Launching a new product - need your help/advice/feeling

[misterT]

Well, i kind of though of a way around this, allowing you to "plug in" all the real components...

Sounds like a breadboard

Designing a "test signal" generator and analyzer would not be a waste of time.. kind of USB -testprobe for analog stuff. I know Ikalogic designs USB logic-analyzers. You can do a nice PC side software etc.

 

[ikalogic]

Sounds like a breadboard

Yes but without the wires, and with integrated power supply, signal generators, and multi-channel scope.

However, an "Enhanced" breadboard was the first seed of the idea, but i downgraded it to an "op amp only" testing board for the sake of simplicity, doability and cost.

 

[misterT]

I do mostly digital designs and the most important tool for me is a logic analyzer. I like my small usb analyzer (not an Ikalogic, sorry). But, I also do lots of interfacing to sensors etc, which require some analog signal conditioning etc. I would buy a pocket size USB "analog analyzer" if it comes with nice PC side software. It should have few signal generator channels and few ADC channels for input. It would not need to be extremely fast.. maybe just the audio-range.. ultrasonics maybe, because I use ultrasonic sensors frequently.

So, maybe you should work on the analyzer side of your product first. Design it so that user can plug the probes on breadboard. You can expand it to "intelligent breadboard" later if you want.. but I only want the analyzer, if the PC software is good. I would pay ~200 dollars.

 

[ikalogic]

@ MisterT

Thanks a lot for sharing that

 

[audioguru]

In my career I designed and constructed many analog and digital circuits and some were very complicated.
I never used a breadboard and that was before simulation software was available. Every prototype worked perfectly and was sold as the final product.
When a circuit was mass produced then my prototype was simply copied.

 

[ikalogic]

Hmmm.. your post kind of proves a point: Undoubtedly, you have an amazing know how in both digital and analog electronics, you helped me (and a lot others) on this forum countless times. Now, that being said, you never used a breadboard, does that make breadboard not useful? (I know you didn't say that either )

Anyway, after all the feedback i got here and there and the results of the survey so far: 50% of the persons asked find this idea very useful, and 50% find it totally useless.

Kind of a risky plan to start designing the HW and Software for such a device, might be a big waste of time. will have to keep this project at the "idea" stage

Damn! i wanted to re-invent the breadboard in a better way! will have to wait for that a little...

 

[audioguru]

Almost every time a student posts a problem on these forums that his circuit doesn't work even when it was connected correctly, it was built on a breadboard.
The high stray capacitance of the many strips of intermittent contacts and long wires all over the place prevented the circuit from working correctly.

Today we have the internet to show circuit designs from IC manufacturers and from ordinary people. The circuits can simply be copied and will work perfectly without testing anything.

Your project to make a product that demonstrates opamp circuits will be useless because opamp circuits are very simple.

 

[ericgibbs]

Almost every time a student posts a problem on these forums that his circuit doesn't work even when it was connected correctly, it was built on a breadboard.
The high stray capacitance of the many strips of intermittent contacts and long wires all over the place prevented the circuit from working correctly.

hi agu,

I think you should specify which type of circuits you are saying are unsuitable for breadboard.

I have used them many, many years for digital and analog circuits with no problems at all.

For any hobbyist on a low budget they can save money by using the same components many times to build a number of learning projects.

E.

The only breadboard testing I have seen on the Forums thats factual, is this one, by JimB

https://www.electro-tech-online.com/threads/a-dog-tracker.278/

 

[audioguru]

I have never seen a student use alternate grounded breadboard tracks as shields.
An ordinary general purpose 2N3904 transistor can oscillate up to 200MHz or more because the minimum fT is 300MHz.
Every audio opamp has plenty of gain for RF oscillations when on a breadboard.
74LS TTL logic and 74HCxxx high speed Cmos logic probably rings like crazy on a breadboard.
The many connecting wires on a breadboard pickup local radio, TV and communications signals as interference.

 

[ericgibbs]

I have never seen a student use alternate grounded breadboard tracks as shields.
An ordinary general purpose 2N3904 transistor can oscillate up to 200MHz or more because the minimum fT is 300MHz.
Every audio opamp has plenty of gain for RF oscillations when on a breadboard.
74LS TTL logic and 74HCxxx high speed Cmos logic probably rings like crazy on a breadboard.
The many connecting wires on a breadboard pickup local radio, TV and communications signals as interference.

You say you have never used a breadboard.?

How can you base those unproven assumptions on a test sample of 'zero'.?

I have run 74F, 74HC, OPA's at 10MHz, PIC's and MCU's at 20MHz. with no problems

 

[ikalogic]

I have never seen a student use alternate grounded breadboard tracks as shields.
An ordinary general purpose 2N3904 transistor can oscillate up to 200MHz or more because the minimum fT is 300MHz.
Every audio opamp has plenty of gain for RF oscillations when on a breadboard.
74LS TTL logic and 74HCxxx high speed Cmos logic probably rings like crazy on a breadboard.
The many connecting wires on a breadboard pickup local radio, TV and communications signals as interference.

If the project i am proposing will make tracks/connexions very short, dense and well shielded, would that make my idea any more viable (for those of use who use breadboard / protoboards..)

 

[ericgibbs]

hi Ibrahim,

All I do is fix the underside face of my breadboards to a copper plated PCB, that is a large piece of un-etched PCB, slightly larger than the breadboard.

