are all the function generators gone?

re.

DDS

o a

s'

lf

y do

als?

Yeah sure... Jon as I've said to you, I don't expect to make much (if any) money from teachscope.. the volume/cost is too low unless someone wants to market it to groups other than physics labs. I saw the 50MHz Rigol scope, for $330?, can't sell teachscope for more than a scope!

Maybe it could be a sparkfun thing?

George H.

Actually, it would be just plain enjoyable. I've had a full year of suing the State of Oregon after a disagreement with an audit (Tax Court -- first the Magistrate Division, now the Regular Division -- last recourse after that is the Oregon Supreme Court .... may be fun.) It's been a handful. But I looked very much forward to this and have read through and collected various ideas from the user point of view, plus examined various ideas (including Walsh function approaches) to this project. I consider it just plain fun -- especially if this is done with a focus like Jörg on absolute least cost while providing the most possible with that.

I've also a 3D printer that was build earlier this year and is working nicely. I considered a proto from that, as well.

Anyway, I'd love to see something done you could either point to or else sell. Money isn't my focus... it's just pure fun to me.

I just did my first Disclosures to the State, yesterday, and expect to meet again in February with the Judge to set a court date. But I have time in the middle. I don't care who takes the lead or even if I'm involved. But I would be pleased to get a peak as things move forward, if possible. Even more pleased if I can help share some of the load.

It's all good, to me. I just want to see it happen.

Jon

appearing

they do

signals?

cost

I am not so sure about that, a lot depends on how we position ourselves in the market. Just the same it would be a tough sell.

That is worth serious consideration, it could be sold case less and no PS, with designs for the case and labeling. Normal PS might be a USB wall wart (~ 3$ US ?) and make the rest onboard, with their test signals as part of the signals to look for.

I am cool with all that. George this is your project so far, what say you.

?-)

I agree that it could sell for the cost of a scope -- depending on what is included. For example, it may provide alternate functions (signal generator?) which later have practical use, as well. And especially with a nice manual.

I think making it "ready to go" out of the box will give it wider utility. It would be nice to give it the look and feel of a typical Rigol instrument, in fact.

But I'm also for anything fun.

Perhaps start by examining existing oscilloscope tutorials (some manuals have been produced for that) and similar manuals for similar tutoring devices like this. See what others have considered and done. Make a list, work through it and organize it with an eye to functional blocks. Then widdle away expensive functional blocks that can't be justified (along with the associated features that require them.) Then see where that leaves things and find the right balance. (Delving enough into certain details may be required to lock down that they can be achieved, of course.)

Jon

ring

they do

ignals?

/- 5

st

Well the only market that I have access to is college and university physics labs. Marketing is expensive, I think part of the reason that we dropped the product was that we were coming out with a new catalog. I'm not sure of the exact numbers, but it was several $k for the printing of a nice glossy catalog page. (in quantities of 5k to

If the pcb could be done all surface mount, with all components on one side. Then you might be able to use the other side as the 'user' interface. pcb silk screeen becomes the artwork, and you just need a five sided (plastic) box as a back side cover.

(I'm thinking of Dave Jones and his uA current meter gizmo.... I could find a link if you don't know of what I speak.)

George h.

do

signals?

Does anyone actually use glossy catalogs, anymore? Vendors obviously publish them because they show up with them but after thumbing through them for maybe a few minutes, either they go back with the vendor or in the trash. The web is an infinitely better marketing tool and a

You're not going to put traces on the back side? That seems a waste. Many devices mount the UI buttons on the main board. It's a PITA to route around them but do-able. A cheap overlay looks a lot better and is more durable than PCB and silk (ick).

Is this going to be a one-off?

Grin, I don't really want it as a project! I'm happy to consult, advise, give you access to 'my' market. I've got more profitable 'fish' to fry.

That said, I wouldn't mind working on it during 'my own' time. (My own time seems awfully limitied these days, given work, house, my wife and two wonderful kids... ages 11 and 13. They suck up a lot of time... which is great, in a few years they'll be gone... well not the wife :^)

George H.

aring

they do

signals?

ost

in

Well that's fine too... If you want to grow it into something bigger. We also have a use for function generators that go below a few Hz. We resell these protek function generators, which leave a bit of a bad taste in my mouth. There is a six or seven digit frequency display, the last few digits of which are meaningless, you change the freq. from 1.00000 kHz to 1.00001 kHz on the display, but nothing changes at the output.

