3904 (again)

A client came to me with a battery design that isn't meeting lifetime spec. The unit is a small transmitter to track people in hospital type environments.

The circuit runs off of a 3v battery. There is a 60k resistor from gpio output to the base of common emitter 3904. The collector is tied to one end of ferrite antenna. Other end is tied to 3v. 3904 is used between oscillator and the ferrite antenna.

My thumbnail analysis is that there is no control over the collector current outside of transistor beta. Battery lifetime is then dependent on beta

The best solution would be a part that has a narrow range of beta. I know that gain selected 3904 parts can be purchased, but they are sorted for minimum gain. I need a part that would be selected for a maximum gain.

I prefer not to open up the design to artwork changes, but am thinking a well bypassed emitter resistor may be the way to go.

Any thoughts?

Regards, Bob N9NEO

Reply to
Yzordderrex
Loading thread data ...

What's the purpose of the transistor? With a 60k base resistor, it can't be the modulator, surely?

Cheers

Phil Hobbs

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

The transistor is a buffer between the crytal oscillator and the ferrite rod antenna. Yes, base drive current is very low. I can't say much more as NDA in place.

Reply to
Yzordderrex

Lots of parts, like BCX70, are sold in fairly tight beta grades.

A bypassed emitter resistor might not control current any better. I'd have to see the circuit, which I can't, to make recommendations.

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

Is the 3904 collector load the ferrite rod antenna and is that tuned to resonance with a capacitor - if so then the load seen is quite high impedence? 60K means only a few tens microamp base drive so even with high beta seems hard to get much RF current into the antenna?

piglet

Reply to
piglet

c. The unit is a small transmitter to track people in hospital type environ ments.

output to the base of common emitter 3904. The collector is tied to one en d of ferrite antenna. Other end is tied to 3v. 3904 is used between oscil lator and the ferrite antenna.

ent outside of transistor beta. Battery lifetime is then dependent on beta

w that gain selected 3904 parts can be purchased, but they are sorted for m inimum gain. I need a part that would be selected for a maximum gain.

well bypassed emitter resistor may be the way to go.

Isn't the current controlled by the inductance of the antenna? Try adding a resistor in series with the loop and measuring the current. I t may not be limited by the current at all. Is there a capacitor anywhere in this loop? If there is no capacitor there will be a DC path that is ena bled half the time. So the current to the antenna is a combination of AC a nd DC with the DC being completely wasteful.

As piglet said, a parallel tuned circuit presents a very high impedance to AC current at resonance. It's still a very low impedance at DC and needs a larger DC blocking cap.

--
  Rick C. 

  - Get 1,000 miles of free Supercharging 
 Click to see the full signature
Reply to
Rick C

pec. The unit is a small transmitter to track people in hospital type envir onments.

o output to the base of common emitter 3904. The collector is tied to one end of ferrite antenna. Other end is tied to 3v. 3904 is used between osc illator and the ferrite antenna.

rrent outside of transistor beta. Battery lifetime is then dependent on be ta

now that gain selected 3904 parts can be purchased, but they are sorted for minimum gain. I need a part that would be selected for a maximum gain.

a well bypassed emitter resistor may be the way to go.

It may not be limited by the current at all. Is there a capacitor anywher e in this loop? If there is no capacitor there will be a DC path that is e nabled half the time. So the current to the antenna is a combination of AC and DC with the DC being completely wasteful.

o AC current at resonance. It's still a very low impedance at DC and needs a larger DC blocking cap.

Oh yeah, the DC current will be determined by the beta of the transistor if the DC resistance of the coil is low enough.

--
  Rick C. 

  + Get 1,000 miles of free Supercharging 
 Click to see the full signature
Reply to
Rick C

So what, is it keyed or something?

If it's class C/D, CW, with a modestly low duty cycle, it can be saturated and load current will be defined by ferrite antenna resistance instead. You're in a better place to tell if that's actually the case or not.

Can also swap it to a 2N7002 and a rather smaller gate resistor, but the higher gate charge would probably end up wasting more power.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Design 
 Click to see the full signature
Reply to
Tim Williams

If Vcc is 3V, 60k gives 40uA of base current, or 4-8mA-ish at the collector.

You can obviously reduce that range with a more complex bias scheme, but if you don't want to change the artwork, why not use an hfe-graded transistor as John suggested, and perhaps a higher-valued (120k?) bias resistor to suit?

It sounds cheesy but it's often done. Garage door transmitters made in huge volumes use components in selected batches. That way they can use cheesy bias schemes and still get good results, for example.

Cheers, James Arthur

Reply to
dagmargoodboat

Knowing nothing, you could try some similar npn's.

2n4401? George H.
Reply to
George Herold

Use a CLD (current limited diode), they are low scale devices but work nicely, that is if you can find a supply for the range you want. They Also have a threshold point so you need to shop a little.

In any case a device like this can be placed inseries. for that matter a feed back transistor could be used to clamp the input of the R.F. output transistor..

Reply to
M Philbrook

First idea: an emitter resistor bypassed with a capacitor.

Second idea: ALD makes MOS devices with adjustable thresholds; combine a emitter resistor with a grounded-gate NMOS with the drain at 3V, so that the (negative of the) cutoff voltage of the MOSFET is the sum of Vbe and Ic * Re. Then Re determines the mean collector current. Emitter-resistor parallel capacitance is still an option.

Reply to
whit3rd

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.