I believe the color subcarrier in a color TV is phase locked to the transmitted signal and, for network studio transmissions, is derived from a cesium clock. From what I have read it is more accurate than WWV and doesn't require extra equipment other than a TV displaying an image with a studio source. Frequency is 3.579545 MHz.
No. One part in 100 million (10^-8) is 100 times better than wwv (10^-6). There is lots of excellent equipment on ebay that can take advantage of this level of accuracy. The main point is that you do not have to think about it very often. The low cost counters that have uncompensated or poorly compensated timebases are basically useless for any serious work. The other nice part about the high stability references is that you can distribute it to all the synthesizers on your bench and everything is coherent. Of course it depends on what you do. For my ham work, one ppm is fine. I do other work where the Rb source is not good enough.
I've seen that discussed elsewhere, and although it would take me a week to find the particulars, (1) the frequency can be off as much as 10 Hz by FCC standards, (2) from what I've read it's frequently off by more than that, even on network feeds, (3) IIRC they don't even use the good clocks on the networks any more, (4) NIST clocks are going to be a couple of orders of magnitude better than the best a network would buy for the purpose of meeting FCC regulations, (5) IIRC the frequency should actually be 3,579,545.454545454545..... Hz, and (6) Doppler shift on the incoming television signal could potentially cause the subcarrier frequency to vary up and down.
--
"...global warming is an apocalyptic faith whose preachers demand sacrifices
of others that they find far too painful for themselves."
-- Andrew Bolt, in Australia\'s Herald Sun
That depends a lot on your definition of "serious". There are lots of things where just being within 100PPM is more than good enough. RS232 is ok up to 5% error. If the so called 60Hz in your motor home was actually
59.9Hz, I don't think you would mind.
A lot of them have worse short term noise than a good OCXO.
?Ref=3107&Site=US&Cat=34342147 "Your search criteria has expired" Furthermore a search on "34342147" (no quotes) gets zero matches. A search on "3107" (no quotes) gets matches that are not better than 1%. Strangely enough, a search on "precision resistors" (no quotes_ is as bad. Worse, a search for "resistors" and wading thru the various types gets *at best* Chip Resistor-Thin Film(67311 items) with 0.02% as the best or tolerance listed. So...... Where are those mysterious 0.01% resistors???
Exactly and completely correct in all aspects. I made a 0.1% resistance reference box: 100 ohms, 1K, 10K, 100K, 1M,
10M and 100M that has been invaluable. I made a voltage reference box using an Intersil (was Xicor) 5V FGA reference powered by a 9V battery; good for source and sink and that initial accuracy of 0.5mV was hard to beat; my HP 5326B verifies the value within its accuracy as well as the 0.5mV of the reference; both in the same region of fuzziness - so not too bad. My handheld DVMs are "fair"; actually the 3.5 digit one os more stable and reliable in readings than the 4.5 digit. That one can be set by only one pot which either makes the resistor readings within spec or makes the DC readings in spec - but not both; i opted for DC reading accuracy. I hate it when i have to fiddle with what seems to be a perfectly good meter, just to make it read correctly (based on two other references). One of these daze, i may be rich enough to get a Fluke 88845A; and if
*really* rich, will pay for a traceable meter!
I just got done calibrating a AM503 & A6302 current probe / amp somebody took a screwdriver to. Without the manual and all required gear (PG506), and cal fixtures it would never have worked properly again. I work in a cal lab and the best part about iso9002 was requiring the sealing stickers (cal void if seal is broken). We never used them prior to iso certification.
Forgot to mention : If you want to cal your own gear, mark any pots / vari-caps and write down any software codes BEFORE changing. Do not adjust the compensation capacitors in any Tektronix attenuators without a PG506 and a procedure.
Jim Yanik - This note isn't for you, as you have seen the damage too.....
The real question is how much precision do you really need in the home "lab"? How often have you needed to use your DMM with how many
*accurate* significant digits? 100 minus some
*very* small percent of the time, 2 significant digits is all you need. Do you _really_ care if your 5.055 volt reading is really 5.06 or 5.04?
Oh hell yes, I want to puff out my chest like everyone else and think I have *accurate* equipment.
But I'm curious as to what home circuits need meters that can read voltage accurately to 3 decimal places?
2 decimal places? The question for current measurement: in what home brew circuit design/troubleshooting do you need accuracy below the tens of mA digit ? *Need*, not
*want*. Do you even trust your DMM on an amps setting for those measurements, or do you measure the current indirectly? How about ohms? Would you trust any DMM, regardless of who calibrated it, to measure down in the miliohm numbers?
