ZBasic Forum

Question 1 - Can a user substitute the 1.0uf electrolics for the Max232 with
ceramic 1.0 uf caps?
Any reliability risk?

Question 2 - Should there be current limiting (1k) resistor(s) in series with the Rx line in case of bit reversals on the Atmel32 due to programming errors?
High on max 232 out and low on pin 14 (ZX-40 as an output instead of an input) or visa versa will short the chip?

Question 1 - Can a user substitute the 1.0uf electrolics for the Max232 with ceramic 1.0 uf caps?
Any reliability risk?

You could, however, that is a rather large value for a ceramic cap. Note that the Max232A (or similar devices like the ST232A can operate with 0.1uF capacitors, typically ceramic.

Question 2 - Should there be current limiting (1k) resistor(s) in series with the Rx line in case of bit reversals on the Atmel32 due to programming errors?

Such limiting resistors seldom cause a problem. However, the serial input line to the AVR cannot be made an ouput as long as the USART is enabled. To get pin 14 to be an output you'd have to mistakenly disable the USART and mistakenly program bit 0 of port D to be an output. I would aver that the risk is fairly low.

1.0µF 805 size ceramic SMT caps are readily available.

Additional ZX-40 Schematic Questions:

Question 1) What is the purpose of the 6.8K pull-up resistor on the com1 tx line (PD1 pin 15)? Is this component actually necessary if one of the Max TTL-RS232 transceiver solutions is employed to implement the com1 interface?

Question 2) A DigiKey PN is suggested for the 14.7456Mhz crystal but nothing is provided for the associated 22pf capacitors. Are these ceramic disk parts and is there a preffered DigiKey PN?

CT

The pullup on TX is described in the manual here: http://www.zbasic.net/doc/ZBasicRef/ZBasicRef111.html. See the last but one paragraph on the page.

The preferred value is actually 27pF and can be pretty much any ceramic capacitor. See the ATmega documentation for futher detail.

dhouston wrote:1.0µF 805 size ceramic SMT caps are readily available.

It's true but for this kind of application a lower powered transceiver is more than adequate so I use MAX202 or ST232/202 instead and use 0.1uF caps. For example the ZX24e uses a TSSOP16 packaged MAX202 transceiver and 5 0.1uF 0603 caps. The fifth cap is used as a bypass across pins 15/16 of the MAX202.

Question 1) What is the purpose of the 6.8K pull-up resistor on the com1 tx line (PD1 pin 15)? Is this component actually necessary if one of the Max TTL-RS232 transceiver solutions is employed to implement the com1 interface?

Some of the level translator devices like the MAX parts have pullup resistors (TTL inputs) and pulldown resistors (RS-232 inputs). If you select a device that does, you don't need additional pullup/pulldown resistors.

The schematic should be amended to show a pull down on the Rx data line. This is needed, assuming that the level translator doesn't have one, to keep the receive line from floating when you remove the serial cable. If you'll always keep it connected, it's not needed.

Question 2) A DigiKey PN is suggested for the 14.7456Mhz crystal but nothing is provided for the associated 22pf capacitors. Are these ceramic disk parts and is there a preffered DigiKey PN?

The load capacitors for crystals are almost always ceramic disc or ceramic chip capacitors. I have used DigiKey p/n 1430PH-ND for the ZX-40. The ZX-24 uses DigiKey p/n PCC270ACVCT-ND. That is an 0603 SMT part. For an 0805 part, use DigiKey p/n PCC270CGCT-ND.

Note that the actual capacitance value that you use will depend somewhat on the component layout. It is easy to pick up a picofarad or two of stray capacitance with point to point wiring on a prototype. Additional capacitance will make the crystal resonate slightly below its target frequency. Too little load capacitance will make the crystal resonate slightly above its target frequency.

The schematic should be amended to show a pull down on the Rx data line.

Should be a pullup shouldn't it? Rx and Tx are normally high logic levels.

Should be a pullup shouldn't it? Rx and Tx are normally high logic levels.

You are correct that on the TTL side the idle state is high. That's why you want a pullup on the TTL inputs to the level converter. However, on the RS-232 side, the idle state is low. Hence, you want a pulldown on the RS-232 inputs to the level converter.