| Kai Klaas 10/06/09 19:40 Read: 814 times Germany |
#169482 - No surprise... Responding to: Chico Magalhães's previous message |
Chico said:
I discovered that putting a pullup in the output the ULN starts to work normally... That's no surprise at all, because the AT89S52's outputs cannot source much current. No more than 300µA at an output voltage of 2V, at "worst case". If you connect an input of ULN2803 to an output of AT89S52, then you must forward bias two base emitter junctions. Makes 1,4V. In parallel to these pn-junctions about 10kOhm is present, which results in a current of 1,4V / 10kOhm = 140µA each. This current must also flow through a 27kOhm resistor, which gives a further voltage drop of about 3.8V. So, each output of AT89S52 would have to deliver 280µA at an output voltage of about 5V, not even mentioning the needed base currents. This is way too much load for a standard 8051's output. Your added pull-up then turned the situation and the ULN2803 started to work properly. Chico said:
Well... OK... BUT... why was working with the other microcontroller (the now burnt one) ??? Manufacturing tolerances! The old one could, the new one cannot. Kai |
| Topic | Author | Date |
| Trouble with a simple project with ULN2803 | Chico Magalhães | 10/06/09 16:38 |
| Start to search step by step | Per Westermark | 10/06/09 17:18 |
| back to breadboard | Chico Magalhães | 10/06/09 17:53 |
| No surprise... | Kai Klaas | 10/06/09 19:40 |
| Thanks Kai | Chico Magalhães | 10/06/09 21:00 |
| In short | Neil Kurzman | 10/06/09 22:17 |
| Then have a look at the datasheet! | Kai Klaas | 10/07/09 10:14 |
| Don't think so | Per Westermark | 10/07/09 11:09 |
| No, Per, the actual reason is... | Kai Klaas | 10/07/09 16:59 |
| 'working' again | Erik Malund | 10/08/09 05:08 |
| But why? | Kai Klaas | 10/08/09 09:39 |
| Lack of knowledge of the unknown | Per Westermark | 10/08/09 12:29 |
| Ah yes | Steve M. Taylor | 10/08/09 13:01 |
| I think people don't just realize how much they do not know. | Andy Neil | 10/08/09 14:17 |
| There is a reason ... | Richard Erlacher | 10/08/09 23:47 |
| why was working with the other microcontroller | Andy Neil | 10/07/09 01:41 |
| Keep in mind it's on a "breadboard" | Richard Erlacher | 10/07/09 14:05 |
| Infinite Improbability Drive? | Andy Neil | 10/08/09 01:01 |
| re: Infinite Improbability Drive | Andy Peters | 10/08/09 11:06 |
| Pah | Steve M. Taylor | 10/08/09 11:36 |
| Vote: Offensive/Flame! | Andy Neil | 10/09/09 00:33 |
| ;-) | Steve M. Taylor | 10/09/09 00:40 |
| Reference for the uninitiated | Andy Neil | 10/09/09 02:49 |
Without pull-up???  | Per Westermark | 10/07/09 02:06 |
| because it was not working, but 'working' | Erik Malund | 10/07/09 07:58 |
| What else is on your board... | Andy Neil | 10/06/09 17:23 |
| It's a transistor array ... not a magic box | Richard Erlacher | 10/07/09 08:35 |
| AGAIN!!! =( | Chico Magalhães | 11/10/09 19:52 |
| What are you connecting to outputs? | Daniel Contarino | 11/11/09 01:09 |
| Yes...! | Kai Klaas | 11/11/09 08:57 |
| new measures with load | Chico Magalhães | 11/11/09 09:37 |
| How? | Kai Klaas | 11/11/09 09:45 |
| Can use an oscilloscope? | Daniel Contarino | 11/11/09 09:55 |
| Laser and oscilloscope | Chico Magalhães | 11/11/09 10:20 |
| Need current control, not voltage control | Per Westermark | 11/11/09 11:26 |
| re Per | Erik Malund | 11/11/09 11:43 |
| Definitely a good way go let out the magic smoke | Per Westermark | 11/11/09 12:03 |
| not even then | Erik Malund | 11/11/09 12:34 |
| The problem is not the laser. | Chico Magalhães | 11/11/09 13:45 |
| Drive the ULN harder. | Per Westermark | 11/11/09 14:22 |
| Factor 10 off? | Per Westermark | 11/11/09 15:55 |
| I guess, you forgot the Ucesat of ULN2803... | Kai Klaas | 11/11/09 19:00 |
| 74HC04 | Chico Magalhães | 11/11/09 20:38 |
| Better to have them.. | Kai Klaas | 11/12/09 08:52 |
| up the garden path | Erik Malund | 11/12/09 12:24 |
| Definitely consider min/max values from datasheet | Per Westermark | 11/12/09 18:15 |
