# 74act high output curve (voltage vs current)

Discussion in 'Electronic Design' started by Mook Johnson, Dec 13, 2005.

1. ### Mook JohnsonGuest

I'm looking to power an array of 32 small 5V 4 - 6mA devices with the
outputs of a 74 ACT bus driver or similar. I'm doing ths because I only
power one device at a time and only for a short period of time (say 200mS
then off for 1 second).

I know the 74AC(T) line is rated for 24mA output current but I would like to
know the output voltage drop as the current increases from 0 through 24mA (a
curve). Or more specifically what the voltage is at 5mA draw. I need the
voltage to be fairly stiff as the output of these devices are ratiometric
with the supply voltage.

The datasheets only specify Voh at a few uA and at 24mA output. I don't
think voltage drop would be linear (might be).

I know the proper way is to use a bank of low RDS p-mos but I'm trying to
reduce parts count.

2. ### Jim ThompsonGuest

I would expect it to look pretty much resistive, until you start to
run out of IDSS.

...Jim Thompson

3. ### Fred BloggsGuest

Even if there was an IV curve it would just be "typical." From the data
sheet of the TI 74AC244 you can infer that RDS,ON of the pull-up PMOS is
no more than 27 ohms from the minimum Voh specifications, and this seems
to be the limit in the Vcc=4.5 to 5.5 volt range. At your 6mA load, this
makes for a 0.006 x 27=0.16V drop which amounts to 0.16/5= -3.2% just
due to the driver. The typical might only be -2.5%, which may or not be
considered "stiff" depending on the load tolerance.

4. ### JoergGuest

Hello Mook,
It'll be different from device to device as Fred had hinted. So if you
have 32 loads driven by four chips and one of the chips happens to
orginate from a different batch or date code you may have a discrepancy
in RDSon and thus voltage.

Why not replace them with some stiff ULN driver? These come in SO
packages. or do you need the speed?

Regards, Joerg

5. ### mook JohnsonGuest

Looks interesting but the forward drop looks like its 0.6V even at low
currents.

6. ### Jim ThompsonGuest

[snip]

WRONG current direction, you're measuring the ESD diode.

Put a 74ACT... into an output high state, load output with a resistor
to ground (various values), THEN you will know how much load it can
support.

...Jim Thompson

7. ### nospamGuest

You can find IBIS models for various parts here

http://focus.ti.com/logic/docs/logictechdoc.tsp?familyId=254&navSection=models&templateId=5985

The models are not that easy to read but do effectively contain plots of
pin current against pin voltage.

I get confused about which way round the pullup and pulldown plots are used
and what the reference for the voltage is but it looks to me like an ACT244
would typically be dropping about 70mV at 5mA either way.

8. ### JoergGuest

Hello Mook,
Shouldn't be 0.6V even with bipolar outputs but you'd have to find one
with FETs. If the load absolutely has to be grounded look for 'high side
drivers', else for 'low side' (easier). Some of them can sink an
incredible amount of current and thus offer a very low RDSon.

Regards, Joerg