Welcome to AAC.

Why do you want to do this? What benefits are you expecting from it?

You say you are an electrical engineer, but what are your actual qualifications (e.g. education, experience)?

 

The TS was last seen in Aug 11.

 

The TS was last seen in Aug 11.

Well, I guess I probably shouldn‘t wait for an answer then!

 

The very first step in creating a new design is to decide what it will do. All efforts prior to that are just general preparation.

 

Looks like the OP left the building ! Some time ago!

 

Unfortunately…. This is a common occurrence from newbie’s that don’t receive the simple answer they were expecting.

Sometimes simple like:
“At Walmart, in the hardware aisle, find the pre-packaged wafers. Make sure that you select either an analog, digital or mixed mode wafer.
At home, preheat your oven at 450F for 5 minutes, and place the wafers inside. Cook them for 45 minutes or until golden brown. Let them sit for 10 minutes before touching them. Slice them with a sharp knife, and enjoy with a glass of milk.
Warning: wafers are not microwave safe.”

 

The fact is that the successful companies do not complain about all the effort that is required, and so when somebody sees a video on the cartoon channel about somebody producing an IC, some folks actually believe it. I suspect that is what happened in this case. AND then they may have got a really good deal on some obsolete production machines .

 

Begin with the fabricator choice what determins a technical demands to be observed. For example, in Netherlands are few powerful firms specialized in such services, and in France is one. Often if the serie is at least sub-million pieces, then funding for such is rather competetive being slight less than million USD. Thus, the funding is first thing to start with. Probably, if the series are smaller and technology may be enough with diffusion, may try the Riga plant RD-Alfa. Their technological level is far below the French, but the salaries are some 10x lowwer thus the expenses are lower as well. (www.rdalfa.eu/)

 

For analog most ICs are designed by using building blocks that you do not get to make changes to. TI has a big list of analog blocks like a PGA103 amplifier building block. It just works and you wire it up in the compiler. Pin52 connects to the output.
The compilers I used does not care if you know what a transistor is or not.

Ron;
Do you have a website link for those?
I searched the TI website and replied with the PGA103 programable gain amp IC datasheet and app notes.

 

STILL, a very important portion of a design is the purpose, at least a partly intended purpose. At least in the design of industrial systems. It may be that when designing consumer targeted systems an exact description of what it should provide is more from the marketing and advertising areas, rather than actual product definition people.

 

If I want to design a chip myself, where to start with? I am a electrical engineer.

Hi,

I do not know if anyone mentioned this yet, but if you can get away with it you can use an ASIC or similar and just write the required programming and have multiple units programmed by the manufacturer. That's a lot cheaper. There are other things like that too such as Microcontrollers and even ROM memory can be used to some extent to perform regular functions.
You might know all this already, but thought I would mention it.

 

This thread should have been allowed to die months ago but for completeness, semiconductor fabs cost from millions to billions of dollars. Rather than buy a fab, most companies make custom integrated circuits by designing it themselves (sometimes with the help of consultants for the really tough parts). An engineer of mine designed an IC using programmable gate arrays and when it worked, he handed the design off to Orbit Semiconductor.

We had to pay for a consultant arranged by Orbit to design the phase-locked loop part of the clock circuit.

By the way, and this might date the design, but the whole design was performed and tested using a 100 MHz i386 computer.

 

I am guessing that the TS saw good prices for some obsolete equipment and thinks, or thought, that just the equipment was all it took.
A lack of adequate understanding, coupled with wishful thinking, often is enough to start the unknowing headed headlong down the "primrose path."

 

check the tinytapout website

 

Whenever I have a requirement for some unusual IC function I first investigate the "Special Functions" listing from the major IC producers. That has always worked so far. And for one of our rather successful products, a student engineer hired by our management suggested combining several standard IC functions into one package to reduce the IC cost of the product. That would have both increased the cost and reduced the reliability provided by the system redundant arrangement.

 

My first turn to now for unusual IC functions or unique requirement is a Picmicro.

 

Evidently I have not needed any functions that were not available from standard products.
Of course, the requirements that all components be available for repairs has a bit of influence on the selections. And ALL of our products are "worth repairing."

 

Whenever I have a requirement for some unusual IC function I first investigate the "Special Functions" listing from the major IC producers. That has always worked so far. And for one of our rather successful products, a student engineer hired by our management suggested combining several standard IC functions into one package to reduce the IC cost of the product. That would have both increased the cost and reduced the reliability provided by the system redundant arrangement.

Out of interest, how did you calculate the change in reliability ?

 

Out of interest, how did you calculate the change in reliability ?

In general, a first time "spin" of a device designed by a beginner is less likely to be reliable than devices produced over many generations by organizations that have a record of reliable products. In addition, I considered that particular individual's skill level. Which was "Fair."

 

In general, a first time "spin" of a device designed by a beginner is less likely to be reliable than devices produced over many generations by organizations that have a record of reliable products. In addition, I considered that particular individual's skill level. Which was "Fair."

Do subjective and experience are as good as most mathamatical methods .
Thank you