This was a supplementary to the aircon question.
Can anyone recommend a plug in power meter which is reasonably accurate?
Amazon offers gadgets around £10-£50 range but it is difficult to see if you get what you pay for.
Cheers
Dave R
The transient current when the air-con starts will probably only last for a fraction of a second - but if your power supply *can't* provide that peak current, it ain't gonna start.
I'm not at all sure that you're going to find a meter which has a fast enough response to see the transient. Certainly none of my meters - both analog and digital - are likely to be any good.
One other instrument which I've got is a hand-held digital oscilloscope. If I needed to do it, I'd power the air-con through a suitable shunt resistor, and connect the scope across the shunt on a voltage range. I would measure the steady running current using a clamp meter, and then use that to calibrate the scope. Then I'd stop the air-con and re-start it, and the scope would hopefully show just how big the transient is.
I've never actually done this, so others may blow holes in it - but it seems to me that it would have a fighting chance.
Thanks - it has morphed into a more general query which is why I separated it from the aircon thread.
It did occur to me that what was needed was not a "read the display" but something with a more or less continuous output to a computer for later analysis. So that peaks, troughs, spikes etc. could be graphed later.
I have no idea if there is such a device.
In more general terms I would like to find a reliable plug in (13A socket) monitor to check everything from the dreaded MH to standby on audio-visual kit and computer consumption.
I do have a clamp meter on the main power feed which has a remote monitor with USB (from Eon) but I was hoping to find something for individual sockets.
Cheers
Dave R
Many moons ago there was a device for a Spectrum, driving a tandy plotter via the interface 1 rs232 that had a shunt device couple by an opto isolator that showed peak current and the waveform. It was in Electronics today international I seem to recall. Course in those days elf and safety was considered common sense... :-) I think though that the shunt needs to be a buig un so it does not heat and cause changes in resistance in the short term to be accurate.
Brian
Fair comment. Let's do the sums.
Let's suppose that we're looking at a device which uses 2kW in the steady state. At 250v (to keep it simple), that's 8 amps.
You could probably get away with a 1 volt drop across the shunt. That would require the resistance to be 1/8 ohms, and it would dissipate 8 watts.
I'm not sure that you'd easily find anything 'off the shelf' to do that, but it would be relatively easy to create the shunt from a suitable number of short lengths of copper wire, connected in parallel.
Many years ago, I made a series of shunts in this way to enable a cheapo multimeter to measure automotive-type currents of up to 30A.
We did something similar once when wanting to assess the inrush on a
10kW transmitter. Used a clamp meter round on of the incomers. It was a reasonably posh one that also had an external trigger output. So fed this into a digital storage scope. Did the job quite nicely. (we had to do multiple power on cycles to capture the worst case - which was when switching coincided with switching near the peak of the mains waveform).It nicely demonstrated why sometimes one or more of the three 45A MCBs built into the amp would trip on power on sometimes - the peak inrush was way more than the 5x allowed on a type B. They had to swap them for either type C or D versions.
Why not use a current transformer rather than a shunt resistor? That way you don't have to electrically tap into the circuit of the device under test.
Wouldn't a zero-crossing switch or relay have done the job, or at least made the inrush consistent?
Yes there are, but they aren't aimed at the low priced consumer market.
'Cos I was using mine on DC.
It would, but this was a control system refit on what was already a mostly 40 year old design. So there was little appetite to make many changes to the power or RF side of the hardware even though all the control systems were becoming microprocessor controlled. Specifying a change of MCB type was an easy and low cost change. (in normal operation, these things would not normally be power cycled for extended periods of time anyway).