Then I connect the copper plated PCB to the 0V rail sockets on the breadboard which effectively gives me a ground plane.

You could do something similar on your PCB layout/PCB.

E

 

[audioguru]

You say you have never used a breadboard.?

I built ONE audio preamp with a TL071 on a breadboard and it oscillated. On a pcb it worked perfectly. I never used a breadboard again.

How can you base those unproven assumptions on a test sample of 'zero'.?

Because so many defective breadboard projects on these forums worked fine on a pcb.

 

[RichTheDude]

I have never seen a student use alternate grounded breadboard tracks as shields.
An ordinary general purpose 2N3904 transistor can oscillate up to 200MHz or more because the minimum fT is 300MHz.
Every audio opamp has plenty of gain for RF oscillations when on a breadboard.
74LS TTL logic and 74HCxxx high speed Cmos logic probably rings like crazy on a breadboard.
The many connecting wires on a breadboard pickup local radio, TV and communications signals as interference.

I am guessing you have never used a breadboard in anger? I have used PICs upto 20 MHz, and generated 6ns pulses with 74ACT logic (the fastest 74 series conventional logic out there). The edges are not as good as PCB's I have designed (when viewed with a 1 GHz oscilloscope) but it worked with no problems.

Breadboards are superb for testing out quick ideas (without the cost and time turnaround of PCBs), and the reusability of parts is great.

I had a guy I was mentoring use a 200 MHz GBP op-amp on a breadboard as a non inverting amp. It took off, and oscillated at 60 ish MHz. Guess what? I showed him how to stabilise it

 

[ericgibbs]

I built ONE audio preamp with a TL071 on a breadboard and it oscillated. On a pcb it worked perfectly. I never used a breadboard again.


Because so many defective breadboard projects on these forums worked fine on a pcb.

This does not mean the breadboards are a problem, most likely you had the TL071 incorrectly wired or inadequately decoupled.??

Most of the breadboard problems I have seen here are due to wrong wiring or no decoupling.

 

[unclejed613]

the only problem i see is that this will be extremely complex. i can see using a few standard configurations and using CMOS analog switches to select component values, but to be able to arbitrarily "build" a circuit sounds more like you would need multiple crosspoint switches in addition to the component value selection switching. at least 5 pairs of crosspoint switches, one pair for each of the following: noninverting input, inverting input, negative feedback path, positive feedback path, and output. the crosspoint switches could be used to select one or more components (resistor, cap, etc) and also select where that component terminates (+V,-V,GND,block input, block output). even with that level of complexity, there still may be certain circuits that just can't be done, that would be much simpler to implement with a soldering iron...

 

[ikalogic]

the only problem i see is that this will be extremely complex. i can see using a few standard configurations and using CMOS analog switches to select component values, but to be able to arbitrarily "build" a circuit sounds more like you would need multiple crosspoint switches in addition to the component value selection switching. at least 5 pairs of crosspoint switches, one pair for each of the following: noninverting input, inverting input, negative feedback path, positive feedback path, and output. the crosspoint switches could be used to select one or more components (resistor, cap, etc) and also select where that component terminates (+V,-V,GND,block input, block output). even with that level of complexity, there still may be certain circuits that just can't be done, that would be much simpler to implement with a soldering iron...

Analog cross point switch matrix... that's EXACTLY what i intend to use. I also came to the same conclusion that it would be very complex. And i wouldn't have came and opened that thread if i didn't have a solution for this . Basically, The user will still need to use a soldering iron to populate the components he needs on a "disposable plug-in board", that will look like a proto-board, more or less. The big advantage, however, is that user will not have to populate the wires: He will draw the circuit on PC, and then the wiring will be done by the device. He can rewire the circuit in 50 different configurations without having to touch the soldering iron.

I know that he will still need to use the iron again to choose different components, but at least, the use:
- won't risk having wiring mistakes (or the risk will be minimal)
- will be able to safely build 50MHz + circuits without having to worry about breadboard capacitance, long wires problems
- will be able to use SMT components (personally, that's one of my big problems, many times i have to solder wires to SOIC leads to be able to use it on my bread board...)
- and all the other advantages stated above...

To be able to make the circuit more affordable (maybe around ~70€) it can be a 14 node circuit (2 of them being VCC and GND). With the same design concept i have in mind, you can have around 8 resistors/capacitors/diode (any 2 leads components).

I took as a reference National Instrument's famous application-note 31, which shows basic op-amp circuits, and there are many circuits here that have far less than 14 nodes.

Anyway, i am defending my idea because it's the logical thing to do! but to be honest, i have my doubts about this project. Not technical, but doubts about whether one would pay 70€ for that. This kind of product has never been made before (correct me if i am wrong). It may be a top seller, and may as well be a big fail...

 

[ericgibbs]

hi ikalogic.

I recommend reading Posts #13 and #14 on this thread, they highlight the uninformed nonsense being posted by some members regarding breadboards.

https://www.electro-tech-online.com/threads/not-200ns-per-cycle-with-20mhz-xtal.134840/

E.

 

[ikalogic]

hi ikalogic.

I recommend reading Posts #13 and #14 on this thread, they highlight the uninformed nonsense being posted by some members regarding breadboards.

Thanks, i am already convinced by the way! From my humble experience, i can confirm your previous post saying that most breadboard problem were due to wiring errors and missing decoupling caps (the least being the the mistake i did the most when i used to work on breadboards )

 

[audioguru]

Why not invent something that is useful instead?