Reply by Harry D April 22, 20182018-04-22

"aioe usenet"  wrote in message news:pbi44n$j5v$1@gioia.aioe.org... 


I had been playing around with a similar idea but stopped when I saw 
https://www.ebay.ie/itm/112168149399
locky_z "my intelligent curve tracer" for 150$ PCB and 220$ complete 
with box. Both include software which seems to be superb:
<https://forums.parallax.com/discussion/137648/my-intelligent-curve-tracer>

Regards
Werner Dahn

Yep, every good lab needs a curve tracer. Have not had one for 20 years.

Ciao,  Harry D.


Reply by aioe usenet April 22, 20182018-04-22
On 20/04/18 05:29, Harry D wrote:
> > > "Tom Del Rosso"&nbsp; wrote in message news:pbb3pj$ppv$1@dont-email.me... > > Jeff Liebermann wrote: >> On Thu, 19 Apr 2018 12:12:30 -0700, "Harry D" <harryd@tdsystems.org> >> wrote: >> >>> Having spent over 50 years in electronic labs, designing and >>> testing, I find myself trying to test zener diodes, FETs, LEDs, >>> transistors and many other semiconductor devices. (...) >> >> Perhaps you should do some market research before taking the plunge. >> There are quite a few products being sold on eBay based on the MK-328 >> or "AVR Transistor Tester" project: >> <https://www.ebay.com/sch/i.html?_nkw=mk328> >> <https://www.google.com/search?q=mk+328> >> Manual: >> <https://elecfreaks.com/estore/download/EF06128-LCR-1602tester.pdf> >> I believe that this is the original project site: >> <https://www.mikrocontroller.net/articles/AVR_Transistortester> >> It will test all the devices that you mention. > > Video of the $7 version from ebay: > > https://www.youtube.com/watch?v=7Br3L1B80ow > > Hard to compete with that.&nbsp; A curve tracer function would be a nice > addition though. > > It made me wonder, doesn't anyone make a ZIF socket with fewer than 14 > pins? > > Hi Tom, > Agreed, but I would not have one of those testers in my lab. As far as > LCR, there are so many testers that can do a better job. All except ESR > to 1.00 m Ohm. That may be my next unit. > When a MOSFET is blown, I run for my Semi-Analyzer and measure four FET > parameters to 1 %. I have never encountered a bad FET passing this test. > Another good use is deciding&nbsp; which diode dropped on the floor. A quick > test comparing their forward voltage (to 1.0 m volt) will easily cull > the devices. > Thanks for your interest,&nbsp; Harry D
I had been playing around with a similar idea but stopped when I saw https://www.ebay.ie/itm/112168149399 locky_z "my intelligent curve tracer" for 150$ PCB and 220$ complete with box. Both include software which seems to be superb: <https://forums.parallax.com/discussion/137648/my-intelligent-curve-tracer> Regards Werner Dahn
Reply by April 22, 20182018-04-22
For Zeners I use a scope. Yup. 

My hand can pick up over 50 volts AC, so I put the Zener on the probe, tip to ground and touch the tip. Voila, a waveform of the 60 Hz in the amplitude of the Zener rating, if it is good. Of course it only works on the lower wattage ones, the bigger ones usually have too much leakage, but those can usually be tested in circuit live anyway. Most higher wattage Zeners are relegated to clamping duty now anyway, with most regulators being active and just using them for a reference. (and then we have Tektronix which used to use one Zener for a half dozen sources or more, what were they made of gold or something ?)

I also used to use a scope for faster testing of lytics. Once you learn how to read it it is the best way hands down, but you have to get used to it. You need to feed a 400 mV, 1 KHz square wave with a source impedance of about 360 ohms and let the DUT shunt it. I have used other test signals but settled on this due to experience, it is the most versatile and accurate. And you set the scope to DC, using a positive or negative going square wave and it will also tell you if the cap is shorted. I designed a small handheld unit with a simple, yet more effective, bar graph LED display that worked on a similar principle but at the time I was working and had no time for such things. The scope method was there and worked, so I used it for decades. The difference is many people don't use the scope all the time, but I did. I literally had a scope before a decent meter. I learned to really use it and it helped alot. I was fast. (it was on a scopemobile along with an isolation transformer and a few other things)

I used to work on 3 benches at once, I would troubleshoot the 3 units and by the time I got to the last one they still weren't done with the first one. Sometimes I fixed the last one and took another one, sometimes I went out for a smoke break. The owner of the company ran the road at times and would pull up in the truck, in the beginning he asked "Do I pay you to smoke ?". I just said "Yes". Later, I replied "Handsomely". And then they used to pay me (and a very few others) to date every Friday so in the afternoon they would ask me how late I'm staying, and then write the check and give it to me. So he pulls up and asks the usual question and I said "Yes, in fact you already have". 

