Harmonic Notch filter

Okay. This tells us that you wish to be thought omniscient.

** Nope.

It asks you to prove your pig ignorant, f****it assertion.

And that proof must be in the context of a typical ANALOGUE notch filter at

50 / 60 Hz.

Quote form the OP:

" The point here is to design a circuit not much more complicated than a normal notch filter. "

Say +/- 1% of the centre frequency over the 0C to 50C temp range.

To be very generous.

You have no hope - f*ck head.

... Phil

You asked me to prove my assertion - which I did, despite the fact that everybody knows that the mains frequency mves around a bit - and now, rather than arguing with any of the points I went to the trouble of making you, just re-assert your original - ill-founded - claims and throw in the extra additional - equally ill-founded - claim that I'm lying. Is there some kind of competion running in the user-groups for the position of Irrational Sociopath for 2011?

Any more of this and people will start confusing you with krw.

-- Bill Sloman, Nijmegen

If astronomers had any brains, they would have used an electronic motor drive that was crystal derived.

--
You can't fix stupid. You can't even put a Band-Aid? on it, because it's
Teflon coated.

In the old days you couldn't buy an electronic motor drive that was crystal controlled. Astronomers are exactly the kind of people who would have built their own as soon as it got to be practicable.

-- Bill Sloman, Nijmegen

"Bill Sloman" "Phil Allison"

You asked me to prove my assertion - which I did,

** You links proved the exact opposite !!!!!!!!!!!!!!

" The frequency has so far rarely deviated more than 0.2 % from 50 Hz, i.e. it was almost always between 49.9 and 50.1 Hz. "

- rather than arguing with any of the points I went to the trouble of making you, just re-assert your original - ill-founded - claims and throw in the extra additional - equally ill-founded - claim that I'm lying.

** Course you are lying.

You tell lie after lie after lie.

Deliberately posting wrong and refusing to ever back down IS LYING.

.... Phil

Astronomers have a *lot* more brains than you do!

Modern scopes are mostly servo controlled with shaft encoder feedback under computer control these days - true even for high end amateur scopes. This allows them to automatically slew to targets, correct for the change in apparent motion with altitude and accept autoguiders.

Industrial stepper drives were used for a while in the late 70's, but tended to cause unwanted structural vibration even with microstepping compared to old synchronous clock motors driven from a sine wave.

But for a very long time cheap and ubiquitous synchronous clock motors were the way to drive telescopes. A few customers even had Rolls Royce cut large 359 tooth gears for the main drive so that their scope would keep almost perfect sidereal time if the mains frequency was at exactly

50Hz. They were the ones that were most miffed about the higher and variable mains frequency at night. This included an optical telescope by Grubb Parsons that at one time in the 1960's was at Jodrell Bank.

It was definitely a step up from the governed clockwork drives that preceded them and were very sensitive to balance and temperature.

Regards, Martin Brown

I once saw a telescope drive made from a clock movement, for some reason, the balance wheel was exposed, so you could see it going tick-tock...

Cheers! Rich

e.

You may think that this is a negligible variation, but you'd neglect it at your peril.

Except that I'm not wrong. We differ on what constitutes a significant change in frequency, but this isn't a matter of fact, but of opinion, and my persistence in holding to my - well-founded - opinion isn't deceitful.

-- Bill Sloman, Nijmegen

Don't bet on it, since I built the reievers used for radio telescopes.

There were tube based power oscilators in the '40s that would drive a syncronous motor, and could be crystal controlled, so it's taken you 70 years to get it right?

--
You can't fix stupid. You can't even put a Band-Aid? on it, because it's
Teflon coated.

I think you are seeing the drift of the crystal which drives the DDS. I once made a 0.5ppm DPLL to track public E1 clocks. Just blowing some air at the board caused the frequency to change because of the temperature change.

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

George isn't seeing the drift of the crystal reference. He's referring to a much larger drift.

I just performed the same experiment with a function generator. If I set the generator output to 10.000000 MHz and zero beat with WWV, the function gen is

120 Hz low, an error of about .001%.

Setting the DDS output to 60.000000 Hz, and watching that output, plus the grid voltage on the scope, the grid drifts first high, then low. At any time the DDS output frequency can be adjusted to cause the grid waveform to remain stationary (for a while) with respect to the DDS. The grid frequency can then be read from the DDS display.

