RPM Meter

Then maybe magnetic isn't the way to go right now. I'd suggest more like

2000-3000 turns, of very fine wire -

Or, just use a white stripe and a reflective optical sensor; it won't matter how fast the shaft is going:

Good Luck! Rich

Right. He only distributes technical excellence now in private emails, because we village idiots don't deserve such wisdom. Darn, I was hoping for further adventures in rusty patio furniture.

John

Look at your ftp or firewall settings, works fine here on Win7 with Google Chrome browser, and on Slackware with wget.

Check for passive FTP enabled?

Grant.

Thing is, being reminded of a topology that may suit a particular case or problem is a valuable reply, doesn't take much time or effort to post, one gets from paper to camera to web site fairly easily these days. Even I can do it ;)

I like ideas, reminders. If I wanted worked design solutions I'd be using text books, application notes and never be reminded of other ways to view a problem. Like that recent issue I had with ground noise, only two suggestions posted, neither worked on its own, but both in tandem worked.

I was reminded one could put some impedance in the ground lead, I'd seen that done before, but forgotten the technique for that case.

Out of 31 different IPs viewed the page one person (I think) made a couple suggestions that worked in tandem.

Whatever fun there is in a flameful reply and/or the occasional flamefest is lost if it drags on for weeks, or starts becoming personal attack. Ideas can be silly, doesn't mean the person is (well, not all the time :)

Grant.

On a sunny day (Fri, 30 Jul 2010 09:14:03 +1000) it happened David Eather wrote in :

in:

amplitude.

accuracy of

the regulator,

needed.

control.

Thats is true, as this seems to be a one time thing, in large quantities it would be the other way around, a calibration step costs money. But even for a one time thing, how much time will you spend looking at that speedometer wondering: 'How much is my average speed, what distance did I travel, what was my maximum speed, did I, or do I, break the speed limit', and maybe even estimated time of arrival? Can all be done with a PIC and LCD, and a couple of push buttons. The real time clock is a bonus that comes free. Couple of evenings coding :-) Better then some TV.

Oh well I have a little PCB with a 555 on the desk now as RC a servo tester. It also has a 78L05 regulator. But I also did a servo controller with a PIC, more features, many more :-)

And don't forget: CHARGE IS CONSERVED :-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |

                   Spice is like a sports car... 
     Performance only as good as the person behind the wheel.

Nope. I've decided only to not designate your crap as "crap"... that seems to be what upsets your supplicants, and ass-kissers... like "markp"... is he an employee ?:-)

I will simply post a link to the real deal and posters will learn or lose at their own option. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |

                   Spice is like a sports car... 
     Performance only as good as the person behind the wheel.

Insulting me now? Be careful Jim, you use giganews to post and the headers contain an abuse reporting url.

Mark.

Circuit topologies are fun all on their own, and are appropriate as a starting point for discussion of an underspecified problem, like the ones we usually get here. Circuits are sort of like improv comedy, or jazz, or slam poetry. The one we had on current limiters was fun [1]

I don't use Spice much, so I start with circuits without values and "simulate" them in my head. I visualize the way I'd want the circuit to work with correct but in fact unknown values, and if it looks interesting, real values could be plugged in later (one at a time even!) to force the dynamics to align with the image. [2] People who "visualize" a circuit's operation in Spice have no choice but to use real component values and then fiddle for results.

I do admit to limited expertise with rusty patio furniture.

I like to drift "off topic" by injecting technical asides into the henpecky personal rants. But that makes some of the hens even madder.

John

[1] I *still* need a good, small, low-drop current limiter for use as a self-resetting fuse. The ideal part would be a SOT-23 whose current limit is set by one resistor, thermal limits, and retries at full current once in a while. How come nobody makes one, at least past 5 volts? Polyfuses sort of suck. [2] Suppose you had a Spice model that worked just right, but the values were hidden. Based on the waveforms, you could determine the actual parts values, one at a time, in any order. That's sort of the idea: visualize the circuit and the waveforms, then start poking in values. Sort of the anti-Spice.

