Yep a lot of things have changed and I'm just wondering if it's my age or something but I've seen a few topics with people seeming to turn to an arduino first without it appears thinking of a discrete component solution.

It's kind of intriguing.

 

Often a small μp will do the job that would normally require a handful of IC's and R/C's , and for a $1 or 2, much cheaper.
I look at a microchip for alternatives if possible.
Max.

 

Yep a lot of things have changed and I'm just wondering if it's my age or something but I've seen a few topics with people seeming to turn to an arduino first without it appears thinking of a discrete component solution.

It's kind of intriguing.

Part of it is probably our age.
I think a lot of times people overlook the one time costs of a micro and the learning curve when it comes to hobby circuits.

 

I suppose it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail.
- Abraham Maslow, The Psychology of Science: A Reconnaissance (1966), Ch. 2, p. 15.

ak

 

Yep a lot of things have changed and I'm just wondering if it's my age or something but I've seen a few topics with people seeming to turn to an arduino first without it appears thinking of a discrete component solution.

It's kind of intriguing.

The cool thing about an Arduino is that some versions have protyping space and or header pins that allow you to add your glue-logic without investing in an additional board (whether you make it your self or not). Additinoally, many "shields" are available to do common things like motor drivers, sensors, ... for the standard Arduino board layout (horseshoe of header pins). The $8 to $50 cost of an Arduino and USB cable to program can be much cheaper and easier than any discrete solution if you are making a one-time solution and want to avoid paying for a custom board design and fabrication (or if you value your time for a DIY solution).

Granted, the one-time learning curve of programming the Arduino is also an investment but any programming experience and some time in front of a tutorial written for a 12-year-old can get you to a solution pretty quickly.

It is all a combination of one-time setup costs vs. number of units you plan to produce and how you value your time.

 

If the distinction is the jump to mcu based solutions instead of discreet logic or analog circuits, I'd have to mirror MaxHeadRoom. At work I constantly run into "small" projects for our test group, or even the marketing team that could be solved with a simple mixed circuit (think: a 555 and some passives). However, by the time I spec all the parts and pack the PCB, a small 8-bit PIC controller on a similar setup would have been A. Cheaper B. Faster to release (especially if it is a controller I already have code for), C. (and possibly most important) re configurable with the click of a button.

I have, in the last few years, put some purely discreet products out to market, and I have done this intentionally for reliability/durability. It's nice to have circuits that don't need a stable power supply and can't get stuck in an endless loop (maybe this is just another age thing). However, to be fair to the argument, the savings in bill of materials and board space probably would have justified a controller.

Now as for an arduino debate: I have never used one and don't see that changing... just not my cup of tea.

 

Not sure if age is a factor, I probably number among the oldest members on the forum.
Max.

 

I think age is a factor. If I were 20 instead of 64 I probably would be trying to use discrete components, and probably pulling them off old circuit boards to save 10 cents here and there.

Once you have invested in learning how to use a micro and have the tools for it, it is generally better than even a circuit that only requires a single 555. Why?

1. Fewer components.
2. I can often test it on a prototype board without building any hardware.
3. Easier to debug.
4. I can change the behaviour in minutes without any rewiring.
5. With a crystal, I can get precision I cannot get with discrete components without a lot of extra complexity.

And proably more reasons that I can't think of right now because at my age, I am losing my memory.

Bob

 

A recent example was a circuit I recently put together for someone's project, originally I thought of a 556, which did work with ~11 components, I then thought of using a 8 pin Pic, and it ended up with 4 components.
And with the pic timers, just as accurate as the RC timing of the 556.
When I started out, if you wanted a flip flop or some kind of timing circuit it had to be done with discrete transistors and other components as this pre-dated the age of TTL/CMOS ic's.
But as soon as they arrived, I dropped the 'hard' way of doing it.
Max.

 

Funny this comes up now. I have a shoe-in one-shot multivibrator circuit a coworker needs for a testbench. I was all but in the bag of 555's when I decided he will probably want to change the absolute ON time as some point down the road. It'll go on an 8-bit PIC later this afternoon...