The owners got sick of Ohio and moved to Florida and opened up shop. I could have went but I have no connections down there and the whole place smells like a dirty fish tanks. So much for that. I had another job anyway, about a third of the driving. I  got sick of getting home an hour after quitting time, with this place usually the heater in the car wouldn't even get warm by the time I made the trip, including stopping for beer. So I only worked for that place again ad hoc, part time.
Reply by April 22, 20182018-04-22
>"This is a standalone go/no-go RC tester that I made a few decades
ago when repairing TVs made up a significant part of my work. https://www.dropbox.com/s/4266z0r318ua82y/RC%20tester.png?dl=0 " I threw something like that together, but my IR receiver was from a VCR and would not run on much less than 9 volts so it has a little transformer and a cord. I also didn't have any LEDs handy so I used a dial light for a stereo, and that required a buffer of course. And the transformer was 12 volts so I had to put in a regulator. In for a penny, in for a pound. But then I really do have too much time on my hands. Maybe I'll put up a picture of it somewhere. I mean, in next year's dictionary when you look up "kludge" there will be a picture of this ting next to it.
Reply by mike April 21, 20182018-04-21
On 4/21/2018 4:20 PM, Harry D wrote:

> > Mike, I noticed you singled out JL in your harangue. I kind of like the > guy and he is on the list. > > Wash your hands after flushing.
I don't recall mentioning anyone personally. I did give an example of a business that appears to be based on good business sense and a contrasting one of a fidget-spinner much too late to the market. Anyone feeling harangued should read it again and take a chill pill. My intention was to inform, not harangue. I deal in my perception of the facts. I never intend to badmouth anybody personally. Emotion is the enemy of logic. I'm not afraid to argue against what I believe to be an ill-conceived action. I attempt to support the logical argument with fact, history, suggestions, education... Typical defensive responses include reference to a counterexample. I'm talking human nature and statistics. 80%+ of small businesses fail. 20% counterexamples won't change the statistics, they ARE the statistics. Personal attacks are never my intention. People who like to argue or attack might feel threatened and rush to defend. Usually, it's somebody on the sidelines who feels the need to demonstrate superiority. That's the way the internet works. Stuff on the internet is forever. I take that into consideration when I speak. That often means that I expand he discussion to include things that others may not have considered. Failures of things that you planned for are less likely. What bites you in the ass are the things you glossed over or didn't consider at all. The clever idea is a tiny piece of a successful business, but might be the entirety of a hobby. Knowing how to do is the easy part. It's much harder to determine WHAT to do from a holistic point of view.
Reply by dcas...@krl.org April 21, 20182018-04-21
On Saturday, April 21, 2018 at 1:13:01 PM UTC-4, Harry D wrote:
> > > Hey Dan, I too have one of his LC meters. Use it for lower range values. His > meter was some of my inspiration. > > Ciao, harry D.
Neil was my load in about 1968 at the Sat V breadboard in Huntsville. Very smart engineer. Dan
Reply by Harry D April 21, 20182018-04-21

"mike"  wrote in message news:pbg7va$7lm$1@dont-email.me...

On 4/21/2018 11:17 AM, Harry D wrote:

