Advertisement

Blog

Choosing Analog ICs

We welcome back guest blogger James Bryant of Analog Devices this week.

You've told us that we need not be too careful when choosing discrete transistors. What about analog ICs?

This is done with a glass slipper. Have your footmen take the slipper to every manufacturer in the kingdom, and hold it by the door. If

is reflected in the glass, you'll be certain of finding your heart's desire within.

More seriously, overspecifying — or calling for arcane and generally unknown devices — can complicate the selection process and increase the cost of ICs, as well as discrete transistors, so you should never demand hard-to-find parts or unnecessary parameters. But ICs are much more complicated than single transistors, making it very important to define all the needed parameters.

This RAQ will talk about general principles, not particular types of ICs, such as amplifiers, references, or converters. The parameters common to many precision analog ICs are absolute maximum ratings, ESD ratings, supply voltages, supply currents, power dissipation, temperature range, temperature coefficients, supply rejection, noise, packages, input impedance, bias current, analog output characteristics, frequency range, and digital interface specifications such as speed, logic type, logic levels, data configuration, and chip enable/shutdown functions. Before working on the parameters for a specific type of device, it is important to decide which of these general parameters are important and to set acceptable limits to their values.

Starting with a list of important parameters, including the device-specific ones that I'll discuss later, arrange them in order of importance. Next, using online parametric search engines from several manufacturers, compile a list of devices that meet all the requirements. Since this requires visits to several websites, and manufacturers will present the data in different formats, it might be nice to use a distributor's website, which should display devices from different manufacturers in a standard format. Unfortunately, parametric comparisons from distributors are rarely adequate. They are improving, but for now it's generally better to use the manufacturers' sites and to compare devices on paper or in a spreadsheet.

If no suitable device can be found, decide which parameters might be relaxed, and by how much, and try again with the new values. Then review all of the device parameters, in case a feature makes one device better or worse for the application.

Finally, list all the devices that will do the job well, and determine which is the least expensive. The cost of using an IC includes the cost of the device itself, and also the cost of any additional components or special power supplies, the cost of any adjustments or calibrations that must be made during manufacture, and the cost of the printed circuit board area that it occupies. Be sure to choose the device that is the least expensive to use, not the cheapest one.

— James Bryant has been a European applications manager with Analog Devices since 1982. He holds a degree in physics and philosophy from the University of Leeds. He is also C.Eng., Eur. Eng., MIEE, and an FBIS. In addition to his passion for engineering, he is a radio ham and holds the call sign G4CLF.

Other articles like this Visit EDN and see How to choose analog ICs, Part 1

23 comments on “Choosing Analog ICs

  1. fasmicro
    November 5, 2014

    I always smile whenever we write about analog ICs and analog devices which one company was creative in its name. I know some differentiate with analogue ICs and analogue devices just to make Analog Devices not to take all the shine. Sure, there are many elements in this game. Choosing an analogue IC requires experience which explains why growing in that sector, career wise, takes years.

  2. goafrit2
    November 5, 2014

    >> Finally, list all the devices that will do the job well, and determine which is the least expensive. 

    I always consider if this company will be in business say in 4-6 years. Most analog startups collapse quickly. You need to model that while choosing parts

  3. Davidled
    November 6, 2014

    Success of analog business might be depending on analog sales engineers who present and decorate the product to customer and sometimes, review their requirement, change and update product appearance in order to meet client's need.

  4. uchiha
    November 6, 2014

    @Daej: Indeed you do need to keep a good relationship with the customer to keep you in the market actively. Also its vital to do research and gain feedback of the product so the improvements can be done based on that. 

  5. goafrit2
    November 10, 2014

    >> Success of analog business might be depending on analog sales engineers who present and decorate the product to customer and sometimes,

    A salesman cannot change a bad product. Success should depend on the designer! Period. If you do not have a great product, a clever salesman will become a stammerer.

  6. goafrit2
    November 10, 2014

    >> Also its vital to do research and gain feedback of the product so the improvements can be done based on that.

    That is important in any business. You need to understand the needs, expectations and perceptions of the customer and how that is changing and evolving. Anything less you can be making products for the landfills.

  7. ue2014
    November 11, 2014

    @goafrit2 – >>>>That is important in any business. You need to understand the needs, expectations and perceptions of the customer and how that is changing and evolving.

    Totally agree with you. Research and development is a key element to any business in modern markets. With that one thing which should be really considered is 'When”. 

    Many organizations conduct researches once there is a problem for their product to see what went wrong. But the correct approch or effective usage of R&D for an organization will work as a supportive proactive approch for the organization to identify the changes in Market and Consumer behaviors anf thinking patterns, so they could adjust them selves also to change without being affected. 

  8. ue2014
    November 11, 2014

    @goafrit2 – >>>>>>>>>>>>  A salesman cannot change a bad product. Success should depend on the designer!

    I also agree with you. Without a good product, a Salesman will have nothing to market. So if any organiation would like to build up a suceess story, they need to have a good product. 

    Also on the other hand, the role of Salesman plays in an organization could not be ignored too. They are the main force of the organization who creates the awareness and profits for the organization. They make your good product availabe to the customers and build a customer base. They also has the ability either to build or ruin your product no matter how good your product is. 

