What happened to Phil Allison?

te:

you

k

Far from it. You've been championing the 555 here for years, as if it was a device that competent designers might still be designing in. Nobody else who posts here seems to have used it for many years.

Since you seem to be the only person around here who still believes that the 555 is "recommended for new designs", based on your unique talents as a a circuit designer, you'd be a better candidate any Chicken Little award.

John

You would like to think that, wouldn't you. You may dream that you may be able to persuade other people to share that view, but it's just a dream John. Time to wake up to reality.

Of course not. That was covered in the previous paragraph, to which you provided an evasive and unconvincing response.

Since "Texan" is not a racial designation, the observation cannot be racist. Considering the enthusiasm with which you cast aspersions, the insult is one that you have earned, so it isn't gratuitous. And I certainly haven't side-stepped the issue of your incapacity to say what you mean, or failure to understand plain English, both of which you managed to demonstrate yet again.

-- Bill Sloman, Nijmegen

Unlike Sloman.

--
Politicians should only get paid if the budget is balanced, and there is
enough left over to pay them.

Bill Sloman wrote: : John Fields wrote: : > BillSloman : >

: >>I know you don't have a high opinion of my expertise - you mostly lack : >>the wit to know what I'm talking about - but I'm curious as to what : > >> --- : > >> If you have to ask, then there are two. : > >> ---

Bill, please just ignore JF. I'm sure I'm not the only one here who appreciated your older habit of (mostly) refraining from personal insults or splitting hairs, and rather concentraing on real argumentation. I have appreciated that, whether the subject was on- or off-topic, and regardless of whether I agree or disagree with you. There are many other such personalities in this NG.

Then there are a lot of uninteresting barking dogs, proliferation of whose posts was obviously the trigger for this very thread, too. If you keep barking back you risk starting to look similar.

Now that we're here: what you think of the paper by Giusi, Crupi and Pace in the 2008 Rev.Sci.Instr.? If I understand it correctly, they are using the trick of cross-correlating the signals through two different amplifiers (it think van der Ziel introduced this method originally) to get rid of the amplifier noises, but they claim to do it for voltage and current noises simultaneously. This appears fishy to me.

Regards, Mikko

In , Bill Sloman wrote in part:

I use the 555, usually favoring the Texas Instruments TLC555.

I even used National's LM555 as a MOSFET driver before MOSFET drivers became inexpensive.

--
 - Don Klipstein (don@misty.com)

Do you have a link for it? I suppose that you could do that by using N pairs of amplifiers, with resistors between the pairs' inputs and the unknown, like this:

In |\

0-*--R2R2-*--| \ | | | >------0 Out1 | | | / | | |/ | | | | | | |\ | *--| \ | | >------0 Out2 | | / | |/ | | | |\ *--R2R2-*--| \ | | >------0 Out3 | | / | |/ | | | |\ *--| \ | >------0 Out4 | / |/

With N pairs of amps, you have 4N+1 free parameters (source noise plus (for each pair) resistor noise, each amps' own voltage noise, and the pairwise sums of their current noise).

The same N pairs will have (1/2)N(N-1) cross-correlations, so the problem becomes soluble when

N**2-N > 2(4N+1)

i.e.

N^^2 - 9N - 2 > 0,

so you need at least 4 pairs (8 amps).

I'm doing something like this in a more deterministic situation to get rid of sensor errors.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

All that work, followed by a face plant. 10 pairs of course.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

If you want to work at the cutting edge, be sure to have plenty of Band-Aids around ;-)

--
"For a successful technology, reality must take precedence 
over public relations, for nature cannot be fooled."
                                       (Richard Feynman)

There's still a place in the world for monostables. As long as you don't care if the timing is off by a factor of 2 either way, they work fine. And obsolescent or not, the 555 will very likely still be available after today's latest magic micro is gone. It also doesn't need programming support. Monostables are good for pulse stretching, for instance, where you don't know when the pulse is going to arrive with respect to the uC clock.

I recently did a laser locker (using both current and temperature tuning to get wide range and good bandwidth) that was almost all analogue, including two feedback loops, laser and Peltier current drivers, and a triangular sweep for lock acquisition. It does need a small micro off on one corner to mind the store, but nothing more than that. You turn the thing on, and a few seconds later the laser is in lock, with a linewidth that should be less than 100 Hz.

If the projected volume were greater, it would be worth implementing some of that in a uC, but nothing like all of it.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

How about projects that are simpler and where parts count and cost is important almost to or outright to pennies?

Make something work with one implementation or another of one or two

555s and lack of a microprocessor, and it generally costs less than using a microprocessor. Much of the time, the story is the same if the alternative to the 555 is an op-amp or basic logic ICs. 555s are also smaller and have fewer pins than logic ICs generally have, and it's my favorite Schmidt trigger inverting buffer when I only need one rather than 2-6. It even has a wider and more predictable hysteresis range than the 40106.

--
 - Don Klipstein (don@misty.com)

A delay line is better, if the pulse stretching you want fits into a delay line you can make or buy.

off

Peltier coolers have the nasty habit that the heat transfer that you get - in watts per unit current - depends on the temperature difference across the Peltier junction. This is usually easy enough to manage in a the digital domain, but trickier if you confine yourself to analog. I've published this point in Rev. Sci. Instruments.

