[Lowfer] Lowfer Final Transistors-Switching Time

[LEE BAHR]

I looked everywhere for 2N6601 and 2N6651 transistors.  Nowhere to be found. 
These seem more unobtainable then MPS6602 and MPS6652.  However, you can 
easily buy zillions of 2N4401 and 2N4403 transistors  for plus and minus 4 
cents each.  I just ordered 200 each from Mouser plus a few NTE128 and 
NTE129.  I also ordered a few Zetex ZTX653 and ZTX753.  I heard they worked 
good, but getting educated on here regarding turn on and turn off times plus 
saturation figures don't make the Zetex ones look good on paper.  However, 
we will see!  I bought a few of the Zetex anyway.  I'm trying to source some 
MPS 6602 and MPS 6652 now.  I might get lucky but have not been given a 
price for them to date.

Bill, where can I find your pp output circuit you mentioned today?  I don't 
have LOWDOWN that far back.

Lee, w0vt
pulsarxp at embarqmail.com





>
>
>
>
> Hi All,
>
> I spent a lot of effort to determine the best complementary output pair 
> when I was experimenting with TX loop antennas for my loop articles that 
> appeared in the LowDown some 7 years (?) ago. The main requirement at that 
> time was that efficiency at a 1W DC input had to be as high as possible. 
> Normally a few percent loss in power would not result in much loss in 
> transmitted distance covered but I didn't want to add another variable in 
> any of the very close loop comparisons I was making. I looked at the 
> 2N2222/2907, the 2N4401/4403, the 2N6601/2N6651 and a number of others - 
> in fact any  switching transistors that I could get my hands on. Turned 
> out the 2N6601/6651 was a cut above the rest in the output configuration 
> that appears in my second article. This circuit contains two additional 
> base diodes from the popular Kohler circuit to reduce the collector 
> conduction at saturation and thus reduce the power-eating turn-off 
> transition where the transistors see both voltage and curre
> nt at the same time. I found that with the right transistors the 
> efficiency of this circuit to be very close to 100% as measured both from 
> the RF and DC powers and the increase in case temperatures from 0 power to 
> a continuous 1W. The second place pair turned out to be the 2N4401/4403 
> but the current gain at upper current levels did not hold up as well as 
> the 2N6601/6651. The popular 2N2222/2907 was acceptable in most cases but 
> did not balance all that well in my circuit with the 2N2907 requiring more 
> base drive than the 2N2222. BTW, I found that using a Vcc in the area of 
> of 24 volts to be preferable over approx 12V because the collector current 
> is 1/2, and you will see faster switching times and less falloff of 
> current gain at the upper current levels.
>
> The last time I looked for a source for the 2N6601/6651 I was disappointed 
> to find that ON Semiconductor (previously a part of Motorola) did not 
> manufacture them. I found some other sources but the data sheets didn't 
> exactly give me the feeling that these were the same device. I would 
> appreciate knowing what luck others have had in finding these.
>
> Bill Ashlock
>