

Luxury! My first scope was 1MHz. I used it to measure frequency response be haviour of a 6MHz circuit.

NT

Reply to
tabbypurr

Sure it'll work. You can do that on a dead-bug proto out to at least 1GHz.

Understood. I use 40-GHz SiGe:C transistors as cascode stages, which gets r elatively entertaining at higher collector currents.

Back before I had 10-50 GHz test gear, I measured one of those oscillating at 12 GHz using a manual wave meter--the offset voltage was sensitive to ha nd capacity, and it went through a whole cycle when I moved my hand by half an inch. 1 inch wavelength -> 12 GHz.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs  
Principal Consultant  
 Click to see the full signature
Reply to
pcdhobbs

Oops, I misspoke, I actually have a 100MHz Rigol digital one, still sitting in a box. Haven't really had a chance to sit down and take the time to learn it yet.

I have two analog scopes, a Kik 60MHz and a Tek 50MHz. They probably cost thousands in 1982; guess I was fortunate to come of age in a time when people were basically giving 'em away.

Reply to
bitrex

Voltage gain is easy to come by I guess, but if large transducer gains is what OP needs then he'll have to do something a little different.

Reply to
bitrex

Designing things is fun, but it's hard to beat the performance and simplicity of a self-biasing 3-terminal MMIC.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

,

atasheet curves for power output, gain, and distortion vs. I_C, V_CE, and f requency. Then you check to see that you aren't going to cook the transisto r.

some low voltage like 1.5 or 2V, enough so that the emitter degeneration re sistor can stabilize the bias over beta and temperature but nor so much tha t the output swing is reduced much. Then you pick I_C from the distortion/g ain curves, which gives you R_E, and then you centre the output voltage. La st, you pick the impedance level of the base biasing scheme so that the out put voltage is OK over beta.

Thank you very much for pointing out the details of the calculation steps i nvolved.

Reply to
dakupoto

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.