Sloman A.W. =93Comment on =91Implementing of a precision fast thermoelectric cooler controller using a personal computer parallel port connection and ADV8830 controller=92[Rev.Sci. Instrum. 74, 3862 (2003)]=94 Review of Scientific Instruments, 75 788-9 (2004).

Sure. My own take on driving a Peltier cooler was published in Measurement Science and Technology, which Americans rarely read. It is cited from time to time.

Sloman A.W., Buggs P., Molloy J., and Stewart D. =93A microcontroller- based driver to stabilise the temperature of an optical stage to 1mK in the range 4C to 38C, using a Peltier heat pump and a thermistor sensor=94 Measurement Science and Technology, 7 1653-64 (1996)

an

I believe you. I just don't believe that there are actually that many situations where the 555 is a good choice. Microprocessors aren't the only way of implementing a - partly - digital solution. Programmable logic parts have long since gotten big enough that you can do useful stuff in a single programmable part.

-- Bill Sloman, Nijmegen

--- Yup, I'd be willing to bet that Dr. Camenzind is flabbergasted, and immensely pleased, with the ways his baby has been put to work.

One particularly cute application which passed through here a while back was with someone who was looking to drive an audio transducer at a higher level than could be afforded by a single transistor doing Morse "H" forever, and someone suggested using a couple of 555's with their totem poles arranged as a full bridge, thereby driving the transducer with about about four times the power available from the single transistor, at the same supply voltage.

I remember that he posted back, not quite awestruck, but certainly quite impressed with the performance increase he enjoyed from his transducer for certainly less than a dollar's worth of parts.

--- JF

Ok, so lets stop using resistors, caps, potentiometers and so on.

Sounds ridiculous doesn't it? Just like your comments.

Did you ever stop and think that maybe the deficiencies you see are actually attractive features to real designers/Engineers?

Have a good day, you old goat.

Why? What's the performance gain, in exchange for using a more expensive, less general, and less common part?

I can't remember ever having designed a 555 into any actual product, but I've often used them when hacking something together in a die cast box for lab use. Quick, simple, no problems. Some of those boxes have been in use for 20 years, off and on, with various modifications along the way.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

t r . k t

ng,

A monostable is a circuit that produces a slowly rising ramp and compares the ramp voltage with some kind of reference voltage. Your monostable period ends when the comparator fires (for the first time). Any noise on either the ramp or the reference adds jitter to the period. Currents circulating through your ground and power supply rails have a nasty tendency to contribute to that noise. Careful layout can minimise - but not eliminate - this. Delay lines do tend to degrade the edge speed of transitions launched into them, but you still tend end up with much higher volts per microsecond that you get with a monostable's ramp.

een

way.

A legacy part in a legacy design. Minimal expenditure on design, poorer performance than a more thorough approach would have yielded. Doesn't make the 555 any less obsolete.

-- Bill Sloman, Nijmegen

Who cares? That just makes the trailing edge a bit jittery--but generally nothing like as bad as you'd get by reclocking it digitally.

I associate the word 'legacy' with happy events such as long lost uncles leaving me a lot of money. I've never understood the current fashion for using it as a swear word.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

g

han

If you'd read his book - which you can now download for free

you know that chapter 11 discusses the 555 in detail, and how the circuit could be improved. He's certainly proud of the design, but he's - tellingly - particularly proud of being able to fit it into an

8-pin package, which was the primary contributor to its commercial success.

Another of the 101 things a boy can do with a 555. I doubt if this was the only way of solving that poster's problem, and I'd be surprised if it was the minimum solution, but since you haven't specified the problem solved in any detail, this is pure speculation.

IIRR the 555's totem pole output doesn't get all that close to the positive rail, so "four times the power available" may be something of an exaggeration.

-- Bill Sloman, Nijmegen

--
Your point being???

two

sing

lso

y

e than

Hans Camenzind's attitude to the 555 is a matter of public record. In his book he doesn't admit to being either "flabbergasted" or "immensely pleased". We've discussed his book here earlier, and you should know better than to parade your guesses as to what he he might think about the design when you should have found out what he actually thought back when he wrote the book.

Only if I washed my hands afterwards.

Since you clearly don't understand 90% of what I post, it won't signify much to you. A tale told to an idiot who doesn't appreciate what it might have signified to a better informed audience.

We don't even have the rail voltage.

With a single 555, the load can be returned directly to the postive rail, and will see the rail voltage minus the saturation voltage of the output - about 0.1V at 5mA.

With your "H" bridge, the load has to be connected between two totem pole output stages. The low side will still be only about 0.1V above the negative rail, but the high side will be around 1.4V below the positive rail. With a 5V rail, you would not put four times as much power into the load as the single-ended driver but only about twice as much.

And "about" implies variation above and below an estimate. Four times the power is the upper limit to the advantage, available only with an infinite supply voltage. You could have legitimately claimed an advantage *approaching* a four-fold increase in power, but that sort of careful use of language is quite beyond you.

-- Bill Sloman, Nijmegen

Thank you for giving a reasoned reply. I would like to ask about one part of it: "The catch with the 555 is that it combines a part nobody really ought to use these days - a monostable - with a rather poor quality bipolar power transistor, and uses the same ground return pin for both devices."

I do not understand why you make a blanket statement indicating that one should not use a monostable. If a monostable does the job, why use something else? If you've got a money making idea that requires nothing more than a 555, would you reject it because it contains a monostable? Would you insist on something other than the 555?

Ed