I was looking at using a microcontroller at one point. I was really only looking at PICs since I don't know anything about what's out there, and while the PICs themselves were inexpensive, the programmers I found were far too expensive. If you could show me how I could go to microcontrollers inexpensively I'd owe you one, since going to a microcontroller would allow me to add other features I'd wanted but couldn't justy (integration of humidity data (this is for a humidor) and even a display ;'] ). It isn't quite as simple as one signal and one control though, I'd be running 3 seperate controls off 4 seperate temperatures, but I have seen PICs that have enough A/D capability to read all those temps.
Except for some really old PICs (and some "esoteric" ones), PICs these days can be serially programmed with 5V pulses, and as such you'll find various almost-free parallel port programmers with a little Googling (try, e.g., "pic programmer schematic").
Personally, I'd suggest the Atmel AVR lines of microcontrollers as being somewhat more programmer-friendly than PICs, and with more or less comparable hardware features (as well as a slew of free "do it yourself schematics" available via Googling or via avrfreaks.com), but if you're familiar with PICs they'll certainly get the job done.
-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | |
formatting link
| 1962 | I love to cook with wine. Sometimes I even put it in the food.
Or, you can go with the Freescale parts, which have free software, and even 'Beans' to make the programming easy. Many have serial bootloaders and internal debug to make programming and troubleshooting easy...
Can you still get 709's? I once got some uA702's (I think that's what they were) in a box of mystery stuff, I think they were even in their individual packaging, and looking them up they must have been even worse than 709's.
In message , dated Tue, 29 Aug 2006, Michael Black writes
I have some 709s for fixing old test equipment. It's doubtful whether they are worse than the 741 or not. Being externally compensated, you can do tricks that you can't with the 741. I've not used the 702.
How about simulating an op-amp made of germanium devices?
--
OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk
2006 is YMMVI- Your mileage may vary immensely.
John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK
The 702 needed +12V and -6V power supplies, and the common-mode range wasn't all that impressive. I never used one, but made a point about one appearing in an ostensubly new design in 1982.
The - very good - engineer who had carried it over from an old design was grateful for the chance to design it out, or at least said he was. I knew that Jerry was brighter than I was so I was just relieved that he hadn't taken the chance to carve me up.
I too did use the 709 - nasty, but at least the output stage isn't as close to oscillating as the 741 output stage appears to be. I've not used the 702 at all, but I have seen designs where it was used - quite effectively.
On 27 Aug 2006 07:37:22 -0700, snipped-for-privacy@triton.net wrote in Msg.
After this thread has been going on for quite a while, let me suggest this: Since the circuit works well with the part you designed in, there's no reason to change it. Reasons to change it would be:
- You'd like to improve on the circuit, and you have asserted that its current shortcomings can be traced to specific parameters of the 741. In that case you can look for a part with better specs in that respect. If that is the case, you need to tell us specificially what you'd like to improve, with an accompanying schematic.
- The part you designed in is hard to get, not manufactured any more, or too expensive for future versions of the circuit. Clearly not the case with the 741.
About 20 years ago I built something which contained several boards with about a dozen 741s per board. This was for some art project of some friends of mine. It worked just fine. Then I wanted to build the stuff more compact by replacing all the 741s with 324s (4 opamps per package). For some reason or another the circuit stopped working. I may have exploited some quirk of the 741 without knowing it, or made some other mistake. I didn't have much of a clue back then. Anyway, I ended up building the whole thing with 741s, and it still works fine today. Morale: As long as something works fine with cheap and abundant parts, even if they're crap, there's no reason to change it.
In message , dated Wed, 30 Aug 2006, Robert Latest writes
The 324 requires 6.2 kohm 'pull-down' resistors on its outputs in capacitor-coupled circuits. See the data sheet. With that, it would probably have worked, but it's not a quad 741, unlike 4741, 4136 and a number of others.
--
OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk
2006 is YMMVI- Your mileage may vary immensely.
John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK
One good reason for replacing the 741 used to be popcorn noise at the input.
The 741 specification doesn't say anything about the noise performance, and the original part was notoriously subject to "popcorn noise"
formatting link
For a while RCA sold a 741 variant which did have a popcorn noise specification.
Modern 741's and other op amps are widely supposed to be much better in this respect
formatting link
but in 1987 I replaced a 741 with an OP-07
formatting link
and got rid of some 300uV of pop-corn noise, whch eventually did seem to improve the quality of the GaAs single crystals being produced on this side of the then Iron Curtain - the Cambridge Instruments/ Metals Ressearch GaAs Crystal pullers involved produced (at that time) some
95% of all the single crystal GaAs sold in the west.
This was in fact just the first step in a more thorough clean-up of the weight-sensing circuitry that drove the heater control loop, but it was a large step, and nothing that followed it produced such an immediately perceptible improvement - for one thing the RF induction heater stopped being bang-bang controlled in step with the step-wise changes in offset that constitute popcorn noise, and went over to the sort of proportionally controlled operation that they'd been designed for .... a very audible difference, as it turned out, and one that eliminated the interesting ridge structure you used to get along the side of the GaAs single crystals.
Looking at the Atmel stuff I've descided to go with that. Forces me to learn a completely new electronics disciplin (not that I really knew anolog electronics) but what is life without challenge ;']
I think I will still need to buffer my incoming voltages from my thermistors so can I get a recommendation for that? Signal range will be well within my 0-12V supply. Next could you recommend some form of DAC that I can output my control voltages on, 8bit should be sufficient and they don't need to be powerful.
Edward
PS. Thank you for all the help, you've saved me from much misdirected effort.
In message , dated Wed, 30 Aug 2006, snipped-for-privacy@ieee.org writes
I believe it's a matter of the die surface being uncontaminated par excellence. I had similar trouble with BC109s from different suppliers and with TBA820 soon after it was released. The latter could vary in flicker noise by 15 dB between samples. It has only a 'nominal' figure for noise.
--
OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk
2006 is YMMVI- Your mileage may vary immensely.
John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK
Everbody makes it, it costs a couple of dollars in small quantities and offers lowoff-set voltage, low voltage noise (at 10nV per root Hz) and all the bandwidth you will need.
The OPA111 and OPA627 FET-input parts are better, but an order of magnitude more expensive.
Don't dissipate too much power in your thermistor - like all negative temperature coefficient devices, too much dissipation leads to current chanelling and unstable resistance. I've got away with 10uW, but anything over 100uW can be too much (and some multimeters will dissipate that much in a resistor under test).
Most single chip microprocessors are designed to make it easy to generate pulse-width-modulated (PWM) outputs, with special programmable set-and-forget counters to do the job. With a 20MHz system clock you can get 8-bit pulse width control on an 78kHz pulse stream, and you can get finer control by alternating between two adjacent pulse widths over a longer period - my millidegree controller used this to push up the output resolution an effective 18-bits.
PWM of the driving transistors means that they won't get hot, which is handy, but you do have to filter the PWM output to prevent the high frequency content getting out to disturb the neighbours - my millidegree controller papers details our solution that particular problem.
For your application, which seems to involve frequencies no higher than you can shake a speaker, the LM741 is likely to be okay.
There are hundreds of newer op amps, most with better specs, like faster response, lower offset, lower input currents. But if I assume you have a strong signal from your auxiliary "laser", a 741 should be okay.
If you notice problems with noise or drifting you might want to try a quieter or more stable or self-nulling op amp, of which there are many. The advantage of the 741 is that it's mighty stable and doesnt have too much bandwidth to encourage oscillations.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.