EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: Electro Fan on June 09, 2019, 08:06:46 pm
-
I'm using a 5315A to measure the 10 MHz signal from a GPSDO. (I of course have no highly definitive way to know if either the GPSDO or the 5315A are accurate, but let's set that aside for a moment.) The 5315A has the C input. On the C input with the gate time knob pointer turned straight up (12 o'clock), the display shows 10.000000 and a 6 for exponent - which I'm pretty sure signifies 10,000,000 Hz. All good.
What I'd like to do if possible is give up the 1 for 10 million and have the counter tell me about tenths (or hundredths, etc.) of a Hz. In an effort to do this I've turned the gate time knob fully clockwise (until the pointer stops at about 5 o'clock). At this point the 1 goes away and is replaced by a solid round LED indicating OVFL, and the display reads 0.0000002 and a 6 for exponent - which looks a lot like 2/10ths of Hertz on the end. Any chance that I am interpreting this correctly and that the 5315A display is signifying 10,000,000.2 Hz?
-
Just found this post by Dave from about 9 years (after ready fire aim, ie ask questions then search):
https://www.eevblog.com/forum/projects/extremely-accurate-frequency-counter/msg1901/#msg1901 (https://www.eevblog.com/forum/projects/extremely-accurate-frequency-counter/msg1901/#msg1901)
- this looks like it could be an answer to the problem.... going to see if I can figure out how to get period mode to solve the problem... but still up for seeing if anyone can confirm my interpretation above. Thx
-
In period mode on the A input, with the level/sens dial pointing straight up (12 o'clock) the display bounces between 99.99999 -9 and 100.00000 -9. Any chance this means something.....useful?
-
Any chance that I am interpreting this correctly and that the 5315A display is signifying 10,000,000.2 Hz?
Yes. That's exactly what it means but keep in mind that that might be beyond the accuracy of the counter so it may not really be useful.
-
In period mode on the A input, with the level/sens dial pointing straight up (12 o'clock) the display bounces between 99.99999 -9 and 100.00000 -9. Any chance this means something.....useful?
First, the position of the knob isn't calibrated so that doesn't really tell you anything. Second, yes the counter display is rounding down and then rounding up so the counter thinks that the input is between those values. Again the counter may not be (Probably isn't) absolutely accurate. If you had an input that you could vary the frequency slightly then you would be able to see the ratio of round-ups to round-downs change so the counter can see the difference but I don't think that that change is display is really useful, particularly on a counter that has lot of displayed digits. If you had a high accuracy counter that displayed only a few digits then the change might tell you something but not on a counter such as this one.
"Any chance this means something.....useful?"
Not on a counter with unknown accuracy.
Keep in mind the old high school math lesson that writing down a lot of digits in a number doesn't necessarily mean that that number is accurate to that many places.
A few months ago I picked up a whole pile (8) of EIP 548A 12 digit counters with the high stability oscillators in them. It was quite interesting to couple them to each other and to my HP and Racal Dana counters and see how they compared to each other. Most agreed very closely but a few didn't. I was pleasantly surprised to find that about 1/2 of the EIP counters agreed with each other and to my option 004 HP 5343A out to the 11th digit. However that still doesn't mean that they were absolutely correct. No matter what source I used, you could see that each counter had slightly different drift and that the source also drifted over time.
Keep in mind that each counter and the signal source each has its own time base and that all of the time bases are drifting and at different rates. With just one source and one counter you can see the drift but you don't know which one is causing it. When you hook up six or so counters to the same source, you can start to see how each time base is drifting over time.