More current than ULN2003 and pin compatible

Frank,

Just this week we released an upgrade for the ULN2003AD, it is called the TPL7407L. It uses NMOS transistors to get a lower VOL than ULN2003's Darlington NPN. Therefore it will run cooler than the ULN2003 in this application. The only reduced specification is maximum voltage, 40V instead of 50V.

7 channels at 250mA is 1.75 ampere of current, so thermal concerns might still exist. 
For manual modifications you may use a glue on heat sink or stack a second IC on top or increase air flow.

Is the ULN2003AD causing any immediate concerns?

Ron,

Thanks for your fast response.

We are using it on two 24V relay coils that require 0.52A each. The circuit is set up with two  Darlingtons in parallel. In the most stringent situation, there will be 4 transistor outputs turned on to deliver 0.52A x2 = 1.04A total current. On each transistor channel, we would need 0.26A or better.

The other Darlington outputs in the circuit have some resistors and LEDs that take little current. All of the Darlington inputs are 15V high.

We can't use the ULN2003AD because  the graph titled: D PACKAGE  MAXIMUM COLLECTOR CURRENT   vs   DUTY CYCLE shows 0.12A for N=4 at 100% duty cycle. 

Our product is specified to run up to 50 degrees Celsius. I see the graph of the TP7407L specified for N=4 at 0.24A at 70C and 0.31A at 25C cutting it close. 

Can you recommend a particular style of heat sink that is very common for this SOIC shape?

Frank

Ron,

I saw that the TP7407L is so new that it is not stocked yet. Do you have any estimate when it will be available?

Thanks,

Frank 

Frank,

The TPL7407L de-rating curve does support 260mA * 4CH @ Ta= 50C, but not by a large margin. This data sheet chart is based data sheet Tja which is based on standard JESD test board performance. If the actual application board has better thermal properties than it should work well. If the application board has many heat producing components near TPL7407L then there will be over temperature concern.

I am checking on TPL7407L sample date. I do not have a heat sink recommendation at hand, but it would be good to see how much advantage a heat sink will provide.

The 15V input signal will not damage a ULN2003A or TPL7407L, however this high input voltage will cause more device heating. ULN2003A input impedance is 2.7k (above 1.5V) and TPL7407L input impedance is 50k (above 6V). Input power for ULN2003A is 7CH * (15V-1.5V)^2 / 2.7k = 473mW. Input power for TPL7407L is 7CH * (15V-6V)^2/50k = 2mW. The TPL7407L advantage here is very clear.

Hello Frank,

I am product marketing for TPL7407L. We have released that part about a week ago and you can now order free samples from our web product folder (http://www.ti.com/product/tpl7407l) under the "Sample & Buy" sections. You have 2 package available (TSSOP and SOIC).

Bruno

Bruno,

Thanks, I'll get some. The footprint on our PCB is SOIC.

Ron,

When I mentioned the 15V on the inputs, I omitted one detail that I didn't think would matter at that time. On our PCB, I already put a 4.75k resistor in series with each ULN2003AD input. When the 15V digital input is present, I measure 4.5V on the input pin. It comes out to 10mW dissipation per input chanel. 

Frank

Frank,

I have not been successful in finding smaller than foot print heat sinks. I checked Newark and Digikey.
The ones I found are over-width. They would work, but trying to remove the IC from the board would be very difficult after the heat sink was glued.

Newark
http://www.newark.com/fischer-elektronik/ick-smd-a-10-sa/heat-sink/dp/34M6429

Digikey
http://www.digikey.com/product-search/en?WT.z_header=search_go&lang=en&site=us&keywords=AE10837-ND+&formaction=on