Here's an EDN article about low dropout constant current source (figure 2 in the article):

I'm wondering whether you could add another resistor between R2 and ground, and in-between them put an N channel MOSFET. You could detect current limiting condition with an ADC input from the voltage at the top of that second resistor and turn off the FET, thereby isolating the supply from the output. You could program it to retry once in a while once it had detected current limiting. Not exactly a 1 chip solution though....

Mark.

On Fri, 30 Jul 2010 09:14:49 -0700, John Larkin wrote: [snip]

[snip] [snip] [snip]

Asides, if they address technical issues and are not too far out of some range of reasonable s.e.d topics, probably offend nobody. But snarky asides [like the "rusty patio furniture" one above] intended as putdowns accomplish nothing positive and make you look petty, stupid and nonprofessional. If you have something useful, accurate, interesting, new, creative, inspiring or educational to say, then say it. If you want to say something petty, untrue, ad hominem, incomplete, obamic, vague or snarky, then don't.

Thanks for the help, but no joy. I'm giving up on it.

[snip]

How about the RPM meter? IIRC you wanted pulses to count. Trivial to do. Contact me if you need help.

Likewise if you want a DC voltage versus RPM.

Let me know. ...Jim Thompson

-- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

| 1962 |

Spice is like a sports car... Performance only as good as the person behind the wheel.

That's where I'm at too, I found spice useful in modeling changes to a commercial (poorly designed) PWM motor controller, even though I had to play with defining the load so it matching CRO waveform, but in that situation it saved much spider's nest soldering to improve matters.

And when I did change the board to new component values, the spice'd effort was good match to the real.

But not for design from scratch, only sections where I'm not sure of the outcome. But then, I've not used LTSpice enough to use it effectively.

I don't think it helps me define component values for say a calibration network, where I'm prepared to put a trimpot and two E24 1% resistors north and south to get wanted range.

Spice will let me check the result, but will it find the component values (with 1 or 2 1% E24 each side of trimpot) if I tell it the trimpot value and reference voltage?

Which is okay in some situations.

This must've been before I joined this group? Or in a thread I killed. I kill many threads by subject because they're way out of my area or about USA issues I know nothing about.

Just shows up the intolerance some have? A little madness helps :)

The sad thing here is that Jim T. could probably design one, but unless a zillion others want it too, it'd never make it to silicon.

Your spec sounds like a low power, low dropout regulator that self protects and can be made to oscillate via thermal lag or feedback.

My first polyfuse 'hands on' experience is the 1.85A one in series with a hobby drill, with a bright red LED across it. It's showing me much about the beastie's behaviour on overload, and near overload while recovering.

They're quite slow, take seconds to recover. So they're very good for protecting stuff like a motor from losing its magic smoke, or a speaker driver, actuators without limit switches. Not so good for electronic protection except as another place to share the heat part of coping with the overload?

But in order to build the spice model, one first visualises the circuit and guess/estimate the component starter values. It'd be fine if the spice schematic could be imported to the CAD for PCB layout and product documentation, BoM, etc.

Grant.

Nor would it protect during startup or watchdog reset while the processor is not active. Protection needs to be simple, reliable.

Add too much complexity and the gremlins will find a weakness and hit.

Grant.

I'm giving up on having a look at ftp://jjlarkin.lmi.net/DoubleTach.jpg. Whatever is going on it seems to be beyond my power to fix (I changed to windows XP firewall, tried multiple FTP download clients - including with passive mode - I suspect it is my network setting which are crewed down for security and which I won't change)

My only bit in the thread was to suggest an analogue solution to the OP's request for a way to check his malfunctioning digital device.

Many thanks for the offer of help.

Here:

If John L. objects to my republishing, I'll happily remove it, so be quick ;)

Grant.

Fair enough, usually ftp sites are easy to get to, but fewer people access ftp these days.

Grant.

Try this, it's a short url to that link:

If the above works, go to

and put in the url you are looking for to convert. This would only work if HTTP redirect works when putting in the full address doesn't, so IMO this solution is unlikely to work, but worth a go.

Mark.

Another thing that it might be is a DNS problem. That might be a common problem to all your ftp clients (you could try to ping jjlarkin.lmi.net from a command box or try the link below). ftp://66.117.156.8/DoubleTach.jpg