Interesting chip. 32 LED driver. SAA1064

Hemi345

Hemi345

Senior Member
I used one of these in my last project. It was very easy to use (especially after reading Westaust's primer on it). That isn't a good price on Ebay though, I got a couple for $2US/ea on Mouser. The only thing I don't like about it is the size. I should have ordered one as SMD version for the final project to save board space and prototype with the DIP.
 
A

AllyCat

Senior Member
Hemi345 said:
I should have ordered one as SMD version for the final project to save board space and prototype with the DIP.
Click to expand...
Hi,

Yes, I agree about the (apparently high ebay) price. However, bear in mind that the SMD version has only half of the power dissipation capability of the large DIL package. And the constant current drivers are linear (not switched mode) so there can be considerable thermal heating in the chip.

Cheers, Alan.
 
B

boriz

Senior Member
21mA*32 @ 1.7v vF and 5v Vdd = Over 2.2 Watts !

Big surface area is definitely usefull.

Thanks Westy. Must have missed that. First time I've seen this chip. Got excited :)
 
P

PaulRB

Senior Member
Hi Boriz, if its driving an led cube you're after, have a look at this discussion we had. I got lots of good advice:

http://www.picaxeforum.co.uk/showthread.php?21926-4x4x4-LED-cube-Plan-B-advice-on-driver-chip-selecton-please&highlight=led+cube

Here's the design I ended up with:

http://www.picaxeforum.co.uk/showthread.php?22115-4x4x4-LED-Cube-using-08M2-and-MAX7219&highlight=led+cube

Note that I did unwittingly buy fake chips on ebay. They do work, but I have found the leds in the cube regularly fail. I don't know if this because I also bought poor quality leds (also on ebay), or my poor cube construction/soldering is damaging the leds, or even that the driver chip is to blame.

http://www.picaxeforum.co.uk/showthread.php?22481-Real-or-fake&highlight=real+fake
 
Last edited:
W

westaust55

Moderator
hippy said:
It's multiplexed isn't it ?
Click to expand...
Yes, as two of 4 digits illuminated at a time so worst case is 16 LEDs on so max power dissipation would be more around 1.1 Watts based on figures by Boriz.



EDIT:
However from the datasheet:

Ptot = P1 + P2+ P3
based on Vcc = 5V, 32 segments operating in dynamic mode thus max 16 on at any one time,
Vo = 0.25 V as output switch voltage, and 21 mA per segment

Ptot = 75 mA + (50* 32/32) mW + (0.25 * 16 * 21mA )mW = 209 mW
 
Last edited:
A

AllyCat

Senior Member
Hi,

Well, it's probably (but not necessarily) somewhere between 209mW and 2.2 watts. :)

Certainly it only has 16 (multiplexed) segment drivers, but the data sheet appears to set a further limit (of 200mA) for the Vcc pin (as do PICaxes).

However, the voltage drop(s) can be much more than 0.25 volt. Firstly, the constant-current output stages are intended to replace the series (current-limiting) resistors (so that power will be also dissipated in the chips), then the internal anode drivers (if used) are a rather inefficient emitter-follower configuration, dropping around 1.5 volts. So (as Boriz said), the power dissipation can be (typically) determined by the (supply voltage minus the LED forward drop) times the average current.

But the potential "killer" is that the chip is specified up to a supply rail of 15 volts, so it's really quite easy to exceed its maximum power rating of about 1 watt (DIP) or 0.5 watt (SMD).

Cheers, Alan.
 
You must log in or register to reply here.
Top