Bipolar transistor bias question

How exactly will that happen? R1 and the BE junction will set the DC voltage at the base. The capacitor can't influence that other than the AC signal that is passed. The DC set point will remain the same, no?

I don't think any of these circuits do what the OP wanted, or did he change that? This thread has gotten pretty long. He wanted something that would detect when the AC input got above 300 mV peak, no? The above schematic cuts off when the input goes negative and is on the rest of the time. Removing the diode and R2 just turn it into an AC coupled small signal amplifier.

--

Rick C

Yes, that worked well and all looked good. The base voltage varies from 360 to 430 to 534mV with temp changes and in all cases the transistor is off. That's what I wanted to know. Thanks.

C1 charges every time you use it to drive Q1(b). When the drive drops back to zero, the C1-Q1(b) node drops to (Vbe - drive voltage).

Cheers, James Arthur

Actually, I don't think that is correct - unless we are saying the same thing. I'm not sure the electrical definition of "use it".

The base node will be driven above and below the DC set point no matter the DC content of the input signal. The only effect of the capacitor is blocking the DC content of the input signal. So what ever the set point is (as determined by R1 and the BE characteristics), if the input signal is applied and the voltage on the capacitor is allowed to settle, the bias on the base is going to be determined by that set point. Q1 will

*not* be off most of the time. Just the opposite. It is on at the set point and will only be turned off by significantly negative portions of the AC component on the input.

--

Rick C

Problem is that a coupling capacitor will get charged by base current on the positive stroke, but then will discharge slowly via the bias, creating a threshold movement.

So a simple transistor stage can't maintain a constant threshold. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     |

Either you guys are missing soemthing or I am. This looks like a 0.3v threshold detector:

. _________+ . | | . R R . | _| . --R--+---(_) . | . | . ___________| 0v

Keep P_diss right down for Vbe stability.

NT

That's right, but AIUI he doesn't need a stable d.c. threshold.

I think it's for a variation on this:

But I'm guessing, & might be mistaken.

Cheers, James Arthur

Maybe I'm missing something, but suppose there are amplitudes higher than 300mV... is the threshold supposed to stay put? ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     |

The capacitor is charged rapidly when driving Q1, and discharges slowly, creating a d.c. offset across C1 that depends on the time constants.

If you don't understand it, LTSpice might clear things up.

Cheers, James Arthur

My main advantage in engineering is due to education and design BEFORE simulators. Most of the respondents to Bowden's post really have no "feel" for how circuits work... totally missing the forward conduction of the B-E junction. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     |

We're all left guessing here, but I think the answer is that Bill has a 300mV or greater a.c. signal, and wants that to drive a switch.

I forgot a resistor--R3--when drawing his original LTSpice circuit:

+2V +2V -+- -+- | | .-. R1 .-. RL | | 5.6k | | 2k '-' '-' | | | R2 |/ +--15k--+----+--| Q1 | | | |>. D1 V C1 --- .-. | --- 2uF --- | | === | | '-' R3 === | | 51k | === ^ Vin

If D1 were schottky, the result is a quasi-temperature stable ~.3V threshold, R3 not needed.

+2V +2V -+- -+- | | .-. R1 .-. RL | | 10k | | 2k '-' '-' | | | R2 |/ +--15k--+----| Q1 | | |>. D1 V C1 --- | ,-' 2uF --- === | | === | ^ Vin

Cheers, James Arthur

This makes assumptions about the waveform I believe. What happens if the input waveform is only positive 1% of the time?

--

Rick C

You guessed it. I started out with 5 transistors and then refined it to 3. So, I thought I might get it down to 2 transistors. The original design was in a magazine Nuts and Volts I think and only used 2 transistors hooked up as a SCS. The magnet induced a small voltage in the coil which triggerd the SCS which allowed a large capacitor (220uF or so) to discharge through the coil and push the magnet for 15 milliseconds or so. Problem was, the pulse wasn't square and drooped off as the capacitor ran down. And they used a low resistance coil of 30 ohms which wasn't very efficient. My latest model uses a coil with better dimensions and has a wire resistance of about 150 ohms which only draws 20 milliamps at a 10% duty cycle which is about 2 milliamps plus another milliamp bias current, or maybe 3 milliamps total. Using a couple 'D' alkaline cells, it should run for 16/.003 = 222 days. I was thinking if I wanted to use the SCS idea, I only need 2 transistors but the problem is biasing the thing so it triggers on 300mV at about a 10% duty cycle. 90% of the time it doesn't do anything. It only sees one sine wave cycle of about a 25 millisecond half cycle time.

.

So what's wrong with the basic 0.3V detector I posted?

NT

I know it's sexy to say 'only two transistors,' but I count a diode and a transistor the same these days. A part is a part. The PNP-thing I sketched up-thread might be worth a gander.

Idea: if your inductance were *much* larger, you could recycle the flyback energy & radically extend the battery life.

I don't know how practical that is. You might need an iron core to get the inductance, which would make the motor cog, but I'm not sure that's a killer.

Dunno.

Or maybe a partial iron magnetic circuit, for better magnetic coupling, but keep it far away from the neodymium?

I haven't thought about it much--I'm pretty swamped, pulling all-nighters.

It's a cool project--really fun to watch it run. You'll definitely inspire some kids.

Nice website, too. Thanks for all that work you put in it.

Cheers, James Arthur

LTSpice is your friend.

Cheers, James Arthur

I'm guessing the circular component was a BJT, right?

(redrawn) . _________+ . | | . R2 R3 | | . | |/ . --R1-+---| Q1 . |>. . | . ___________| 0v

Looks fine to me.

Cheers, James Arthur

The base voltage should remain constant at about 400mV as the supply voltage falls from 3.2 to 2 volts. That's why I used a diode.

LTspice thinks a 2N3055 will accept a couple of milliamps into the collector with 300mV on the base, but my experiment with a real part suggests that 400mV is actually needed, but my part is from "CDIL" and their model is from OnSemi

--
This email has not been checked by half-arsed antivirus software