# ULN2803 input resistors?

Discussion in 'General Electronics Chat' started by ghall426, Sep 2, 2012.

1. ### ghall426 Thread Starter Member

Jul 22, 2009
11
0
I see most circuits direct drive the ULN2803 Darlington array with 5V, however the datasheet shows VI(on) as 2.4V maximum for a 200mA load. The maximum rating for input voltage is 30V.

I always thought the chip should be driven within the recommended operating conditions. So should I use input resistors to keep the voltage under 2.4V for a 200mA load or is 5V fine?

2. ### absf Senior Member

Dec 29, 2010
1,869
510
The reason I use ULN2803 or ULN2804 is to reduce space to exclud the 1-10K resistors. If driven with TTL I use ULN2803 and if driven by CMOS 6V-15V then I'd use ULN2804.

Allen

• ###### uln2803.PNG
File size:
44.9 KB
Views:
114
Last edited: Sep 2, 2012
3. ### #12 Expert

Nov 30, 2010
18,082
9,650
That's the maximum that the chip will require to do its job, not the maximum before things start melting. If you give the input the alleged maximum voltage, you would be giving it the absolute minimum voltage that would guarantee a proper output.

Sometimes it would help if the datasheets would finish the sentence, "Maximum to do what?"

4. ### Ron H AAC Fanatic!

Apr 14, 2005
7,049
674
Whose datasheet are you looking at?
The 2.4V spec means the input needs to be at least that high to ensure saturation of the output. 5V logic drive will be fine, as long as the ULN2803 input does not load your driver to the point that its output voltage is too low to drive any other logic you have connected it to.

5. ### ghall426 Thread Starter Member

Jul 22, 2009
11
0
Doesn't that make this a "minimum" then?

If I need at least 2.4V, not to exceed 30V, it makes sense that 2.4V should be designated as a minimum and 30V as a maximum.

Jul 22, 2009
11
0
7. ### Ron H AAC Fanatic!

Apr 14, 2005
7,049
674
Post a link to the datasheet. There are several versions of datasheets out there, from different mfrs.

Jul 22, 2009
11
0
9. ### #12 Expert

Nov 30, 2010
18,082
9,650
Yes, that's what I said. The maximum the chip will require to do its job is the minimum you have to give it.

10. ### ghall426 Thread Starter Member

Jul 22, 2009
11
0
It's impossible to have a maximum required value. Maximum means the most. You shouldn't exceed a maximum. All valid operating conditions should be between a minimum and a maximum.

What you mean to say is "The minimum the chip will require to do it's job". The value should be listed on the datasheets a minimum right?

It won't let me post a link. It's a Texas Instruments ULN2803A from Mouser.

Last edited: Sep 3, 2012
11. ### #12 Expert

Nov 30, 2010
18,082
9,650
Perhaps Texas Instruments needs to hear from you about how to write datasheets.

• ###### 366828_DS.pdf
File size:
90.8 KB
Views:
24
Last edited: Sep 3, 2012
12. ### Ron H AAC Fanatic!

Apr 14, 2005
7,049
674
I see why you are confused. It means that the maximum input voltage required to saturate the output is 2.4V. It might work with less voltage, but it is not guaranteed. It certainly will work with more.
What that means is that, to guarantee saturation, the minimum input voltage is 2.4V.
Same thing as #12 said. I just used more words.

13. ### ghall426 Thread Starter Member

Jul 22, 2009
11
0
So am I right or are you being sarcastic? I totally agree that the voltage should be above 2.4V to saturate the transistors and operate the driver, I guess the terminology is just confusing to me.

By the way, this isn't the only datasheet I've seen this on.

14. ### #12 Expert

Nov 30, 2010
18,082
9,650
Trying to be funny about how impossible it would be to make a Semi manufacturer change anything.

ps, I added a copy of the datasheet from ON Semiconductors a coupla posts back.

15. ### ghall426 Thread Starter Member

Jul 22, 2009
11
0
There must be a reason it's noted as a maximim since the Motorola sheet you posted shows the same thing.

16. ### MrChips Moderator

Oct 2, 2009
17,322
5,335
There is often confusion with the use of the terms "minimum" and "maximum" when quoted in data specifications. Sometimes even manufacturers are not consistent with their usage.

In the case with the TI ULN2803, Vin(on) = 2.4V MAX is correct.

What the spec sheet is quoting is not the max allowable input voltage but the range of statistically measured input operating voltage. It is saying that ALL of the devices will operate if the input voltage is greater than 2.4V.

Sometimes you might see something like 2.1V TYP, 2.4V MAX which would make sense.

Sometimes you might see the MAX number being lower than the TYP number and that is when it gets confusing. For example, if the manufacturer were quoting clock frequency, you might see something like TYP = 66MHz, MAX = 60MHz

The bottom line is 5V input is better for 2.4V MAX input.

absf likes this.
17. ### Wendy Moderator

Mar 24, 2008
21,413
2,946
I have unmoderated the previous 3 posts from the OP. Since he was under 10 posts, and was posting links, the forum software automoderated the new posts. His post count has jumped, and I suspect it will not be a problem in the future.

#12 likes this.
18. ### ErnieM AAC Fanatic!

Apr 24, 2011
7,953
1,826
Looking at the TI data sheet the Input Voltage is given as 30 V Maximum. VI(on) is a different parameter showing you what input voltage is required to get a certain VCE, and as such is not a maximum limiting value not to be exceeded. The limit is the 30V maximum.

A 5V input will be nominally limited by the internal 2700 ohm resistor and 2 BE drops to around 1.3 mA. It is completely fine to use this as a driving voltage without any additional components.