Choice between CMOS 4099 or 74HC259

What do you think to gain?

Dunno! I am a novice. (Okay, I have picked up a soldering iron occasionally and built some circuits.) The reason I asked was because the 4099 doesn't seem widely available any more, so I thought that's because it's been superseded.

MM

MM,

The way I understand your question is that you want to replace a 74HC25 with a 4099. If this is the case, then you should know that aside fro anything else, they are not pin compatible. Secondly the Clear functio appears to be inverse between the devices.

Digikey seems to have plenty 4099s, so you don't need a circuit redesign.

Generally you need to consider several things when replacing one technolg with another. Firstly the setup times and write times are normally muc longer in 4000 series CMOS. The 74HC series is quicker so you probabl could make the substitution (if pin equivalent), but you do need to mak sure the timing is OK.

Driving 4000 series from any TTL output normally requires a pull u resistor to ensure logic high level compatibilty (provided they are bot run from 5V) unless the 4000 series device has Schmitt trigger input (which the 4099 does not). Drive levels of the 4000 series into TTL ca prove problematic. You need to investigate the output drive capabilty o the CMOS part.

The 4000 CMOS series can operate with a supply voltage from 3V to 15V This can be advantageous in some circumstances, but obviously interfacin with 5V systems would need extra attention.

-Aubrey

This message was sent using the comp.arch.embedded web interface o

Thanks for that, Aubrey. Okay, I now have found Cricklewood Electronics that also do the 4099. Like I said, I'm a novice so when some of the main catalogues didn't stock it I assumed it had become outdated and little sought after.

By the way, the 4099s will be driving LEDs, one per pin. This is in conjunction with MIDI.

Since I'm here, what's the next best choice to the 16F628(A)? I was intending to use it in preference to a 16F84 because the former comes with an inbuilt USART. However, that utilises RB1 and RB2 as RX/TX respectively, which thus lose me those pins as outputs. Pin 8 (RB2/TX) apparently cannot be configured as a bog-standard output when the USART is invoked, even if TX is not required. And all the other pins that could be outputs are already used up.

Therefore I am now considering the 16F877 instead, which has many more pins to play with. However, this PIC is twice the price and has a lot of additional functionality/capacity that I don't need. Is there something in between? For example, do any of the 28-pin PICs have a USART?

MM

MM

For the PIC, do you need an A/D or not?

AS far as the 4099s are concerned, they cannot drive much current. Mak sure you use low current LEDs.

-Aubrey

function

redesign.

technolgy

much

make

both

can

of

interfacing

No.

What about the bog-standard ones I can get from Maplin? (Each 4099 may have to drive a maximum of 8 LEDs, one per output pin.)

MM

As far I concern there is variation of 74HC that could replace 4099 (try 74HC59X or 74HC57X in that range). I would try make use of 74HC since 4099 belong to ice age now. They bound to phase out very soon, just like window 3.1 and 95.

It can work on 3.3 V or 5V, better output drive, faster and more stable device than 4099.

Riscy.

MM

Take a look at this URL fro 28pin PICs. Scroll down looking for UARTS. (O course you could bit bang a serial port)

As far as current is concerned, you should really consult a data sheet. lot depends on the range of temperature you are going to operate in an what voltage you can let the output rise to (in the case of current sink) The outout voltage will increase as the current increases. If you are usin the output to drive other logic, this can prove a problem.

Using the TI data sheet, assuming that you are only operating at 25 degC and that you are sinking current through the LED, you will need a resisto of ((5-0.4)/0.51)Kohms in series with the LED. The device can sink 0.51m guaranteed in the above circumstance with a 5V supply. I doubt you wil even see a dull glow from a "bog standard" LED. You may want to try lo current devices like the HLMP1700 which can work down to 0.5mA withi limits.

You seemed to imply earlier that this was an existing circuit; if it i you are going to have to be a lot more specific to provide help. If not it would be simple to add some form of driver to increase the current.

Make

Bit banging is what others have done with the 16F84, but given that there are alternatives with inbuilt USART it seems churlish not to use one!

Thanks for that. Crikey! Microchip certainly do have a plethora of MCUs!

No, the 4099 would drive just LEDs, up to eight, one per output.

How about if a TTL chip were used instead of a CMOS? Mind you, I'm sure I've hooked up an LED to a CMOS output in the past (with resistor) and there's not been a problem.

MM

Plain TTL, or even LS TTL is as old and outdated as 4000 CMOS and they che current as well. There are a multitude of devices you can use in the 7 format though. Even the 74HC series will be better from current sink poin of view.

It all depends what you want to accomplish as part of the bigger picture If this is part of self education and you need easy access to parts, b all means use 74HC parts.

If you want more esoteric latching parts with higher drive capabilty, tr looking at some parts by Allegro

Texas Instuments and ON Semi used to make compatible stuff, I am not sur if they still do. Try also ST and Philips as well. They sometimes hav some very interesting devices for esoteric applications.

-Aubrey

(Of

and

sink).

using

degC,

resistor

will

low