> > Mike, You must be long of tooth to have stored all that negative BS.
Another way to look at it is that I've described a process to AVOID or reduce the chances of failure. Would you rather be a financial partner in a business that threw products over the transom to see what failed? Or one that did the sorting on the front end and produced winners? If you had no emotional attachment and the opportunity to invest real cash money in this project based on the information disclosed, would you do so? I'm not being negative, the product (as disclosed) speaks for itself if you ask the questions. You decide whether to invest. You quote that 80% of businesses fail. I posit that far more fail from errors in planning and execution than from bad product concept.
> Reminds me of a lunch I had with Andy Grove at an Intel seminar. Andy, > what is the best way to start a new company? Answer, "well there are two > ways. One, get some PhDs with fresh Ideas and sell them to the VCs. Then > a tall office building, carpets and pretty girls. Bankruptcy will follow > shortly. Second, design a product that will fill a market need. Make a > few in your garage. If there is interest, expand to a two car garage. > Keep pushing and you have a 20% change of surviving." This is my > paraphrase, AG said it better. > Mike, stay away from sharp objects. > Cheers, Harry D. >
That's entirely consistent with what I said. The second way can work very well when your idea has some of these characteristics: Your product contains 'secret sauce' that can't be copied easily. Is unattractive to potential copycats Your market is small and localized so you can stay under the radar You've researched and KNOW that you won't be infringing on patents/copyrights/etc. Your product is so compelling that people who didn't know they needed it could easily justify purchase. You've taken steps to assure that your house can't be taken away over a frivolous lawsuit. You're willing to do design, development, manufacturing, marketing, sales, support functions for zero financial compensation for as long as it takes. You have sufficient other income to fund the project AND the rest of your life. Surely you can think of lots of other potential ways that might succeed. The second way is a bad idea when your idea has some of these characteristics: Your idea is already available in the market from hundreds of sources who are capable of producing it in high volume and shipping it to your house for less than your burdened manufacturing cost. There are multiple organizations eager to copy any improvements in competitive offerings. Your product contains nothing that can't be copied easily. Your competition is a ghost that can't be held accountable for anything. The low-hanging-fruit portion of the market is saturated. Most users don't NEED the improvements that your product has. Users who do need the improvements may also want certifications, and documentation and support. Selling expensive picosecond pulse generators to government funded research facilities might be a great idea. Selling a $10 fidget-spinner, made from commodity components, that's GUARANTEED to spin twice as long as the 50-cent one on the closeout shelf at Walmart may not be such a good idea. Understanding the market is critical. Understanding the life-cycle of a product is critical. My definition of a hobby is: Spending way more time and money than you can logically justify doing something that gives you pleasure. Nothing wrong with that. Finding out that you had a hobby AFTER you mortgaged the house to fund it will not give you pleasure. Of course, I've presented a bleak picture. Certainly, the OP has no INTENTION of doing any of these things. Will be interesting to see the postmortem two years down the road disclosing all the issues that were never intended. I have a long history of identifying issues with concepts, designs, implementations, markets...AND proposing solutions. Problem is that people don't want change...they seek validation. Anything less than validation is viewed as negative BS. I'd offer to help, but It's time for my shift at Happy Burgers. ;-) Mike, I noticed you singled out JL in your harangue. I kind of like the guy and he is on the list. Wash your hands after flushing.
Reply by Jeff Liebermann April 21, 20182018-04-21
On Sat, 21 Apr 2018 10:57:51 -0700, John Robertson <spam@flippers.com>
wrote:

>On 2018/04/20 6:35 PM, Jeff Liebermann wrote: >> On Fri, 20 Apr 2018 17:52:43 +0530, Pimpom <Pimpom@invalid.invalid> >> wrote: >> >>> On 4/20/2018 4:45 AM, Jeff Liebermann wrote: >>>> >>>> However, there are a few features that I wouldn't mind seeing added. >>>> 1. An IR (infrared) emitter and detector for testing TV remote >>>> controls. >> >>> This is a standalone go/no-go RC tester that I made a few decades >>> ago when repairing TVs made up a significant part of my work. >>> https://www.dropbox.com/s/4266z0r318ua82y/RC%20tester.png?dl=0 >> >> I use a digital camera to view the emitted near IR.
>Depends on the vintage of digital camera, earlier ones did not have an >IR filter, but that is more common now to be coated on the lens. As an >example most phones have IR filters AFAIK.
One would think that should be true because it's so commonly repeated. However, that hasn't been my experience. I just tried the following cell phones and tablets with my Samsung TV remote control: Samsung S6 Works great and very bright Lumia 535(?) Visible Nexus 7 tablet Visible Lumia 928W Dim but visible Moto G v1 Barely visible So, it seems to vary with the type of phone. However, I don't see any relationship to the age of the phone. It's much the same story for digital cameras. All work but with varying near-IR sensitivity levels. Trust, but verify.
>An argument for never throwing out any potential test gear... >John :-#)#
I spent my first 50 years collecting all this junk. I plan to spend my next 50 years getting rid of the junk. -- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
Reply by mike April 21, 20182018-04-21
On 4/21/2018 11:17 AM, Harry D wrote:

> > Mike, You must be long of tooth to have stored all that negative BS.
Another way to look at it is that I've described a process to AVOID or reduce the chances of failure. Would you rather be a financial partner in a business that threw products over the transom to see what failed? Or one that did the sorting on the front end and produced winners? If you had no emotional attachment and the opportunity to invest real cash money in this project based on the information disclosed, would you do so? I'm not being negative, the product (as disclosed) speaks for itself if you ask the questions. You decide whether to invest. You quote that 80% of businesses fail. I posit that far more fail from errors in planning and execution than from bad product concept.
> Reminds me of a lunch I had with Andy Grove at an Intel seminar. Andy, > what is the best way to start a new company? Answer, "well there are two > ways. One, get some PhDs with fresh Ideas and sell them to the VCs. Then > a tall office building, carpets and pretty girls. Bankruptcy will follow > shortly. Second, design a product that will fill a market need. Make a > few in your garage. If there is interest, expand to a two car garage. > Keep pushing and you have a 20% change of surviving." This is my > paraphrase, AG said it better. > Mike, stay away from sharp objects. > Cheers, Harry D. >
That's entirely consistent with what I said. The second way can work very well when your idea has some of these characteristics: Your product contains 'secret sauce' that can't be copied easily. Is unattractive to potential copycats Your market is small and localized so you can stay under the radar You've researched and KNOW that you won't be infringing on patents/copyrights/etc. Your product is so compelling that people who didn't know they needed it could easily justify purchase. You've taken steps to assure that your house can't be taken away over a frivolous lawsuit. You're willing to do design, development, manufacturing, marketing, sales, support functions for zero financial compensation for as long as it takes. You have sufficient other income to fund the project AND the rest of your life. Surely you can think of lots of other potential ways that might succeed. The second way is a bad idea when your idea has some of these characteristics: Your idea is already available in the market from hundreds of sources who are capable of producing it in high volume and shipping it to your house for less than your burdened manufacturing cost. There are multiple organizations eager to copy any improvements in competitive offerings. Your product contains nothing that can't be copied easily. Your competition is a ghost that can't be held accountable for anything. The low-hanging-fruit portion of the market is saturated. Most users don't NEED the improvements that your product has. Users who do need the improvements may also want certifications, and documentation and support. Selling expensive picosecond pulse generators to government funded research facilities might be a great idea. Selling a $10 fidget-spinner, made from commodity components, that's GUARANTEED to spin twice as long as the 50-cent one on the closeout shelf at Walmart may not be such a good idea. Understanding the market is critical. Understanding the life-cycle of a product is critical. My definition of a hobby is: Spending way more time and money than you can logically justify doing something that gives you pleasure. Nothing wrong with that. Finding out that you had a hobby AFTER you mortgaged the house to fund it will not give you pleasure. Of course, I've presented a bleak picture. Certainly, the OP has no INTENTION of doing any of these things. Will be interesting to see the postmortem two years down the road disclosing all the issues that were never intended. I have a long history of identifying issues with concepts, designs, implementations, markets...AND proposing solutions. Problem is that people don't want change...they seek validation. Anything less than validation is viewed as negative BS. I'd offer to help, but It's time for my shift at Happy Burgers. ;-)
Reply by Harry D April 21, 20182018-04-21

"mike"  wrote in message news:pbbvbp$ru$1@dont-email.me...


There's a lot of missing information about price and who you expect
to buy it.
Based on what you (OP) have said, this is a VERY BAD IDEA on many levels.

Just to put you in the frame of mind...
Would you have liked to have been the first vendor for the Pet Rock
(and had the sense to get out at the peak)?
How would you have liked to have been the 240th vendor of the Pet Rock
five years later with a marketing strategy based on more compelling
writing on the included paper?

You have two choices.
1) Waste time having fun, the definition of hobby.
2) Create/market/sell a product, the opposite of a hobby.

If it's the first, go forth and have fun.
If it's the second, you gots some work to do.
And almost none of it has to do with engineering the hardware.

You absolutely, positively, MUST write a business plan.
Scribble it on the back of an envelope, but do it.

My suggestion is that you start near the end and work backwards
toward the design.  If you do that, you'll waste less time and money
before scrapping the project.

I'd start with the discount cash flow analysis.
Plot out the timeline for the project and the rate of cash burn
you'll be spending in each phase.
Assume you'll be borrowing all that money and paying interest.
Factor in the cost of money.