I found the grid to typically be off by +-.01 Hz, but as much as +-.05 Hz during about 5 minutes of watching. This is an error of .016% to .083%,

The oscillator in the DDS gen is much more stable than that. When I was listening to the beat note between the DDS output and WWV on 10 MHz, a .01% drift would have resulted in a 1 kHz change in the audio beat note. No such large change was occurring; in fact, I could detect no change in the audio beat note frequency by ear.

For which observatory and wavelengths? Front end or IF stage?

Most of them have in significant house capability for design of the main amplifier and correlator electronics. They tend to use mass produced dishes and then add custom interchangeable front ends.

We did aperture synthesis in permitted bands from 151MHz up to 31GHz.

I already told you synchronous motors got phased out around the 60's and

70's in favour of steppers. But computer controlled servos and before that in some sites paper tape controlled tracking was used.

Regards, Martin Brown

We built complete, custom microwave receivers for several, but they didn't tell us the customer's names. They used a LNA at the feedhorn and fed the output to our equipment. We were rarely told who the customer was, unless it was a turnkey package or a rush order when they would tag something to give it the highest priority.

What I described was for those motors, before a computer was available.

--
You can't fix stupid. You can't even put a Band-Aid? on it, because it's
Teflon coated.

So the first IF gain stage receiver at the dish after the cryogenic LNA?

What a strange place you worked.

The only production slots in our factory that did not have a customer name on them were the ones being built on spec to avoid line stalls.

I can't see NRAO subcontracting out microwave receivers without going to the suppliers to acceptance test them (and I certainly can't imagine them paying extra for quick delivery - less for slow delivery maybe). The VLA would want sets of 30 which would be a bit of a give-away.

For a price that was nothing like justifiable for the task. The old synchronous clock motors were good enough. The mechanical gears caused periodic errors that were way bigger than systematic mains drift.

But the fact remains mains drift was noticeable on a cold winters night.

Regards, Martin Brown

ed text -

Well it could be, you can try it. The frequency first is up above 60 Hz, then below for a while, then above. (This was done with a Keithly

3390 function generator.) If I have a chance, I'll take some data and write down numbers.

George H.

s n . r e

to a

et the

n is

e grid

e the DDS

tionary

read from

during

s 1%

such

o beat

Thanks Phantom, (I stopped keeping track of this thread once it sunk to the flame war level.) That's basically what I say. (I didn't have any outside checks on the

10MHz though.) I remeber being confused at first why it wasn't exactly 60 Hz.... and then I watched it change.

George H.

It was at whatever frequency the customer ordered. I didn't work on many tuner modules, but I did on the low phase noise synthesizers they used.

It was an engineer to order company. Can't you wrap your head around that? There might be 200 separate work orders for a customer's customized equipment. Why did the production and test departments need to know, or would even care who the customer was? Was management to put fake names on orders for certain government agencies? The only jobs I knew the names of the customer were the single receiver built for NASA for the International Space Station. The complete turnkey earth station for NOAA for their Wallops Island facility. The pair of earth stations for the European Space Agency, one fixed site and the other mobile to monitor their launches.

Sigh. Sometimes a bid would be open for six months or more, while the customer waited for funding. Normal time from order to at their door was 45 to 60 days. Sometimes they had less than 30 days before the funding was awarded and would be lost. Since we had base models, we could usually pull something that was ahead of schedule and add the custom modules to ship it on time. It wasn't uncommon to have the head of production, test or engineering show up at my benches and ask me to do a special job in a hurry, since I loved the dirty jobs, and rush orders to keep from getting bored out of my mind. Your mistake is thinking of a production line instead of building specialized products that were each customized for each customer.

NASA, NOAA and the ESA didn't come to the plant to inspect the equipment, so what makes NRAQ so special? Some of our equipment was used by NASA to track deep space probes and at least on receiver was online for over 30 years with no repairs. Early delivery bonuses were common. One was a half million dollars if we could ship in 30 days instead of stated 60 days. We made that bonus because I worked an 18 hour day and cherry picked boards instead of repairing any that failed. I finished 72 boards that would normally been about 18 for that extra 10 hours. Then I finished the ones that failed to meet specs the next week.

Sigh. The system I described was independent of line frequency. If they couldn't afford a couple hundred dollars, they should have shut down.

--
You can't fix stupid. You can't even put a Band-Aid? on it, because it's
Teflon coated.