PI1 PCB error

[joop]

Tracing the PCB of PI1 on page 177, I found that the output if the 74HC221 is connected to the drain of the 2N3819. It should be connected to the gate. The PI2 PCB is correct. Here's the 2N3819 pinout.

[mikebg]

I found the opposite in the NET . Since I have several pieces 2N3819 from different manufacturers, I decided to determine by measuring what is correct. It turned out that I do not have enough knowledge on how to determine which lead is the source and which is the drain. Can someone tell me how to measure this?

[Qiaozhi]

Please note that we have stated in several places in the book that transistor pinouts can vary depending on the country of manufacture, even for what appears to be an identical part. This is demonstrated by the information found by mikebg.
However, in this case Joop appears to be correct, as the drain is shown connected to the middle pin. If you look at the PI-2 layout you will also note that the pinout is correct for the same part.

Thanks Joop. I will add that one to the errata list.

[joop]

Here's my working PI1 build. Notice the position of the 2N3819 (red arrow).

[ApBerg]

Hello Joop

Thanks for the info, can you tell how much current the circuit takes, also I like to know that for your Arduino PI-2 mod... ? I like to make a '9 volt batt' PI but most take to much current.
The PI-2 I made, with the 12F1840 out the BOOK, draws some 100mA so to much for a single 9 Volt batt.

Best regards (en de groeten uit Friesland)

Ap

[ODM]

Junction fets are usually different beasts than mosfets by being symmetric. That symmetric structure means the drain and source can be exchanged. There are usually listings for gate-source voltages etc, but not for gate-drain.

[joop]

Thanks for the info, can you tell how much current the circuit takes, also I like to know that for your Arduino PI-2 mod... ? I like to make a '9 volt batt' PI but most take to much current.
The PI-2 I made, with the 12F1840 out the BOOK, draws some 100mA so to much for a single 9 Volt batt.

Howdy Ap!

I get about 80 mA from PI1. The Arduino-PI almost 100mA. Most of the current comes from charging the coil I guess. I have done a quick test, when I change the TX pulse from 100 to 50 microseconds, current is halved! But sensitivity becomes less, I guess because the coil does not get fully charged.

Groetjes uit Houten!

[ApBerg]

Howdy Ap!

I get about 80 mA from PI1. The Arduino-PI almost 100mA. Most of the current comes from charging the coil I guess. I have done a quick test, when I change the TX pulse from 100 to 50 microseconds, current is halved! But sensitivity becomes less, I guess because the coil does not get fully charged.

Groetjes uit Houten!

Hello Joop,

Thanks for checking !

Best regards.

Ap

[Carl-NC]

80mA is correct, as I recall.

[mikebg]

Junction fets are usually different beasts than mosfets by being symmetric. That symmetric structure means the drain and source can be exchanged. There are usually listings for gate-source voltages etc, but not for gate-drain.

ODM, does this mean that in order to determine which leadl is the source and which is the drain, I should connect the JFET as an amplifier and test the gain (parameter Y21)? I used this method, which shows equal resistances:

[ODM]

ODM, does this mean that in order to determine which leadl is the source and which is the drain

It means that drain is source and source is drain, in a jfet. It's approaching Zen, just with physical symmetry! At least one old Motorola app note (Field Effect Transistors in theory and practice) says that the names of terminals depend on the direction of current flow. That's why a single JFET works as an analog switch, with some precautions on signal amplitude, and signal/gate voltage difference.

MOSFET internal structure is drastically different, with the gate being insulated and parasitic diode/bjt components being present.