At this point, your mouse cursor is racing toward that "reply"
icon to TELL ME that
you're not borrowing any money.  You'd be wrong.  Opportunity
cost IS borrowed money that you could have left in the bank
earning interest or working on a different product.
Take a chill pill and read on...

The graph will be headed down until revenue exceeds the
costs of maintaining the product...fixes, improvements, sales, marketing,
distribution, taxes, fees, insurance, the list is endless.
Of course, you've considered the cost of product liability insurance,
business fees/taxes/licenses, EMC compliance, safety certifications.
If you don't it's a hobby and may become a VERY EXPENSIVE hobby.
Nothing wrong with a hobby, just don't confuse the two.

The graph will eventually head up until it reaches zero.
That's the time to break even.
The PROFIT accumulates and earns interest until product end of life, or
somebody
takes your market away.  In the first case, you can sell off inventory.
In the second case, you get to scrap it.
If at the end of the run, the graph is above zero, you done good.
If it's below zero, you wasted a bunch of your life.

The time value of money shows up as your graph deviating, in the undesired
direction, from straight line segments.

You'll also note that delays near the end are FAR MORE EXPENSIVE
than they would have been for the same delay inserted at the beginning.
Delays near the end can, and often do, convert profitable ventures
into losses.
You've known that since the day your mother taught you that an ounce
of prevention is worth a pound of cure.  I posit that upper management
never had mothers...but I digress.  Do the damn business plan.

Next thing to do is make a second discount cash flow graph.
Assume you got a job at minimum wage, maybe flipping burgers.
Spend as much time at that job as you'd spend on the project.
Put the money in the bank and earn interest.

Lay the two DCF graphs over one another.
If the PROJECT line EVER crosses the burger flipping line,
you MIGHT have a profitable project.  If the project line
at end of life exceeds the burger flipping line,
you've got a winner...unless you don't get everything perfect.

Notice that if you delay the project or don't meet sales targets,
the DCF can go horribly wrong very quickly, while the burger flipper
just keep raking in that minimum wage.
If you have marketable skills that exceed minimum wage, by all means,
use that higher rate in the analysis.  It will be much harder to
exceed the burger line.

If your optimistic estimates of schedule, volume and price don't
pencil out, you should abort the project.  Things rarely go
better than plan.  You can guess what happens if you don't
even have a plan.  Did I mention, do the damn business plan.

In the real world, you'd want the project to EXCEED the burger
flipping line by a large margin.  Shit happens.

Full disclosure...
I've had several "SUPER-DUPER" ideas over the years.
None, zero, not any had a projected outcome that exceeded
burger flipping.  There was no way I could have mustered the
resources to take advantage of the opportunity in a manner
that suggested a profitable outcome.
If I'd been a Chinese manufacturer with an established sales channel
and experience going from 0 to customer in weeks, the story
may have been different.  Those guys, by the way, will be the
ones who take your market away.

Yes, I'm conservative.  I'd rather live the quiet hobby life instead
of the stomach ulcer, heart failure life.  I try not to buck the trend
for projects that don't pencil out.

But I digress...
Back to the topic...
You have to decide what you're selling.
What do you offer that people will be willing to pay for.
Often, what people will pay for is not what you think of
as your primary deliverable.  There's an example at the end of this.

It's important to separate emotion from reality.
Who is gonna buy it.
As an ebay hobbyist:
I know that there are many clones.
I know that specs/features get exaggerated beyond capabilities.
I know that you can never trust anything.
I bought the cheapest available.

As an engineer for whom time is money,
I'd buy from a reputable vendor.
I'd buy a product with proper standards compliance/certifications.
I'd expect EVERY spec to be met.
I'd expect the test algorithms to be documented sufficiently
for me to trust what I was measuring.

What do they expect from it?
The unit I bought could test capacitors for ESR.
Was the number accurate to 1%?  I doubt it.
Could I use it to identify failing caps?  Youbetcha.

I plugged a bunch of different components on it and was
amazed at how well it detected good devices.  I was
far less impressed with the results on defective devices.

I expected to be able to measure the ESR of a battery.
No dice, unidentified component.
OK, put a big cap in series and measure the sum.
No dice.
OK, match two cells, hook 'em up reversed so the sum voltage
was zero.
No dice.

