Circuit behavior

[Homebrew]

I built the circuit as shown and when i use a 6v 60mA bulb it lights to full brightness instantly, when i change the bulb to a 6v 160mA it glows faintly and takes almost 30 seconds to reach full brightness....can someone shine some light on this for me please (pun intended) lol

 

[sagor1]

You are probably drawing too much current from a 9V battery. At 60ma, it can supply enough current to light the bulb. At 160ma, you are maybe drawing too much current and the internal resistance of the battery is causing a voltage drop. As the battery overheats from excessive current, the internal resistance may drop.
Measure the voltage at the bulb when you try this, see what it shows. Also, measure the battery voltage with both bulbs, see if there is a difference.
You don't specify what kind of 9V battery you are using, I'm assuming the standard alkaline 9V type...
Also, having a transistor handling the current draw, and with a sagging voltage on the base, the transistor gain may limit the current as a start, then slowly increase if the battery voltage slowly increases.

 

[Homebrew]

I am running the circuit from my bench power supply set at 9v, i measure the voltage across the bulb and with the 60mA bulb the voltage is full instantly, with the 160mA bulb the voltage creeps up over several seconds.

 

[ZipZapOuch]

The base current is only 8.7v/10000ohms = 0.8mA

The gain is not huge on this part ( from my search, it's an old germanium PnP) and your collector current is about 20mA or less.

Change the base resistor to 350 ohms to goose the base current to about 2.5mA.

Here's the datasheet I found.

AC128 Datasheet | PDF

AC128

www.scribd.com

 

[Homebrew]

Thankyou so much ZipZapOuch.....i dropped the value of the base resistor to 5K and it is almost instant response, now i just need to study the circuit more and play with some other values to understand the circuit, i am new to Electronics and i like to know how things work.....thanks aagain

 

[ZipZapOuch]

Also, since you're using a 9v supply and your bulbs are rated for 6v, you should add a current limiting resistor in series with the bulb that dissipates 3v @ 160mA (use ohms law).

For the 160mA @ 6v bulb, you want to "drop" 3v at 0.160A, you need a 3A/0.160ohm = 18 ohm resistor (or 22 ohm).

For the 60mA bulb, use a 3v/0.060A = 50 ohms (or 47 ohm as a standard value, or two 100 ohm resistors in parallel)

 

[Tony Stewart]

I built the circuit as shown and when i use a 6v 60mA bulb it lights to full brightness instantly, when i change the bulb to a 6v 160mA it glows faintly and takes almost 30 seconds to reach full brightness....can someone shine some light on this for me please (pun intended) lol

I am impressed that you actually have a vintage Ge transistor. Zip Zap is correct that your base resistance is too high relative to the load resistance. That ratio ought to be the current gain times the load R = Rb, if we ignore Vbe drop for a bulb because the bulb resistance will rise as much as 12x the cold resistance at full brightness.

So why did it take so long? Your transistor junction was getting as hot as the bulb initially with only 4V across the bulb and Bipolar transistors love heat ( unlike FETs ) and this increases the mobility and current gain, in your case perhaps as high as 350 when hot. Unfortunately, like all things, aging takes a toll on hot parts and reduces their normal life expectancy 50% roughly every 10'C rise due to the Arrhenius Effect in Chemistry. So the current gain was increasing after the transistor got hotter, then the Vce voltage drop would reduce, and the lamp voltage would rise.

 

[Diver300]

In addition to what others have said, the bulb will have a much lower resistance when cold. I would expect a bulb like that to initially take around 5 times the normal current if connected directly to a 6 V supply. The transistor is limiting the current and so it is taking longer to start up.

 

[Nigel Goodwin]

Thankyou so much ZipZapOuch.....i dropped the value of the base resistor to 5K and it is almost instant response, now i just need to study the circuit more and play with some other values to understand the circuit, i am new to Electronics and i like to know how things work.....thanks aagain

If you want to learn things I would suggest using a MUCH more modern transistor, instead of one that's very low specification, probably 60+ years old, and been obsolete for many decades - germanium devices were also easily damaged, and had a tendency to be unreliable. Assuming you haven't damaged the transistor by the abuse you've been giving it?, then you could easily sell it as they are popular with people making antique guitar pedals.

Pretty well any small general purpose silicon transistor would do, my 'go to' device was always the BC107, but that's pretty well an antique now (but the spec and reliability is up there with current devices) - something like the BC327 or BC337 would be good.

I would also suggest using NPN transistors rather than PNP, as the entire electronics industry moved to NPN (and -ve chassis) back in the 60's, and most examples will use NPN. It makes little real difference, but it means you've got to reverse almost every example you try and follow.

 

[danadak]

PNP's still enjoy significant usage, many ICs as well, like LDO regulators, use them
extensively.

From ChatGPT :

an esitmate of worldwide shipments npn versus pnp transistors

ChatGPT said:​

There are no precise public estimates breaking down worldwide shipments of NPN vs PNP transistors individually, as most industry reports categorize them broadly under discrete bipolar junction transistors (BJTs) or even lump them into broader "discrete semiconductors."
However, based on industry usage and design preferences, we can provide a reasonable estimate and rationale:

Engineering & Market Context:​

• NPN transistors are far more commonly used than PNP in most applications.
• The vast majority of digital and analog circuit designsfavor NPN due to:
  • Better electron mobility (faster switching and higher efficiency).
  • Simpler interfacing with common ground-referenced logic systems.

Shipment Estimate (Based on Usage Trends):​

From various industry sources and anecdotal evidence (e.g., component distributors, design practices), the ratio is roughly:
NPN: ~80–90%
PNP: ~10–20%
This aligns with:

• Common discrete transistor types sold (e.g., 2N2222, BC547 = NPN).
• Power transistor arrays (e.g., for motor drivers) that mostly use NPN.
• Typical complementary pair designs where NPN is dominant, and PNP is only used where necessary (e.g., push-pull stages).

If We Estimate Volume:​

The global market for discrete transistors (excluding power modules) is tens of billions of units annually.
Example:

• If 50 billion BJTs are shipped yearly, and 85% are NPN:
  • ~42.5 billion NPN
  • ~7.5 billion PNP

This is a rough estimate and would vary by industry segment (e.g., automotive, consumer electronics, industrial).

 

[ZipZapOuch]

PNP's still enjoy significant usage, many ICs as well, like LDO regulators, use them
extensively.

From ChatGPT :

Very true. Any high-side switch application or high-side current regulating - PNPs have solved many problems for me. P-channel MOSFET for high-some switching but I prefer PNP for current regulation