To me, the design of the circuit being mesured has to take care of all of that crap. If it is so poorly designed that a 10 mV departure from nominal (that is missed by my innaccurate meter) will keep it from working, that suggests other problems. Yes, the home "lab" person wants extreme accuracy to as many decimal places as he can get. But when does he ever really need it?
None of this is to argue against having the best instrumentation you can afford, or references to check it against, or paying for calibration and so forth. But for myself, I need a dose of reality from time to time when I start drooling over some accuracy specs that I will never need at home. My bet is that most of us are seduced by that same muse.
You surely didn't mean tens of _mA_, did you? I build stuff with PICs as you know, and some of it is designed to run on batteries and needs to go for long periods of time unattended. The current draw for a 12F683 running at
31kHz is 11uA, sleep current is 50nA. If I could only measure current to "tens of mA", I'd never know if the PIC was setup right for low current draw and I certainly couldn't have any idea of expected battery life. I wouldn't even know if it was sleeping until it ate thru some batteries in a few days instead of six or eight months. I think I have a need to measure fractions of a uA.
When he needs it he needs it, what can I say? Do I really "need" a new DSO? Well I've managed to get by all this time without one, so maybe you think I don't really "need" one. I see it like this though, I don't get allot of time to tinker anymore. I'd like to spend it more productively. Instead of fumbling around and trying to devise silly methods to make my existing equipment do something it wasn't designed to (like going off on a tangent to build a PIC circuit that will trigger my scope early so I can try to see some pre-trigger history).
You may, but not accuracy below the tens of _mA_ digit. When you need accuracy below tens of mA, you measure voltage across a resistance. It doesn't make a lot of sense to look for your meter to be accurate to 8 decimal places for your .00000005 amp reading.
Here's how you do it with accuracy at the tens of _mV_ digit:
For 11 uA, put a 10K .01% resistor in series with the supply and measure .11 volts across it. The voltage would range from 0.109989 to 0.110011. Keep only
2 decimal places. Your computed current, worst case, would be off by 1 uA
For 50 nA, use a 2 meg 1% resistor and measure .10 volts across it. The voltage would range from .099 to .101 taking the 1% into account. Throw out the last digit. Your current computation would be off worst case, by 5 nA.
With a voltmeter accurate to 2 decimal places. I don't know why you would
I asked, looking for concrete cases. Your case with the PIC is an excellent example of when a person needs to know about really small currents. It definitely fits into the difference I had in mind between "needs" and "wants". But it does not mean he needs accuracy out to 8 decimal places. He needs it to
2 decimal places, as was shown. Three decimal places would be nice. :-)
I have no opinion on that, and it would be irrelevant if I did. I don't know what your situation is.
Well, you're free to argue against having the best instrumentation you can afford, or having references to check it against or getting it calibrated or whatever, if that's how you feel. I tend to err on the side of wanting the best even when it is not the best fit for what I really need.
Here is a good "trick" to measure low currents with your DVM. Make a switchable shunt box with (at least) the following full scale ranges: 200nA (shunt resistor 1.11 megs), 2uA (shunt resistor 101K),
20uA (shunt resistor 10.0K), 200uA (shunt resistor 1.00K). Put a twisted pair of leads (red, black) with banana plugs (red, black) running out of the box via a small grommet, to plug into your DVM set to the 200mV scale; a pair of (red, black) banana jacks with 0.75 "spacing is mounted on the box for your test leads. Hint: add to the legend the parallel resistance of the system (200nA/1M, 2uA/100K, etc) as a reminder of the resistance of this current meter scheme. Added hint: the 200MV scale is good for 20nA full scale, just remember the meter resistance is 10 megs.
Now come on, the 8 decimal places is only assuming that the scale is in an Amps range. The meter would be in the 500uA full scale range where 50nA is only 2 decimal places.
Those are fine ways to measuring static current levels, but they will not work for me. Until the PIC goes to sleep, the current draw is much higher. So much so that it would never power up thru a 2M resistor.
If your volt meter has a 1V maximum at full scale and one can live with 10% error, then I agree. If it has a 100V range, then you need .01% accuracy on your equipment to make your measurements, right?
Aren't you arbitrarily relocating your base measurement scale to uA or nA and then claiming that you're only being accurate to two decimal places? You are still measuring current to the same "8 decimal places" in terms of whole Amps, you just moved the decimal around.
IMO, it's not about decimal places at all, that's just a matter of scale. It's about accuracy. 10% ain't good enough, and that's only accounting for the error in your shunt resistors. :-)
Actually, I'm all for that part.
And this is what I do as well. I'd rather have a margin of overkill than to be constantly living with sacrifice by saving a couple of bucks on a NRE. What I wasn't "sure" about was whether _you_ really felt that way. ;-)
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.