Granted, I think auto hold will probably completely replace the functionality I would get from regular hold and add some, I just wish I knew whether or not the regular one was available.
Not sure if display hold is available outside of the min/max mode, but I've never seen the point of display hold. If I had my hands and eyes available to press a button, I wouldn't need it.
Next, I'm a little confused about the difference between using the frequency feature while in DC vs AC. The full manual gave some situations when you would use which, but it isn't super clear what the real difference is, and it's a bit confusing that there is a frequency measurement in the DC mode.
Table 6 of the manual sums it up pretty well: AC uses a trigger level of about 0V (slightly higher since RMS converters are non-linear close to 0V), DC uses a positive trigger level (10-30% of full scale). AC frequency measurements won't work for signals with a DC offset (eg. pulse train), DC may have issues with small AC signals.
I've always been a little irked that DMM's only give you rms for AC. Granted rms is the most useful, but peak voltages can be useful and some cheap meters actually calculate rms by taking peak/?2 so clearly a peak measurement isn't a hard thing to implement along side rms.
I believe this is incorrect, the cheap meters are average responding, which means they calculate the RMS value (for pure sinusoidal signals) from the average value of the rectified signal.
This Fluke article (obviously completely unbiased) seems to confirm this.
On this note, I had an idea. I figured that I could use my 87V to measure the peak voltage of an AC signal by measuring it in DC volts min max avg mode using the faster 250 microsecond time scale. I was under the impression that min max avg just did the corresponding measurement once every 100ms or 250µs and then extracted the data and DC would take literal voltage measurements.
My impression is that the normal min/max mode takes a sample every 100ms, but although the peak mode can detect peaks of at least 250us, I don't think it samples continuously at that rate, which is why it can't calculate an average. The fact that the peak needs to be repetitive for it to capture 250us pulses hints that there's some sort of equivalent time sampling going on. They probably use something like an analog comparator set to trigger at the previous peak level to trigger the measurement, and need >> 250us to sample and re-arm.
So I try it out on a wall socket. I get ~118-119 rms and a max of about 162 (the min was negative and a little smaller magnitude). This was pretty close, but not quite the ?2 factor I expected. So either the signal isn't quite a sine or this idea isn't working as a peak measurement.
This could easily be distortion, loads like bridge rectifiers tend to draw the most current at the top. Look at the 'how clean is your mains' thread from some time ago for some examples.
I did some reading...and I ran into some information that suggested that the PEAK min max was actually meant for just this, measuring PEAKS (duh!). I had thought peak was just a new word for fast, as in peak acquisition speed. To test this, I tried the same thing but using min max avg in AC mode. Sure enough, regular min max avg gave the rms min max avg, but when I went into peak mode it jumped up to ~162.
Again this is confusing since I would expect the AC peak min max avg mode to just do faster data on the rms value, though maybe the term rms doesn't make sense at those timescales (granted I don't see how it would make sense at 100ms then either).
I believe you are correct, peak min/max is intended for instantaneous(ish) values, so no averaging (which RMS is) is taking place. Min/max samples at a much slower scale than most AC signals, so it has to use some sort of average. Averaging instantaneous values wouldn't make much sense, since that would be 0V by definition for a pure AC signal, so they have to use some sort of RMS conversion for that. But I haven't seen this documented, and I haven't done tests to verify this. You could use a function generator, and set the frequency proportional to the 'sample rate' (eg. 100Hz for peak min/max, 1Hz for normal min/max) to test this. If your hypothesis is correct, you would expect lower values (compared to the 'average' AC value) with min/max, about sqrt(2) lower.
So what exactly is peak min max avg and how does it differ between DC and AC?
No difference between AC and DC I think, although I would rather consider DC very low frequency AC, since there's no such thing as a peak DC value (that would be an AC component on top of the constant DC value).
On a similar note, I'm also a little confused as to what the DC and AC measurements actually measure. At first I had assumed AC measured the true rms of the signal and DC literally measured voltage, so an AC signal would read quickly varying in DC mode and I could use the avg feature to get the DC offset.
If the signal is slower than the sampling rate (say .4Hz), this would be the case.
After reading some, it appears that the AC feature measures the true rms of the AC component only (no offset).
Yes, it's usually AC coupled.
They say you can get a true total rms by taking the AC and DC measurements and doing the orthogonal square square root sum up, but this leads me to believe then that the DC function actually measure the DC offset of the signal.
It's intended to measure the DC component (although it also measures very low frequency AC, as above). For example, a 5V power supply with 1Vrms 120Hz ripple would measure 5V on DC, and 1V on AC.
So if you measure an AC signal using the DC mode does it actually give you the time average of the signal?
I believe so, which would be zero for pure AC. Not sure how they determine the average, it may just be a low-pass filter.
Also, I couldn't find much on the smoothing feature. I assume it smooths the signals in some way, like a watered down version of the low pass filter function, but it's functionality isn't entirely clear to me.
I don't think they give detailed specs, but I consider it a form of digital low-pass filtering, or something like a windowed average.
Home
Help
Search
Login
Register
