Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

But I'm a Hardware Guy

Status
Not open for further replies.

Mikebits

Well-Known Member
There was a time when you had hardware guys, and software guys (and gals too). I have always been on the hardware side of things, in fact I enjoy that side of the fence. It seems as though now you can no longer just be hardware as so many things ranging from sensors to power management have communication interfaces requiring some amount of software, or firmware if you prefer. So if you want to do any sort of testing these days, you need some amount of programing just to get things done.
I find doing firmware related things as a chore and I really do not like doing it. Perhaps that is why I am not very good at it.
So what's my point. I really don't have one, I think they call this a rant :)
My advice to new would be engineers is, don't pigeon hole yourself like so many of your predecessors have done in the past, or you may find yourself playing catch-up or risk being left behind.
End of rant
 
:) A good rant though MB.

There is the opposite side of the fence- software specialists who hate hardware and do not know a thing about it.

Have you really had a go at software- it is not as uninteresting, or demanding as you may think- although writing assembler can be a bit tricky.

I am basically a hardware person too, but I also find writing/learning about software fascinating, especially to realise a hardware function, say with a microcontroller chip.

Single-chip, low-cost microcontrollers have revolutionized electronics since they first came out, but in the last ten years, or so, this has accelerated with AVR and PIC microcontroller chips and the Arduino microcontroller unit (MCU), and Raspberry PI single board computer (SBC), plus simple, free, integrated development environments (IDEs) and high level programing languages.

I hate to say this, but I suspect the future trend will be automatic hardware generation. We have seen the start of this with programmable gate arrays (PGAs).

What I suspect will happen is that you will generate a specification for a device, then send a defining file to a manufacturing house and the whole unit, connectors and all, will be automatically manufactured, in a similar way that printed circuit boards (PCBs) are manufactured now.

In the case of small circuit, say a typical 555 pulse width modulation circuit (PWM), that will all be done in one silicon chip, although the Rs and Cs will not be used- the whole thing will probably be digital.

Just some thoughts.:)

spec
 
Last edited:
Well, I have taken a college course in C/C++, and also in LabView. Much of my testing platforms are on a Cypress PSoC development board, so I can write software, I just don't like doing it. The programmable hardware devices were once a thing for the digital domain, but now we are seeing more programmable analog devices, the Cypress PSoC for example. A lesser known device but really nifty part touted as an FPAA, is from a outfit called Anadigm.
So I suspect you are right in your prediction about the future of hardware design.
 
In the case of small circuit, say a typical 555 pulse width modulation circuit (PWM), that will all be done in one silicon chip, although the Rs and Cs will not be used- the whole thing will probably be digital.
I needed a 555 for a circuit.... It was quicker and more efficient for me to use a pic12f1840... I could get the output timing spot on using compare... The device was placed rotated where the 555 was and worked better!!! Only 1 pin needed attention.
 
I needed a 555 for a circuit.... It was quicker and more efficient for me to use a pic12f1840... I could get the output timing spot on using compare... The device was placed rotated where the 555 was and worked better!!! Only 1 pin needed attention.
Yes, it is often the case that a microcontroller chip will perform better than a collection of components. The trouble is that most people in hobby electrons do not want to go that route. But if you design/recommend a circuit with a 555 your like votes go up.

Of course some members, due to their location, can't easily get a microcontroller and haven't even got a broadband internet connection (essential to do microcontroller development).

For a while now I have wondered if there couldn't be a standard ETO microcontroller, that ran a program that could make the microcontroller perform certain functions. For example, there could be a timer sub program. pulse width modulation (PWM) sub program, etc.

spec
 
Last edited:
But if you design/recommend a circuit with a 555 your like votes go up.
I don't care too much for "Social media like system"... Not here to be "liked" so to speak, I'm here to help!!
 
Of course some members, due to their location, can't easily get a microcontroller and haven't even got a broadband internet connection (essential to do microcontroller development).

Not strictly true, and certainly not in the past - as micro-controllers long pre-date broadband connections :D

As far as the hardware/software argument goes, my whole reason for PIC's was to replace complex hardware with a PIC and a bit of source code, and this was back in the DOS only days of PIC programming - and as I mentioned before, I wrote the worlds first Windows PIC Programmer software, long before MicroChip :p

Certainly now (due to the huge sizes of downloads) you need broadband, or at least access to it via a third party, so they can download the required files for you. Once you've got the files (MPLABX etc.) you can happily work with no Internet connection at all. I've commonly done my downloads at work, and taken the files home on CDROM or USB drives.

As for availability of micro-controllers?, as long as you have a postal service then pretty well anywhere can get them - I suspect most places don't have the option of 'popping in the corner shop' to buy them :D
 
Hi MB nice rant. there is a third player here , the field engineer ... The guy / gal that visited sites to sort out the other twos mistakes... some customers got quite angry I remember having to go 'Mod' boards with fine wire and the like to overcome errors , even had to piggy back chips ... happy days...
( don't take it badly ) ,
 
...
For a while now I have wondered if there couldn't be a standard ETO microcontroller, that ran a program that could make the microcontroller perform certain functions. For example, there could be a timer sub program. pulse width modulation (PWM) sub program, etc.

spec

It is called an "Arduino"!
 