 

It depends on the job. I whipped together a simple float charger in about 30 minutes with 7 parts I already had.
If I wanted it to do anything except one job, a small PIC would quickly be more versatile and use fewer parts.
I think the question is about why some people grab an MCU before thinking about how to do it in analog...analog design takes a lot of education. You can get involved with a lot of solutions quicker with an MCU than you can by trying to learn analog design.

 

Part of it is probably our age.
I think a lot of times people overlook the one time costs of a micro and the learning curve when it comes to hobby circuits.

I'm not against mrcos, I've been playing with PICs for the last 18 odd years, I guess now that micros ( and tools needed )are so much more cheaper and available than years ago.

 

I'm not against micros.
I guess now that micros ( and tools needed )are so much more cheaper and available than years ago.

I'm against them.

They are cheaper and more available to us nerds, but when a manufacturer uses a microprocessor to turn on the dome light in my car, and it won't shut off, the microprocessor is NOT cheaper and more available than a door switch. When some programmer in China decided that the minimum tire to agitate my clothes in the washer is 20 minutes, changing that is NOT more available than a wind-up timer which I can set to any amount of wash time I want. When my cell phone checks to be sure it is connected only to the charger that came with it, that is NOT more convenient than plugging it into any 5V wall wart I have laying around.

I am against using microprocessors to over-complicate simple things like turning on a light bulb, 6 feet away, then charging me $100 for what amounts to a fancy $3 door switch.
I am against using microprocessors to make sure appliances are not adjustable by the owner, and can't be made adjustable at any price.
I am against using microprocessors to be sure I can only buy the original Brand Name parts for a vastly inflated price.

Microprocessors are NOT cheap and available when I need to pay over $100 each to repair my possessions.

 

I'm with you, I was just talking from a hobbyist point of view.

And there does appear to be many things that could be done without a uc that are.

I'm not a big fan of "driver aids" that are thrown at motor vehicles, there should be more onus on the driver to actually pay attention, play "what if" whilst driving.

 

I'm against them.

.

It seems you are against crappy engineering using controllers.

 

It seems you are against crappy engineering using controllers.

Even when the engineering is good, I am against pretending microprocessors are, "cheaper and/or more accessible", when the inevitable result of installing them is to make repairs economically unfeasible or simply impossible.

A $500 controller on a $300,000 production line? Sure. A $100 controller to turn on the dome light in my car? I don't think so.

 

Yep a lot of things have changed and I'm just wondering if it's my age or something but I've seen a few topics with people seeming to turn to an arduino first without it appears thinking of a discrete component solution. It's kind of intriguing.

I think in most of these cases-- and there are a lot of them, not just a few-- the person simply doesn't know how to design a discrete component solution, whether analog or digital; they may know programming (to some degree), but anything electronic is utterly beyond them. So they immediately resort to an Arduino.

But for anyone with microcontroller experience as well as circuit design experience, use of a microcontroller in any given design is just a matter of whether it makes sense or not. Sometimes it does; sometimes, perhaps more often, it doesn't.

 

IMO It is impossible to generalize.
If it were not for micro's we would not have miniaturization.
I imagine no one would appreciate using a cell phone the size of a small shoe box anymore, which was once the case.
Max.

 

Almost all the electronics and electrical work I do is with analog operations where brute force and less than precise tolerances are in play. uC's are just to fiddly and fussy about that sort of applications.

The only major smart system I have at home is the digital boiler controller I built years ago and even then given the application for me putting in a basic PLR/PLC unit was a way more practical and robustly functional than my wasting the time building an equivalent system around some uC device.

My brother however went the uC route for his boiler controls (Because he didn't want to pay me $200 - $300 to build him a reliable and robust plug and play PLR based controller) and its a fiddly fussy POS to setup and work with that has way too many hours tied up into it for what it can do and do poorly.

 

Funny this comes up now. I have a shoe-in one-shot multivibrator circuit a coworker needs for a testbench. I was all but in the bag of 555's when I decided he will probably want to change the absolute ON time as some point down the road. It'll go on an 8-bit PIC later this afternoon...

Why use a thumbtack when you can use a sledge hammer and a railroad spike!