Readily Available 8ch, 10-12bit A/D

I am seeking an 8 channel(5 minimum), 10-12 bit ADC... Other than that the specs are pretty loose.

I have found some perfect ICs @ maxim but both samples and orderes require a 4 week lead time.

I am lookign to be able to get the ADC ASAP...

Any suggestions?

Thanks.

If you want it any faster you are going to have to buy it from Digikey, Mouster or some other place...remember shipping...

What are you trying to do?.

Bear in mind a PIC16F877 has eight channels of internal 10 bit A2D, so it's usually much simpler to use that (or something similar) rather than an external A2D and seperate processor - it's also probably cheaper than the seperate A2D?.

have a look a microchip

**broken link removed**

I forgot two other factors

1) I want it in a DIP package
2) The analog voltage input range has to be from 0-12V minimum

Most ADC's I have seen only range from 0-Vcc +/- .6V

Can the input voltage not be scaled into the appropriate range? A resistive divider would do it, or an op-amp circuit if there are impedance issues.

The application requires reading one of about 30 ADC values... 12V was used so each value could fall in range of about .43V=12V/30

If I were to scale down to 5V then each of the 30 values would be associated with a .16V=5V/30 range

With this new smaller range, I feel like there is not enough room for error and I could end up reading 'value 24' when in fact 'value 25' was physically present. Is .16V plenty of 'buffer' or am I correct in giving each value a bigger voltage range?

I'm not sure I follow your reasoning. for a given number of ADC bits, you will still have the issue of LSB accuracy. Yes, noise is an issue but 160 mV is pretty high - with reasonably decent design, you should be just fine. because you will always have some noise, you will want to do some processing on the results anyway.

buurin said:

The application requires reading one of about 30 ADC values... 12V was used so each value could fall in range of about .43V=12V/30

If I were to scale down to 5V then each of the 30 values would be associated with a .16V=5V/30 range

With this new smaller range, I feel like there is not enough room for error and I could end up reading 'value 24' when in fact 'value 25' was physically present. Is .16V plenty of 'buffer' or am I correct in giving each value a bigger voltage range?

Click to expand...

Attenuating the input makes no difference to the accuracy (you've still got exactly the same 10 bit resolution), however you still haven't given any details of what you're trying to do?. But, from the little you've said so far, it sounds like a PIC would do the job easily?.

The ADC input voltage is being produced from a voltage divider using 1% precision resistors and I was concerned that due to the nature of the circuit there may be quite a few points of error... simplified, one ADC input would look like this:

**broken link removed**

R11 is user selectable(1 of 30 values)
R5 is known
This schematic is simplified, the transistors are actually turned on/off via microcontroller
The diode is there for reasons beyond the scope of the simplified schematic

PS Those points of error I was concerned with are the voltage drop across the diode and the 12V transistor.

So you won't give details of what it's supposed to do?, you draw incredibly bizarre 'simplified' parts of circuits with silly components who's reason you won't explain?, so how do you expect to get help?.

if you are worried about the 1% resistors being "off", don't. any errors will be linear.

I have to admint, your circuit baffles me. for one thing, your PNP appears upside down. what are you trying to do here? Are you trying to make a uC controlled voltage divider? I think FETs would be a better solution.

You will suffer the voltage drop of the diode but why not divide the voltage at the point where you have the ADC input shown? that will also divide the diode drop so it is proportionally smaller. I assume you know that the transistors will contribute voltage drops of their own...

It might be a good idea to tell use what you are really doing. I can't help but think this is the hard way...

edit: aw Nigel, you're bein harsh on the laddie. he's probably just trying to avoid giving away his idea for the next big thing...

I apologize for the frustration I may have caused here w/ my lack of info... I know you guys are only trying to help me and I am unquestionably hindering that effort due to my vagueness.

As philba said, I am trying to protect my 'brilliant' idea

If you're still willing to help me I will go back to the drawing board and come back with a much more realistic schematic and explanation.

You MUST give info on what you want to do.
Issues of input impedance, ADC speed, offset, and ADC precision are going to change the situation. Any advice given without knowing these factors is not reliable.

buurin said:

If you're still willing to help me I will go back to the drawing board and come back with a much more realistic schematic and explanation.

Click to expand...

Remember the acronym GIGO (Garbage In, Garbage Out) - in almost every single thread like this one there's never been enough information to give any useful help.

Usually the 'brilliant' idea is nothing of the sort, the entire premise is wrong, and the method of trying to implement it is completely wrong. But as we're all completely in the dark we don't find this out until the thread is many pages long.

This wastes bandwidth, and everyones time - so if you want help you need to provide details! - or risk the thread being locked or deleted!.

Nigel Goodwin said:

Remember the acronym GIGO (Garbage In, Garbage Out) - in almost every single thread like this one there's never been enough information to give any useful help.

Usually the 'brilliant' idea is nothing of the sort, the entire premise is wrong, and the method of trying to implement it is completely wrong. But as we're all completely in the dark we don't find this out until the thread is many pages long.

This wastes bandwidth, and everyones time - so if you want help you need to provide details! - or risk the thread being locked or deleted!.

Click to expand...

to be fair to the guy, he only wanted help finding ADCs. Of course, his odd requirements triggered the ensuing discussion. The simple fact is that national, microchip, maxim, linear and so on have a ton of resources available.

