PDA

View Full Version : Interfaceing Propeller to 74LS245's



lynchaj
01-09-2009, 01:07 AM
Hi!· I am thinking about a Propeller project and would like to confirm what I think I learned from the hardware interfacing sticky message.

I would like to attach 5v TTL 74LS245's to a 3.3v Propeller.· Is the circuit in this thread

http://forums.parallax.com/showthread.php?p=585920

safe and reliable for long term operation?· Specifically the voltage divider network that Tracy Allen posted with the 1K series resistor and the 10K pull up resistor on the 3.3v side.· I would like to skip the 3.3 zener diode as it appears to be not necessary.

Thanks and have a nice day!

Andrew Lynch

Leon
01-09-2009, 01:13 AM
Why don't you use 74HC245s? LS logic isn't used much these days.

Leon

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Amateur radio callsign: G1HSM
Suzuki SV1000S motorcycle

Ale
01-09-2009, 01:45 AM
lynchaj: The HCT version would be better suited than the LS due to less drain, better load drive (equal to HC series) and TTL compatible threshold. The HC has a 2.5 V threshold (supposedly) or 3V, I think HCT is better due to bigger margin. It works for me very well since months! (series resistors are still needed if you are going to read). There are true level translators, they do not come in DIP packages though. (TLVC series if I'm not mistaken).

lynchaj
01-09-2009, 02:04 AM
Ale said...
lynchaj: The HCT version would be better suited than the LS due to less drain, better load drive (equal to HC series) and TTL compatible threshold. The HC has a 2.5 V threshold (supposedly) or 3V, I think HCT is better due to bigger margin. It works for me very well since months! (series resistors are still needed if you are going to read). There are true level translators, they do not come in DIP packages though. (TLVC series if I'm not mistaken).
Hi!· Thanks!

I agree that the 74HCTxxx chips are more modern and in most ways better than their 74LSxxx predecessors.· HCT uses less power, less current, improved fanout, etc.· However, the 5v to 3.3v interfacing issue is my primary concern since I would like the Propeller to drive a legacy circuit and I need a way to reliably "insulate" the effects of the 5v system on the Propeller.· I am mostly interested in the 74LSxxx chips because that is what the legacy circuit uses and I would like to preserve it unmodified.

At least in my background, mixing chips of differing source voltages and levels is·not recommended·because even though it may work temporarily it can have long term stresses on the lower voltage parts which reduce their life spans and reliability.· I would like to make sure the voltage divider approach works long term and not just experimentally or under special conditions like taking special steps to make all pins inputs or outputs when not being used.· The voltage divider approach looks sound but I would like to confirm it and make sure there are no special conditions associated with it.

For the most part, 74HCTxxx and 74LSxxx are level compatible and interchangeable.· Personally, I would like to avoid the HC chips because they introduce level mismatch issues I would rather avoid especially with the legacy circuit.· Ideally there would be a 74245 like bus transceiver DIP chip that would allow 3.3v operation on one side and 5v TTL on the other.· Those devices do exist but are only SMD AFAIK.

Thanks and have a nice day!

Andrew Lynch

Post Edited (lynchaj) : 1/8/2009 6:09:31 PM GMT

kwinn
01-09-2009, 09:08 AM
Try the 1K resistors. They should work (have in my case for 6+ months). There are level translators available, and you could also use a transistor array if necessary.

sosaraujo
02-01-2009, 05:04 AM
Hi,
Then, I should be able to build this circuit?
www.sosaraujo.com/imagenes/bus-transceiver.pdf (http://www.sosaraujo.com/imagenes/bus-transceiver.pdf)

Thank you

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
-------
Adrian Sosa Araujo.- San Luis, ARG - Time Zone (GMT -3)
El planteo de un problema es mas importante que su solución.

Ale
02-01-2009, 06:37 AM
The circuit looks fine to me, similar to the one I use but with OE tied to ground so I do not need pull-ups. Do not forget some decoupling cap (.1uF ceramic) for the power pin of the 245.

sosaraujo
02-01-2009, 02:31 PM
I've made some changes on the schematics. Two pull-down resistors and a decoupling cap.
www.sosaraujo.com/imagenes/bus-transceiver.pdf (http://www.sosaraujo.com/imagenes/bus-transceiver.pdf)

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
-------
Adrian Sosa Araujo.- San Luis, ARG - Time Zone (GMT -3)
El planteo de un problema es mas importante que su solución.

Ale
02-02-2009, 04:29 AM
Adrián: El circuito está bien, los pull-down para prevenir estados indeterminados durante la inicialización son también una buena idea. No te olvides de usar preferentemente un HCT245 al menos que tengas que manejar lógica LS después, probá que da mejor resultado. (Los HC tiene otras tensiones de salida y pueden manejar mucha más corriente).

Saludos desde bueno Alemania, aunque yo vivía en Pcia de Buenos Aires http://forums.parallax.com/images/smilies/smile.gif

Ale

sosaraujo
02-02-2009, 06:05 AM
look who's here! Gracias Ale. Thank you Ale.
Saludos desde San Luis. Ya sabés, si necesitas algo de por acá..... lo que sea, menos integrados, que no se consiguen y tengo que pedirlos a BsAs. :)

and continuing on the subject...
what if I use a cd4034B (or MC14034B) instead a MC74HCT245A ? with a Vdd=3.3V .

thank you.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
-------
Adrian Sosa Araujo.- San Luis, ARG - Time Zone (GMT -3)
El planteo de un problema es mas importante que su solución.