Here was the beastie, a Marconi HF1141 :
The main cabinet is about 6' tall (look at the set of 3 "normal" MCBs on the top right by the power switches to get a feel for the scale). The lump grafted on the top was a modern fully synthesized "digital drive" radio transceiver (all DSP based). The lump under it was basically a
400W pre-amp, some tuning circuitry (its a tunable HF amp), a 10kV step up transformer and a 'kin big valve!(for those familiar with the typical LC tuned resonant circuit used for radio tuning, you can see all the bits there - what looks like a coil of
15mm water pipe is the main inductor[1], and all the motors and gearboxes at the top of the bottom middle section are used to vary the two vacuum capacitors (in white cylinders in the middle). The caps were only a few hundred puff if memory serves, but rated for many kV operating voltage!). [1] The inductor also had a series of shorting bars that could tap it off at shorter lengths - these were operated by a bank of large solenoids under software control. Which all made a rather satisfying series clanks as they were operated in quick sequence to adjust the inductance.
Yup that does change things somewhat... perhaps I should have read the "automotive-type currents" bit a little more carefully ;-)
If you're trying to measure DC amps, a visit to a local flea market may prove fruitful. I bought a DC/AC clamp meter from "Maplin Man" for 8 quid (afaicr I blagged him down from ten quid).
The DC amps ranges are 0 to 200 and 0 to 1200 amps with the 200A range showing to tenths of an amp readings. However, when trying to measure the current drawn over the 12v and 5v rails of an ATX PSU, you need to double check the calibration since it can easily drift a few tenths of an amp over the space of the first few minutes after pressing the "DCA Zero" button. This drift effect can be ignored once you're dealing with tens to hundreds of amps such as when testing lighting load and generator charging and the obvious 'cranking amps'.
When scouting the various stalls at flea markets, you're likely to see the more common AC clamp meter type. Tracking down a DC clamp meter is a little more difficult. Obviously, you're looking for one with a digital display (which isn't exclusive to AC/DC clamp meters - there's plenty of AC only clamp meters with digital displays).
However, the stand out feature, if the labelling doesn't already make it obvious, is the presence of a "DCA Zero" button. Without such a button, there's no way to compensate the inevitable drift in the hall effect sensor for DC offset to zero the meter just prior to taking a reading.
Most of these meters use a PP3 battery so it might be wise to carry a fresh PP3 (and maybe some AA and AAA cells for the less common types that don't use a PP3) to test any prospective purchases. If the meter has a DC amps range calibrated in tenths of an amp, you could try the effect of the earth's magnetic field when you fully open the jaws and turn around. I get a half amp reading when I do this test on my own DC clamp meter. Alternatively, a small bar magnet would be useful aid to test/confirm the DC amps measuring feature.
HTH & HAND :-)
That LC resonant circuit is most likely the 'classic' Pi "tank" circuit. Effectively a resonant matching 'transformer' with the rather useful characteristic of an LPF. Were the two tuning caps still labelled "Plate" and "Load"?
Indeed it was... with the slightly less commonly seen added wrinkle of having both variable L as well as C components.
Quite likely - I can't remember for sure, I did not spend much time studying the circuit diagrams for the RF side of it - I was too busy finding bugs in the digital hardware, and rediscovering secrets of the fine details of the control interfaces to the analogue side.
(amazingly the chap who did the original RF design still worked there over 40 years later - but the spec he had produced for the hardware software interface (not surprisingly) had one or two gaps or errors in it)
That is some beast!
Compared to the one it sat next to in the rig room it was *tiny*
a 1MW HF Amp, it had a 200kW audio amp that was used as a pre-amp on the input stage ;-)
A pre-amp?? Wouldn't fly. I think you're describing an AM transmitter and the audio amp was providing the plate mod for the HF amp.
Yup, same design as used for some of the Voice of America transmitters.
Quite possibly - my poor description...
Accepting a voltage drop of 100mV and 1/80 R shunt would get you into more tractable territory. Most cheap multimeters have a dedicated shunt for 10A fsd anyway but how accurate they are on mains I don't know.
Most multimeters already have a suitable current range for 10A max.
If I were the OP I'd fit a Owl onto the main house supply lines and then work on reducing the household base load by finding the old modem PSUs hidden behind sideboards etc. The current clamps are not as accurate as a dedicated 13A plug device but they let you see the entire household consumption and seeing it will allow you to shave 10% of your electricity bill if you want to. Finding the power hogs is the first step. (for which you might also want one of the 13A monitor devices)
HomeOwnersHub 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.