0
\$\begingroup\$

I have a circuit in which the C2001 (aka 2SC2001) transistor failed after many years. I Googled about, and tried to replace it with a BC337. The BC337 got kind of warm, and didn't work. B, C, and E pins were all in the correct PCB holes, I checked and double-checked, and triple-checked. The BC337 tested OK with my hFE meter.

So, is there something 'special' about the C2001 trannies? I have seen them referred to as 'switching' transistors somewhere, I don't know what that specially means. This is being used in a flyback circuit, which switches at approx ummm 140 kHz or so.

Thanks, any help appreciated!

My previous, sort-of related question is here: What type of voltage booster circuit is this?

\$\endgroup\$
3
  • \$\begingroup\$ I have short-term access to another C2001 - putting that in the circuit works just fine, so I know the transistor is the issue. That 'spare' C2001 belongs in another PCB, I do need a spare, hence keen to find proper equivalent. Thanks! \$\endgroup\$ – Chalky Sep 19 '17 at 9:09
  • 1
    \$\begingroup\$ How do you know that your transistor is the only problem in your circuit? Is it possible that there is another problem which caused your transistor to fail and which makes your new transistor hot? \$\endgroup\$ – Chupacabras Sep 19 '17 at 13:32
  • \$\begingroup\$ Hi there @Chupacabras, when I replace it with a good C2001 the whole circuit works just fine. When I replace it with a BC337, the circuit doesn't work at all. \$\endgroup\$ – Chalky Sep 19 '17 at 23:10
2
\$\begingroup\$

What could cause transistor to overheat? Well, it's whether excessive base current, or excessive product of Ic x Vce.
This is your circuit:
enter image description here

Base current is not a problem, it is limited by R128 to 4.5mA. So, transistor could overheat by high Vce or Ic.
Say hFE=200. Then maximum Ic=900mA. But it will be lower, say it is 500mA. Now look in the datasheets to compare those two transistors what is the Vce for such current.
This is from 2SC2001 datasheet:
enter image description here

Vce=0.3V
So power dissipation would be 150mW, well within specification.

This is from BC337 datasheet:
enter image description here

It is off the chart, it is a bad sign for this situation. Vce would be much higher than 1.0V, say it is somewhere around 1.5V. So power dissipation would be 750mW, which is above maximum allowed power dissipation.

I do not say your circuit is drawing 500mA, peaks could be higher or lower. But you can see that in some situations there are differences between BC337 and 2SC2001.

Another relevant difference could be in hFE. It depends on your batch. There are huge differences:
2SC2001 (hFE from 90 to 400)
enter image description here

BC337 (hFE from 100 to 630) enter image description here

So you could compare BC337 and 2SC2001 with very different hFE (you maybe compare BC337 with very low hFE, and 2SC2001 with higher hFE).

And the difference can be even bigger in hFE because hFE changes with Ic:
enter image description here

\$\endgroup\$
0
\$\begingroup\$

As far as I can tell from the datasheets, the 2SC2001 and BC337 are equivalent. The only difference I saw was the much higher f_t of the BC337. I know that some circuits depended implicitly on the f_t being low as a kind of low pass filter, though yours doesn't look like this. It could be that the higher f_t causes your circuit to oscillate somewhere in the 100-200MHz range. Seeing that your circuit has no feedback resistor anywhere on the 2SC2001, I would suggest to try adding a 10R resistor in the base of the 2SC2001 (resp BC337) after the 22R resistor that goes to the feedback path and see whether that improves the situation. If not, you have to take out a good oscilloscope and look at the waveforms.

Edit: Adding schematic to show what I mean:

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
6
  • \$\begingroup\$ Or maybe it could be due to differences on the base-emitter break down characteristics (not shown in the datasheet, except \$V_{BR(EB0}\$) of the two BJTs ? \$\endgroup\$ – next-hack Sep 19 '17 at 9:58
  • \$\begingroup\$ Hi, Attila, thanks so much for prompt reply. Can you clarify position of additional 10R resistor? Between the base of the 2SC2001/BC337 transistor and the junction of the existing 22R and 1k8? \$\endgroup\$ – Chalky Sep 19 '17 at 10:17
  • \$\begingroup\$ @next-hack: The switching transistor shouldn't go into breakdown. This would severely limit the lifetime of an already stressed component. \$\endgroup\$ – Attila Kinali Sep 19 '17 at 10:58
  • \$\begingroup\$ @Chalky: It would take too much space to explain the exact mechanism here, but the gist is, the transistor acts as an impedance conversion element at high frequencies (due to beta being frequency dependent). Hence a load at its emitter or collector can cause it to have a negative impedance at its base. If the base circuit presents an inductive source to the transistor this will form an oscillator somewhere in the VHF/UHF range. The additional resistor dampens this oscillation. \$\endgroup\$ – Attila Kinali Sep 19 '17 at 11:02
  • \$\begingroup\$ Thanks for that, but just want to clarify where do I put the resistor? \$\endgroup\$ – Chalky Sep 19 '17 at 12:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.