S,

Grin, knock yourself out! In some ways I fear that (many) physicist's have already abandoned 'scopes. I was at a summer meeting and saw demo experiments where the interface was National Instruments (DAC)-> labview-> computer. With no ?real time? display of the signal. It seemed crazy to me not to have a ?scope to give (at least) real time diagnostics of what was happening. But clearly a scope was not a tool that they felt comfortable using... sigh.

George H.

Wait... let me apply my Sherlockian psychic intellect... her name is Patricia!

pearing

they do

ignals?

o +/- 5

st

s in

Grin... we're perhaps a throw back company in lots of ways. We send out a nice catalog in a three ring binder. The hope is that people will put new catalog pages into their binder. We do a lot of marketing at trade shows/ conferences. There it's nice to hand out the catalog page of some insturment if there is any interest. Mostly these are ~$5k to ~s15k insturmnets, so a dollar glossy page is not much waste if one of every hundred generates one more sale. (I'm making up numbers.)

PS,

(If I can get someone else to make it then I don't have to put traces anywhere. :^)

OK, maybe that was a bad idea.

George H.

ne

icia!

chuckle... OK Sherlock what color is her hair?

George H.

.highlandtechnology.com  jlarkin at highlandtechnology dot com

Jon Kirwan a écrit :

Jon, which 3Dprinter is it? A DIY one? (URL?) How much effort and money did you put in it?

It's the MakerBot Thing-O-Matic with the MK7 extruder. It was shipped as a kit of parts and assembled. we currently run the ReplicatorG software on the PC and use the SailFish GCode interpreter on the driver boards. We bought the automatic build platform, but in use found it to be almost unusable. So we reverted to the manual platform and use a Kaptan built surface over the aluminum heating surface. Our purchase price was $1100 (they placed it at discount for a short time.) But we also spent some additional money on ABS and PLA wire.

My youngest son did almost all the work in assembling it. He is sufficiently 'autistic' that the federally appointed psychologist assigned to evaluate him qualified him as permanently disabled. I don't accept that, but admit that he is sufficiently limited that he requires care (he cannot be trusted to shop for himself or to safely ride a bus on his own.)

He is able to read datasheets and do basic electronic design tasks (I don't need to tell him anything about pullups or pulldowns in a micro, for example, or how to check and verify hysteresis thresholds for the inputs against the schematics

-- he is thorough and doesn't need to be guided much), programs in C, C++, Lua, and several different assembly code languages, is learning Laplace from me, and can solve first and some second year physics problems handed to him. His mathematical imagination is very good, he's competent at calculus, and I can set interesting problems he's never faced and his intuition takes him, usually in seconds, to the right place.

We bought this because of my own experiences circa early 1975 when I purchased and built an Altair 8800 and then, when they came out, two 4k dynamic ram boards (which didn't work because they were designed wrong... forcing me to learn more about electronics over tears having already spent months of spare cash for them.) Although there are differences, there are also many similarities between the 3D printer environment now and around the time of the Altair 8800 and the IMSAI 8080

-- before the Apple appeared (which was shopped at first as a [expensive] game machine before visicalc appeared and carried it into a financial success by providing a real reason to spend that money.)

The "visicalc" concept hasn't yet hit the 3D marketplace. I don't know what it will be, either, except that it won't be only software but will also be something related to mods to the extruder. One idea I'm considering right now is to add a camera to observe the threads of plastic being laid down to provide closed loop control of the extruder -- which is lacking now. But that, even if useful, isn't enough. There will have to be the right software to go with that for some solution to a common, yet specific application area that will drive sales of this lower cost end of the market. 3D printers, for now, are much like useful microcomputers were in the late 1970's -- obvious uses in business (accounting, record keeping, etc) that were traditional. The main thing was that smaller companies with less annual revenue could start affording their own computer. There wasn't much packaged softare (mostly Microsoft papertape BASIC and/or CP/M) around. So a lot of custom activities abounded.

It was visicalc and the imagination to bring that novel approach forward that rapidly caught attention and moved microcomputers from the homes of geeks and into the enduring attention of business managers. I watched it happen over a very short few months, in fact, seeing the change of clients coming into the computer stores at the time over a matter of months.

My wife and I decided that this was an experience Lee (our son) was ready for and an opportunity that only appears once in a while. I don't believe MakerBot sells the kit, or any kit, anymore. So the oppotunity to build from a well planned kit no longer exists, so far as I'm aware. Sad to say. Still, of course, there are many, many plans. But these are too difficult for someone young and able to built and learn from kits but not yet ready to work from a hodge-podge of parts and documentation.

