Relay driver - ULN2003?

The voltage drop of a darlington is a bit much at 5V, but for 12 and

24, sure. These are still very popular parts, I wouldn't (don't) worry about it. Allegro probably just wants to make more $1.50 parts rather than 15-cent parts.

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

There are a lot of "smart" low-side and a number of high-side logic drivers which can handle a respectable amount of current, enough to replace not only the ULN2003 but the logic driving it. For example the TPIC6C596 which I love as an "output expander".

Sprague (oops, I mean Allegro) also sells "smart" driver chips.

I doubt that the other makers of ULN2003's (off the top of my head, TI, Toshiba, and STMicro) will stop soon.

But I could be wrong. Nobody makes CA3046's anymore, for example...

Tim.

What's a good octal relay driver with integral shift register and latch? For 3.3V ("TTL-level" from 3.3V cmos) logic. My relays draw 33mA, but that's too much for HC, etc., to drive.

--
 Thanks,
    - Win

TPIC6C596 looks interesting, but I haven't used it. We were going to, but then cut over to bipolar-drive latching relays to zap thermal EMFs, so we had to do something else.

But it needs 5 volts and isn't (legally) 3.3v compatible on the logic inputs. More danged level shift issues!

John

ST still makes 'em, never even seen the Allegro version. I use them quite often for relay drivers and so on. Perfect fit for an 8 bit port to open collector lamp or relay drive. I don't see any reason to not use them in on future projects.

Jim

Right. Yes, but TI gives its 250mA-capability specs at 4.5V min, yet shows curves for operation down to 3V. I'm only driving 35mA relays, so the higher 1.8-ohm Ron isn't a problem. The $1.17 at DigiKey doesn't look too bad either. As for Spehro's suggestion of hc595 plus uln2803, yes the steering-diode pin has its appeal, but at 35mA it's lessened. Plus a bad mark, the Darlington drop puts me out of spec for the dinky struggling 5V relay.

--
 Thanks,
    - Win

Why not use a HC595 and ULN2803? Two cheap parts instead of one part, but you don't have to worry about that 250mA+ through the ground pin.

I've used the TPIC6595 and similar (single-sourced) TI parts in products, but the voltage doesn't quite match, as John L points out, and they can get grumpy about layout issues. They also force the DMOS output transistors to absorb the inductive 'kick', rather than steering it out of a pin where you can do as you will with it.

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

--
If you can stand to do a level conversion on the serial data, the data
clock, and the latch clock inputs, ( to 5V from 3.3V ) an Allegro
A6275 would work for you.

Ok, I'll see your TI tpic6a595, 6b595 and 6c595, your ST 6c595, your Allegro 6b595, your TI c6595, *ahem* and raise you a SamHop smb6b595. Aha! Gotcha! You can fold now.

--
 Thanks,
    - Win

huh?

martin

On 19 Jul 2005 06:09:47 -0700, " snipped-for-privacy@gmail.com" wroth:

I've recently used a MAX7301 driving a bunch of National's DS3668 chips. An SPI port can give you up to 28 outputs or a mix of outputs and inputs. Works fine at either 3.3 or 5.0 volts for the logic.

Jim

Hello James,

And TI, and Toshiba, and...

Regards, Joerg

Hello Martin,

Intersil's site hung again but that didn't surprise much. Digikey says that the last order bell had been clanging. The usual: Obsolete item. Quantity available zilch. Call for a copy of the obituary...

Regards, Joerg

Yes, that's very interesting, thank you for the pointer, John. ST seesm to be second-sourcing some of the TI CMOS op-amps too.

The part number I mentioned (without the C) has much lower Rds(on) than the TPIC6C595 (and more pins-- 20 rather than 16) but that's certainly not necessary for every application. One very nice feature you get with the extra pins is separate power and logic grounds.

Here's an interesting application note:

Definitely split the ground plane on this one. ;-)

data sheet:

Looks like typical Rds(on) at 3V/125°C is 3 ohms. If we can scale that by 2:1 for worst case (??) that would only be 200mV drop at

TPIC6C595 data sheet:

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

"Winfield Hill" schreef in bericht news: snipped-for-privacy@drn.newsguy.com...

ST's L9822N on 4.5V, perhaps. Nah...

--
Thanks, Frank.
(remove \'q\' and \'invalid\' when replying by email)

Hey, Spehro -

If you meant the TPIC6C595, there is also the STPIC6C595 by ST.

Just thought you'd like to know.

Cheers, John

Doesn't seem to have built-in diodes (to suppress the back-emf from relay coils).

--

John Devereux

They are avalanche-rated. This has advantages and disadvantages.

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Good points, Spehro. Thanks.

John