The designer of the tester had a vision of what the device
should do that differs from mine.  Nothing in the spec,
or lack thereof, prepared me for that.
I think I paid about $10 for the thing and am well pleased
with it.  If I had a production line shutdown depending on that
diagnostic, I'd have been pissed.

For a device this general, you can't please everybody.
And the tighter specifications don't matter much.
Any process that really depends on that accuracy
will also require traceability.

Is more really better?
Sure, if you're matching 20V zeners, you could make a case.
But how many people do that?

What if, instead, you were matching Vbe on low noise microwave
transistors, do you really want those three leads thrashing all over
the place breaking down the device trying to determine what it is?
Yes, nobody in their right mind would use this to do that, but
not everybody is in their right mind.  I'm sure you can come
up with other cases where 24V is not a good thing.  After watching
the waveforms while it analyzed the component, I'm not sure
that I like what it's doing to the parts.

In case you missed it...
Unless there's a lot going on that you haven't mentioned,
I predict that this is a BAD IDEA.

Yes, you can design/develop/manufacture/market/sell a product
at a profit.  I suggest, based on available disclosure,
that this is not one of those cases.

I have a story to tell.
Hopefully, it's more helpful than a long, contentious thread about politics.

Long ago in a galaxy far away, I attended a "club" of entrepreneurs
who met to exchange ideas and provide support to one another.
Attorneys were present
to help you, for a fee, draft your patent applications and perform
other necessary legal services.

I listened to a bunch of presentations; most of them horror stories.
The people who made money were the lawyers and the people collecting
membership fees.
The entrepreneurs, not so much.
They all thought their idea was gonna make them rich.  To a man-1,
they were blinded by their brilliance and didn't have a clue about
what it took to make a buck.  Sure, some of the products were innovative.
What was lacking was the business plan.

The outlier understood that you can separate the user
of the item from the one paying for it.
He designed a plastic extrusion, cut it into 1" sections and sold it to
advertising distributors.
Businesses contracted with the distributors to have their advertisement
printed on it and gave it to customers.  Customers used it to clip
a pencil to the sun visor of their car.

It was free, so customers took a lot of them.
It was profitable for the advertising distributor.
It was cheap advertising for the businesses.

The entrepreneur took orders for blank plastic from the few distributors
and forwarded orders to the plant in China making the plastic.
As I recall, he made about 5-cents each, but the volume was high
and sales cost was near zero.

There was no contact with actual user.
Nobody cared about the quality of the plastic.
There were no specifications.
Life is sweet when the end user doesn't pay the cost and didn't
even know he needed one.
How well the product held pencils was not an issue.
A competitor offering a product that held pencils "better"
at the same price wouldn't get anywhere.
The market is so huge that clones wouldn't impact sales.

The entrepreneur was not selling pencil holders.
He was selling efficient advertising.
And he was not involved in either manufacture of the plastic
or the sale of advertising.

One more story, since I'm on a roll...

One guy had a contract with local chain stores to supply
lawn care products.  I forget the details.
Iron is an important factor in green lawns.  The guy
spread iron filings on his lawn and it got green.
Since the iron didn't disperse, his lawn was still green the
next year.  OK now, we've got ourselves a product.
He was so sure of his product that he offered a 5 year money
back guarantee.
He built a machine to bag iron filings and set up to produce.
OOPS, you can't just buy iron filings.  He had to buy a train car
full of it and rent a space to store it.  No problem, this is gonna
make him rich.  The stuff flew off the shelves.

The first year was great.
Two years later, people started wanting their money back.
Vendors took it off the shelves.

He had to pay back all the purchasers who demanded it.
He still had almost a full train car load of iron filings
that he couldn't get rid of.

He had let his optimism override rigorous validation of his
claims and it came back to bite him.


I've got many more horror stories about how very smart people
let their enthusiasm override logic.

Mike, You must be long of tooth to have stored all that negative BS.
Reminds me of a lunch I had with Andy Grove at an Intel seminar. Andy, what 
is the best way to start a new company? Answer, "well there are two ways. 
One, get some PhDs with fresh Ideas and sell them to the VCs. Then a tall 
office building, carpets and pretty girls. Bankruptcy will follow shortly. 
Second, design a product that will fill a market need. Make a few in your 
garage. If there is interest, expand to a two car garage. Keep pushing and 
you have a 20% change of surviving." This is my paraphrase, AG said it 
better.
Mike, stay away from sharp objects.
Cheers, Harry D.