We initially "forced" him to apply an hour a day on the project. (He needs such guidance for anything "new.") But he soon was self-driven -- more than we'd expected -- and it was assembled over the course of a month and a half, I believe. We began serious testing of standardized objects in April, I think.

There are many interesting problems yet to solve. The Thing-o-Matic is quite good regarding precision in 3D space and the backlash is very small. But the extruder head is hot

-- near 240C or so for ABS -- and during a movement which does NOT involve extruding ABS but instead just moving to a place where extrusion will start up again -- the ABS plastic does 'dribble.' Too low a temperature at the head and the ABS applied to the object has problems. Hot enough to solve those problems leaves the ABS able to drip. It's not much, but it shows up on every project that involves such "gaps" in extrusion. We just "knock" the tiny bits off. But it's something to consider in a new design, for example. There are many, many other issues. For example, no amount of precision knowledge about the diameter of the ABS plastic thread driving wheel, the thickness of the ABS plastic wire, knowledge about the stepper motor itself, etc., can completely and precisely predict the exact amount of ABS plastic being extruded out over a LARGE project. ANY error simply accumulates over time. And there is ALWAYS error. So this is one reason why I'm considering a tiny camera at the extruder head.

I consider the experiences we've had, and continue to have, to be absolutely wonderful and priceless. I am reminded very much of my early days near to, during, and shortly after the Altair 8800 days. And it's been nothing but joy to work through issues, learn about problems, consider solutions, try them out, etc. Also, the software is getting a lot better, too. And NIST provides complete docs on GCode, as well as standard parser and execution code -- not for 3D per se but for all numerical control machinery.

I am researching the construction of a much larger and more versatile machine, now. And I'll be taking these experiences into account, where I can. It's really neat, actually.

About the time required for the unit my son built -- I'd guess at about 40 hours or so of somewhat unskilled labor, plus occasional help from me (but very little of that, since our purpose was that he did this for himself and would feel success from it and would own that success.) I would have required perhaps 15-20 hours, I think. But I've no real idea since I pretty much left it to him, except where he'd drag me over for a clarification of something. I spent a lot more time since then observing details of its functioning and documenting problems I'd like to think about solving for a new machine.

Jon

appearing

do

signals?

They can get a real-time display from a suitable .vi, but typically the lab guys have it set up to autoscale so it's very hard to get a analog-y feel for what's going on, despite the rolling chart display. For example, a signal that is trending will look like it has much less noise than a signal that is stable with exactly the same noise level.

appearing

do

signals?

Sounds like they might use a binary DDS and do ugly rounding to decimal. That problem can be worked around, but the DDS clock frequency gets messy, or you need a huge, like 64 bit, phase accumulator.

I toured the Cornell EE dept when The Brat was at Cornell. I counted about 40 computer screens and one oscilloscope screen.

ppearing

they do

signals?

to +/- 5

ost

ne

es in

o PS,

It was worse than that. There was a prof, who had made a different mechanism for doing the bouncing gold wires, quantum conductance mesasurment. (The bouncing wires give transient steps in the resistance of 12.7(?)k ohms(e^2/h,) (Really thin wires work best.) Instead of bouncing wires, he had the wire on a beam, that he flexed from the back with a fine pitched screw. I thought it might work, but the dam# labview program, could only capture the time record and then display it later. So he'd move it around and then go see what he 'caught'... no better than bouncing wires. I so wanted to push the prof out of the way, grab a scope, and see if the mechanism worked. I'll have to build my own someday.

George H.

I previously misunderstood your question... see Fig 5 in the patent... it's a synchronous "phase flipper" ;-) ...Jim Thompson

ppearing

they do

signals?

to +/- 5

ost

ne

es in

I didn't look at it that close, but it seemed to tick over on what might have been binary numbers. So I can forgive maybe one 'funny' digit, but this had three! You'd never know as a user, unless you had something to compare it with.

In other ways the protek is OK, I've used the sweep function and burst mode, (I think I was even able to send an external trigger for the burst.) And like I said the only cheap DDS I know that has milli Hertz resolution.

George H.

o PS,

Maybe they keep the 'scope guys in the back room, their lab benches will be messier.. clutter looks bad to visiting parents.

George H.

ed text -