At my last job there was a noticeable separation between me a hardware guy and our software guys regarding doing service work on various systems.

The software guys would burn up a day or more trying to write a perfect software fix for a buggy input signal trying to find a solution to filtering out unwanted false tripping events due to inherent line noise issues without then also code filtering out too much of the wanted signal out. (magnetic rotary pulse encoders on the end of 200+ foot cables run next to who knows what high powered and improperly grounded electrical and mechanical gear were a common issue. )

As a hardware guy I would just spend 10 minutes and add a pair of high speed diodes and .001 - .1 uf capacitors across the power supply rails and the problem input and kill the noise before it hit the PLC input while leaving the base single fully intact. :rolleyes:
 
It is called an "Arduino"!
Afraid not! Unless you know of an Arduino with the software suite that I described.

Perhaps I didn't explain too clearly. I meant a microcontroller with a pre-loaded software suite that could simulate a few of the basic hardware chips like the 555.

No programing would be required, instead the pre-programmed microcontroller would be configured by hardware links to perform a certain function.

spec
 
Last edited:
Afraid not! Unless you know of an Arduino with the software suite that I described.

Perhaps I didn't explain too clearly. I meant a microcontroller with a pre-loaded software suite that could simulate a few of the basic hardware chips like the 555.

No programing would be required, instead the pre-programmed microcontroller would be configured by hardware links to perform a certain function.

There have been a number of pre-programmed PIC based devices sold, replicating various chips such as the 555 - but each device only replicated one specific chip.

I presume such devices failed?, as it's an incredibly small market you're aiming at - it makes far more sense to use them as intended, and program them to do what you want.
 
There have been a number of pre-programmed PIC based devices sold, replicating various chips such as the 555 - but each device only replicated one specific chip.

I presume such devices failed?, as it's an incredibly small market you're aiming at - it makes far more sense to use them as intended, and program them to do what you want.
I agree with everything you say about microcontrollers/microcontroller boards/microcomputer boards from a technical point of view. In fact many, ETO responses could just be, 'Use an Ardunio': timers, PWM, LED chasers, you name it.

But, many members, especially new members, shy away from any contact whatsoever with a processor and associated IDE and specifically the dreaded programming. Also many members want the experience of hooking up a few chips and building a traditional discrete circuit- they often say this.

Also, some members do not have the facilities/funds/access that members in the developed countries have. One member for example has no (update) mains electricity. In some areas post can take 60 days, if it arrives at all.

But, what I am suggesting could become a standard item that has most of the advantages of a processor approach without needing any programming or other support.

This would not be a commercial exercise, but you never know it may take off (in a small way).

spec
 
Last edited:
Also, some members do not have the facilities/funds/access that members in the developed countries have. One member for example has no electricity. In some areas post can take 60 days, if it arrives at all.

But, what I am suggesting could become a standard item that has most of the advantages of a processor approach without needing any programming or other support.

That seems rather contradictory :D

If they can't get mass produced worldwide products from major manufacturers, they certainly won't be able to get obscure 'one off' specially made pre-programmed processors.
 
That seems rather contradictory :D

If they can't get mass produced worldwide products from major manufacturers, they certainly won't be able to get obscure 'one off' specially made pre-programmed processors.

I give up:banghead:

spec
 
Perhaps I didn't explain too clearly. I meant a microcontroller with a pre-loaded software suite that could simulate a few of the basic hardware chips like the 555.
Liking the concept , can I add a switch pack , one chip could do 256 functions...
custompic.jpg
 
Ah, an objective response.:)

That is a great idea- preset switch rather than wire links like I said. Wire links could also be used though. It would depend on the application.

The principle being that all you would need is the pre programmed microcontoller so that would be only one inventory item for people with supply and cost problems- they just buy one batch of reprogrammed chips. This would also be an advantage for other electronics hobbyists too.

You never know there might be a whole sub culture using the ETO multifunction chip (ETOMHC). :)

I have a lot of building work to do at the moment, but while mixing concrete and laying bricks I have been thinking about a simulation of a 555, mainly the method of inputing the mode (bistable tmonostable) etc and the time value, say 1uS to 24 Hours. So far I cant see a problem.

I think the address of that simulation should be 555 (in binary) on your switch.:cool:

If you or other members have any suggestions for other circuit functions perhaps we could arrive at a good list.

spec
 
As an a professional electronic engineer I've really struggled with this conundrum for a few years now, as I assess my career path. In larger companies, their teams are still separated; HW, FW, and then SW. In smaller companies they generally seem to like people who can do it all.

I've been a hardware and firmware engineer for a while now, going on 10years. I do the schematic designs, route the pcb's, taken devices through all their approvals, do tons of firmware, write the PC test apps, I've written embedded web servers etc etc. I enjoy the firmware side of things more than hardware, but in all honestly I feel inadequate with both hardware and firmware, although feel more competent with firmware. I feel I am not an expert in either, more a "jack of all trades" engineer, and see many shortcomings in my skills which I actively work on.

No one's complained though, never had a problem with my employers, but all I'm saying is it would be nice to feel like an expert in something...if that makes sense? Hence I feel the need to try to push towards some speciality at this point, be it hardware or firmware.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top