| Nothing more instructive than... | Kai Klaas | 11/12/09 18:50 |
| but he won't | Erik Malund | 11/13/09 07:04 |
| Yes, that's true... | Kai Klaas | 11/13/09 08:06 |
| a better way | Erik Malund | 11/13/09 11:40 |
| Datasheet | Chico Magalhães | 11/13/09 12:44 |
| OH BOY | Erik Malund | 11/13/09 14:23 |
| Logic level MOSFET... | Kai Klaas | 11/13/09 15:11 |
| A new test I made, bizarre... | Chico Magalhães | 11/14/09 00:00 |
| It seems the power supply... | Daniel Contarino | 11/14/09 00:17 |
| Lack of information | Per Westermark | 11/14/09 03:04 |
| yep. It IS very much! | Daniel Contarino | 11/14/09 03:44 |
| Laser and peltier | Chico Magalhães | 11/14/09 07:38 |
| Dont konw wich regulator you use. | Daniel Contarino | 11/14/09 08:12 |
| well... no regulator at all... | Chico Magalhães | 11/14/09 11:09 |
| PC supplies are not very well regulated | Richard Erlacher | 11/23/09 22:35 |
| For 3V you should have a 3V regulator | Per Westermark | 11/14/09 08:23 |
| That's the explanation! | Kai Klaas | 11/14/09 08:56 |
| 3v | Chico Magalhães | 11/14/09 11:20 |
| Minimum load | Per Westermark | 11/14/09 11:40 |
| Supply and 7805 | Chico Magalhães | 11/23/09 15:16 |
| you want help?? | Erik Malund | 11/23/09 16:35 |
| No laser Datasheet | Chico Magalhães | 11/23/09 22:52 |
| then return it and DEMAND one with a datasheet | Erik Malund | 11/24/09 06:28 |
| Already mentioned lack of datasheet | Per Westermark | 11/24/09 06:40 |
| suggested as first step | Erik Malund | 11/24/09 07:11 |
| laser | Chico Magalhães | 11/24/09 07:13 |
| Hard to tell but I think you should mind the voltage | Per Westermark | 11/24/09 07:40 |
| Perhaps you should use an adjustable regulator | Richard Erlacher | 11/23/09 17:27 |
| I do not think so | Erik Malund | 11/23/09 19:36 |
| and how would YOU build a current source? | Richard Erlacher | 11/23/09 22:30 |
| line anyone else - a 'cookbook' design | Erik Malund | 11/24/09 06:29 |
| can you do that without a regulated voltage source? | Richard Erlacher | 11/24/09 07:16 |
| yes | Erik Malund | 11/24/09 08:11 |
| Guys, Chico has a Laser modul, already... | Kai Klaas | 11/24/09 08:52 |
| two possibilities | Erik Malund | 11/24/09 09:15 |
| assumptions are dangerous | Per Westermark | 11/24/09 09:25 |
| But I didn't assume, that it has a good current control! | Kai Klaas | 11/24/09 10:32 |
| Stable is one thing - what voltage is another | Per Westermark | 11/24/09 11:50 |
| If only Chico had sent us a link to his Laser modul... | Kai Klaas | 11/24/09 12:25 |
| probably highly unstable | Erik Malund | 11/24/09 12:54 |
| Yes, and a good linear regulator can supply that! | Richard Erlacher | 11/24/09 22:56 |
| That was my point! | Richard Erlacher | 11/24/09 22:56 |
| we can't know for certain, but ... | Richard Erlacher | 11/25/09 23:52 |
| What is an ST2221 | Richard Erlacher | 11/24/09 22:53 |
| ST2221a != ST2221 | Per Westermark | 11/24/09 23:18 |
| didn't you notice the reference to an earlier post? | Richard Erlacher | 11/24/09 23:32 |
| you will need one of these | Erik Malund | 11/25/09 06:04 |
| Yes, if you have a good, stabile supply without much noise | Richard Erlacher | 11/25/09 23:46 |
| A current source isn't a magic beast | Per Westermark | 11/26/09 02:57 |
| ... but only if ... | Richard Erlacher | 11/26/09 15:59 |
| and I thought this was turkey day | Erik Malund | 11/26/09 18:33 |
| Are you assuming CC has to be passive? | Per Westermark | 11/26/09 18:52 |
| that would be ridiculous | Erik Malund | 11/27/09 06:41 |
| Ask Richard. | Per Westermark | 11/27/09 06:50 |
| other than Richard? | Erik Malund | 11/27/09 07:08 |
| Now define "constant" | Richard Erlacher | 11/27/09 16:22 |
| Consider this ... | Richard Erlacher | 11/28/09 16:02 |
| Of course, it is... | Kai Klaas | 11/28/09 19:30 |
| Current control != Constant Current | Per Westermark | 11/28/09 22:05 |
| irrelevant | Erik Malund | 11/30/09 08:43 |
| Why do you continue in this way, Erik? | Richard Erlacher | 11/30/09 16:30 |