  9. uchiha
    November 13, 2014

    @ue2014: Do you mean to say that sales team or an individual will go towards the extent of lying to the customers ? I don't think it will work because if so nothing will be true when it comes for marketing  

  10. ue2014
    November 13, 2014

    @uchiha – >>>>>>>>>>>>>>>>>>>> Do you mean to say that sales team or an individual will go towards the extent of lying to the customers ?

    I was not refering to any senarios like this nor eighter I have mentioned / meant anything on same kind. Good Product and Good Sales team both are important to an organization. 

    Good Product is required to Market. Sales Team should be there to do the marketing. Even bad products are being sold in the market everyday. But I do not believe its Ethical Marketing. That's why a Good Product & Good Sales Team both are important. 

  11. PCR
    November 18, 2014

    Exactly uchiha that is right for the all businesses, satisfied customers will never allow business to go wrong they will always direct o the success.

  12. PCR
    November 18, 2014

    True ue2014, its worth spending time and money in the stage of product development, rather than spending onto find out what went wrong.
    Simply it's better to be proactive rather than being reactive.

  13. ue2014
    November 19, 2014

    @ Ranasinghe – >>>>>>>>>>>>>>>>> True ue2014, its worth spending time and money in the stage of product development, rather than spending onto find out what went wrong. Simply it's better to be proactive rather than being reactive. >>>>>>>

    Agree with you. There is two sides of the story. Spending time in the Product Development Stage would minimize the errors on product but will not be able to cut it 100%. Therefore, there has to be amendments / adjustments made after product being developed anyway at certain percentage.

    On the otherhand, once the product is in the market, it is important to know that whether product serves the exact purpose it was introduced and how well market react to it. Then again, Research is important. 

  14. uchiha
    November 21, 2014

    @ranasinghe: Yes it's a good point. Loyal customers will not let their brand name to get tarnished. It's something that gets created with time itself. 

  15. PCR
    November 24, 2014

    Yes ue2014, Research is important for sure to identify the future business  thou it is a super brand, otherwise they will be wipe from the market with no time. 

  16. goafrit2
    December 4, 2014

    >> Totally agree with you. Research and development is a key element to any business in modern markets. With that one thing which should be really considered is 'When”. 

    It is also important to know that research does not solve all your problems. Apple does not waste time on it. There are many things no customer can understand because it has not imagined it. You just have to see that product to like it. A research may not make sense. I am not sure anyone will say YES if someone told him/her that leasing his/her room to a stranger for small change a night could work. But Airbnb is worth billions of dollars doing so!

  17. goafrit2
    December 4, 2014

    >> They also has the ability either to build or ruin your product no matter how good your product is. 

    It will be very bad if your salesman is ruining your product. I see no reason of having him or her in the system to start with.

  18. goafrit2
    December 4, 2014

    >> On the otherhand, once the product is in the market, it is important to know that whether product serves the exact purpose it was introduced and how well market react to it.

    You can easily measure that – just look at the sales numbers. If the product is selling, it does not matter the purpose. When they made Viagra before it had a name, the real “purpose” was not even the reason customers were buying the stuff.  So, the purpose while important is not as vital that people are buying the stuff. 

  19. ue2014
    December 4, 2014

    >>>>>>>  I see no reason of having him or her in the system to start with. .>>>>>>>

    Exactly. You need to be very contious about the people you hire, specially for the sales team as they would be the mirror of your organization. If any salesman is not delivering the expected results and does not adjust him/her self to match the brand Image, the organization must be vigilant to take corrective nessasary action before its too late. 

  20. ue2014
    December 4, 2014

    >>>>> When they made Viagra before it had a name, the real “purpose” was not even the reason customers were buying the stuff.  So, the purpose while important is not as vital that people are buying the stuff. >>>

    Could be the situation some times. But atleast as an organization needs to understand why people buy their product, even it might not be the original purpose it was supposed to serve. That will definietly help the organization in future developments and marketing activities for the product. 

  21. fasmicro
    December 6, 2014

    >>  If any salesman is not delivering the expected results and does not adjust him/her self to match the brand Image

    I do not expect salesmen to do magic. When a company has poor products, do not expect them to be magical in their results. How could salesmen have saved the old Compaq computers with the quality of their products?

  22. fasmicro
    December 6, 2014

    >> But atleast as an organization needs to understand why people buy their product, even it might not be the original purpose it was supposed to serve. 

    Of course the makets of Viagra later understoond. The point is that do not say if it is not that reason, it is a bad thing. As they say in politics, the poll that matters is the one on election day when people vote, it applies here. You may have thought something else but when you see the reason customers are lining up to buy, change and promote the product accordingly.

  23. ue2014
    December 7, 2014

    @fasmicro – >>>>>> I do not expect salesmen to do magic. When a company has poor products, do not expect them to be magical in their results. >>>>>>

    100% agree with you. If your product is bad, you can not expect the salesman to do magic. Even if they do, that would be uneithical. Through my comment of ' If any salesman is not delivering the expected results and does not adjust him/her self to match the brand Image.' I was refering to the otherside of the senario. 

    If you do have a quality product which has Brand Recognition and Value, you should not allow your Sales Team to damage it. 

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.