I do agree that when they try to obscure what they are doing, it usually means that they can't possibly get decent help since no one, including the poster, has any real understanding of therequirements. Pretty frustrating to see a diagram like that which shows the protective inventor clearly needing more than help in selecting an ADC.

But, I do think they should simply be subject to darwinian forces rather than a boot in the posterior.

I had typed up the following reply before philba's post, which I agree with pretty much 100%. I could certainly use more help than choosing an ADC and at this phase I dont have clear requirements...

I am working with what I know and trying to make things come together. My final product will unquestionably be a hack but it will certainly be useful to me and prove my concept. I have built plenty of products from electronics, to software, to automotive and my working prototype is NEVER something even remotely close to something actually marketable.

If I came here asking you to solve the issues of my application and refused to disclose what the application is, I believe then it would be fair to be so critical of me but I did not come here for design help (though I could certainly use it)

Here is my message from yesterday when the boards were down.
--------------------------------------
I am going to get off topic but would like to point out some things.

I am a huge advocate of open forums like this and I use dozens of them on a regular basis... At some I am the 'Newbie' and know little about the forums topic while at others I am the guy with thousands of posts and am quite knowledgeable about the forums topics... Here I am new and admittedly electronics are by no means my forte.

It is amazing to me how attitudes vary across various fourms -- At one end of the spectrum we have forums where every question gets answers no matter who you are and how you go about asking. On the extreme other hand we have the forums where everyone with a post count less 5 billion gets flamed and ridiculed for their questiond and/or ideas. These attacks make it very intimidating for new users to post, which ultimately deafeats the purpose of having open forums.

I cannot comprehend how you expect everyone seeking help to divulge their ideas, ESPECIALLY on a forum where 90% of its users are fully capable of bringing the ideas to life. I have no problem with being told I am being too vague and that more info is needed -- I do however have a problem with my ideas being referred to as 'the next big thing', 'far form brilliant', 'the premise is wrong', and the 'implementation is wrong'... You can argue that you are not talking about my ideas specifically but there is no question that it is inferred. These kind of attitudes are what give engineers a bad name (know-it-alls, etc...)

On the topic of wasting bandwidth and time... Being told repeatedly that more info is needed, especially after I said I am going back to the drawing board, is just as much of a waste of time and bandwidth... Perhaps you are just trying to increase your post count, or perhaps you want to get on 'Mr. 10,000 post Electronics God's goodside by reiterating what he said -- Whatever your reasoning, you've provided no useful help or insight.

You can lock this thread / ban me if you so desire... I'd rather not be part of a forum that treats its new users in such a manner.

Summed up... Treat users as if you are sitting at the table with them in a professional setting making design decisions. Asking critical questions and giving critical feedback is a skill -- the anonimity of these forums should not allow you to forget that.

I've redone my 'explanatory' schematic and I believe it makes much more sense. I will post it an an explanation in a bit if anyone is interested

buurin said:

I cannot comprehend how you expect everyone seeking help to divulge their ideas, ESPECIALLY on a forum where 90% of its users are fully capable of bringing the ideas to life. I have no problem with being told I am being too vague and that more info is needed -- I do however have a problem with my ideas being referred to as 'the next big thing', 'far form brilliant', 'the premise is wrong', and the 'implementation is wrong'... You can argue that you are not talking about my ideas specifically but there is no question that it is inferred.

Click to expand...

I can't comprehend how you expect to get useful answers when you don't ask useful questions?.

As for my references to your 'ideas', obviously as I've no idea what it is I can't comment on it - but what I did comment on was historical threads on these forums. In almost every case where people made claims like your's and refused to give sufficient details, eventually (after threads running to pages and pages) we find out the entire premise was flawed, and they were approaching it the wrong way - completely wasting everyones time and energy.

As you suspect, I'm certainly 'inferring' it in your case - because your post fits the template so exactly!.

Also, another reason for 'inferring' is the extremely basic type of question you ARE asking - with such a low level of electronics knowledge it's highly likely you're approaching the problem in a 'less than optimum' way?.

buurin said:

I cannot comprehend how you expect everyone seeking help to divulge their ideas, ESPECIALLY on a forum where 90% of its users are fully capable of bringing the ideas to life. I have no problem with being told I am being too vague and that more info is needed -- I do however have a problem with my ideas being referred to as 'the next big thing', 'far form brilliant', 'the premise is wrong', and the 'implementation is wrong'... You can argue that you are not talking about my ideas specifically but there is no question that it is inferred. These kind of attitudes are what give engineers a bad name (know-it-alls, etc...)

Click to expand...

Have you ever helped someone to solve a math problem without knowing exactly what the problem is? They are just asking how to do bits and pieces of the problem and you actually have no idea if they are going about the overall problem correctly? It's really hard. There are a lot of people coming on this forum asking all the wrong questions and going about their task in an overly elaborate (and ultimately non-functional) way, so sometimes the premise is just plain wrong and the implementation could be done much better for much simpler. Taking constructive criticism is one thing, but labelling all criticism as non-constructive is not very productive.

Treating others profesionally like everyone is at a design table is important...but that respect also goes both ways- is also not very reasonable to ask a very specific (yet vague) question and then get worked up about them saying that you aren't giving them all they need to know. We can only give critical answers, if you ask critical questions.

All that aside, you seem to not be thick-skulled and are willing to modify your designs. It's sometimes hard to get a message through on a forum so you sometimes you exagerate what you say so the point gets across...and sometimes the person on the other side has no problem receiving the message so it seems overpowering.