| Minimalistic const current regulator | Per Westermark | 11/30/09 19:23 |
| It's a matter of definition | Richard Erlacher | 12/01/09 09:07 |
| Ideal constant current source... | Kai Klaas | 12/01/09 09:30 |
| That's where this whole thing began ... | Richard Erlacher | 12/01/09 17:28 |
| Don't jump around wildly when cornered | Per Westermark | 12/01/09 23:43 |
| That's not what was said | Richard Erlacher | 12/03/09 14:26 |
| That's why we should avoid to offend each other... | Kai Klaas | 12/03/09 15:00 |
| Still trying to smash in open doors | Per Westermark | 12/04/09 01:44 |
| The world is flat! | Andy Neil | 12/04/09 02:11 |
| VERY "poor choice of wording" | Erik Malund | 12/04/09 16:17 |
| It wasn't MY wording to which I referred, Erik | Richard Erlacher | 12/05/09 09:21 |
| Memory lapse? | Per Westermark | 12/05/09 14:49 |
| You missed the point, Per | Richard Erlacher | 12/05/09 22:16 |
| I didn't even bother .... | Erik Malund | 12/05/09 18:06 |
| Why, then did you mention it? | Richard Erlacher | 12/05/09 22:18 |
| Lots of ridiculous claims later... | Per Westermark | 12/06/09 04:00 |
| ABSOLUTELY NOT | Erik Malund | 12/06/09 06:20 |
| Cunning ? | Steve M. Taylor | 12/06/09 14:14 |
| Perhaps, but ... | Richard Erlacher | 12/07/09 10:56 |
| What definition? CC/CV both have line regulation tolerances | Per Westermark | 12/01/09 09:56 |
| same old, same old, same old | Erik Malund | 12/01/09 12:48 |
| There you go again ... waving your hands ... | Richard Erlacher | 12/01/09 17:34 |
| already answered | Erik Malund | 12/01/09 12:45 |
| is that what you think? | Richard Erlacher | 12/01/09 17:47 |
| language, language, language | Erik Malund | 12/02/09 14:42 |
| a rose by any other name ... | Richard Erlacher | 12/03/09 13:44 |
| nearly any linear regulator can do that | Richard Erlacher | 11/27/09 14:19 |
| Yes, indeed | Richard Erlacher | 11/27/09 12:53 |
| confound it!!! | Erik Malund | 11/27/09 14:21 |
| Will an AC supply be adequate for you? | Richard Erlacher | 11/27/09 16:21 |
| that does not deserve an answer | Erik Malund | 11/27/09 16:37 |
| If not, then show us what is adequate! | Richard Erlacher | 11/27/09 16:53 |
| Now see here ... | Richard Erlacher | 11/27/09 16:46 |
| Not according to the datasheet I have ... | Richard Erlacher | 11/27/09 16:29 |
| of course it does | Erik Malund | 11/27/09 16:40 |
| So what does that schematic mean? | Richard Erlacher | 11/27/09 17:09 |
| re your babble about a constant supply voltage | Erik Malund | 11/27/09 17:21 |
| If that were the case ... | Richard Erlacher | 12/02/09 19:30 |
| Time to return to earth. | Per Westermark | 12/03/09 00:59 |
| What are you trying to prove? | Richard Erlacher | 12/03/09 14:13 |
| Perhaps, but one with no supply? | Richard Erlacher | 11/27/09 16:26 |
| NOBODY, NOBODY can make this work .... | Erik Malund | 11/14/09 11:53 |
| How about LM317? Look it up! | Richard Erlacher | 11/14/09 12:46 |
| As I said here... | Kai Klaas | 11/14/09 16:04 |
| Just a minute, Chico ... | Richard Erlacher | 11/11/09 23:05 |
| Double check | Daniel Contarino | 11/11/09 11:34 |
| I see no pullups | Erik Malund | 11/11/09 06:23 |
| "The commom of the resistor array is connected to +5v" | Chico Magalhães | 11/11/09 08:55 |
| Not what I mean! | Daniel Contarino | 11/11/09 09:17 |
| He probably needs pullups on both inputs and outputs. | Richard Erlacher | 11/13/09 12:55 |
| be specific!!! | Erik Malund | 11/13/09 14:27 |
| Now I understand ... | Richard Erlacher | 11/13/09 17:33 |
| au contraire | Erik Malund | 11/13/09 19:10 |
| Gate Threshold Voltage, what is? | Chico Magalhães | 11/26/09 05:54 |
| Did you really try Google? | Per Westermark | 11/26/09 06:04 |
| I tried google... | Chico Magalhães | 11/26/09 06:38 |
| Vgs > Vgs(th)! | Kai Klaas | 11/26/09 08:46 |
| I GAVE UP! | Chico Magalhães | 01/12/10 22:55 |
| Don't give up! | Andy Neil | 01/13/10 01:30 |
| Yes, don't give up! | Kai Klaas | 01/13/10 07:41 |
DONT GIVE UP, did you try ... | Erik Malund | 01/13/10 08:17 |