UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Test Gear and Workshop Equipment

Notices

Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders.

Closed Thread
 
Thread Tools
Old 17th Jul 2017, 4:44 pm   #1
MrBungle
Dekatron
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
Default Homebrew AC voltage references?

I've got a DC voltage reference I knocked up from a REF02 and checked on a friend's calibrated 6.5 digit meter. This is being used to cal my equipment.

However this is DC only. Does anyone know a suitable method to calibrate or generate an AC signal of a known amplitude precisely (within 0.5% is really fine) other than comparing it to another meter which is what I'm doing at the moment?

This excludes spending lots of money on expensive Fluke calibrators even if I would like one
MrBungle is offline  
Old 17th Jul 2017, 5:07 pm   #2
Guest
Guest
 
Posts: n/a
Default Re: Homebrew AC voltage references?

Any proper sine wave AC source and then a 'perfect' rectifier (op amp and diode job), the DC will be root two bigger than the DC (or two times that depending on the circuit).
 
Old 17th Jul 2017, 5:19 pm   #3
MrBungle
Dekatron
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
Default Re: Homebrew AC voltage references?

That's a really good idea. I'll look into that. Thank you
MrBungle is offline  
Old 17th Jul 2017, 5:57 pm   #4
Guest
Guest
 
Posts: n/a
Default Re: Homebrew AC voltage references?

Let us know if (when!) it works, untested brain thought.
 
Old 17th Jul 2017, 6:01 pm   #5
chriswood1900
Octode
 
Join Date: Jan 2007
Location: Dorridge, West Midlands, UK.
Posts: 1,474
Default Re: Homebrew AC voltage references?

An interesting project but not a trivial one to do. Most AC calibrators seem very complex, but this varies depending what you want to cover i.e. voltage range frequency and accuracy. To generate 500v AC at 20khz accurate to 30uv is no small task and that is why calibrators cost so much whereas, 5V RMS at 100hz could be done relatively easily.
I'm not sure if you spotted this but it gives a few starting points.
http://www.nutsvolts.com/magazine/article/the-ac-volt
Good luck
chris
chriswood1900 is online now  
Old 17th Jul 2017, 6:27 pm   #6
MrBungle
Dekatron
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
Default Re: Homebrew AC voltage references?

I've done some LTspice simulations and I reckon I can get enough mileage out of the following from merlinmaxwell's suggestion to get a 1% accuracy which is good enough for the analogue and 2000 count meters and a confidence check for more accurate ones. 0.5% is unlikely to happen. The basic idea is to calibrate an HP 427A, Fluke 8024B, 8010A and 8600A which are all simple single control calibration for all AC voltage ranges. The 8010A is a little more difficult as it has more complex true RMS and frequency compensation.

1. Build a peak detector based on an ideal diode opamp configuration with offset null control on the opamp. Using a very low pA bias current amp, a decent film capacitor and buffering the output, also with offset null. Opamp to be selected! I have a couple of RCA matched transistor arrays which should maintain some thermal consistency there!
2. Set the offsets with my HP3478A (which is a little dicky as it was a wreck but fine on mV ranges)
3. Whack the circuit into the function generator.
4. Measure the peak voltage out with a calibrated DC meter and calculate the RMS voltage.
5. Trim the AC metering so that it matches the calculated RMS voltage.

If I get an op amp with reasonable gain-bandwidth product this might do for higher frequencies as well.

This is *good enough* for me which is the objective here to say "yes this is close enough for me to actually say that the meter is good enough for regular use as a hobbyist.
MrBungle is offline  
Old 17th Jul 2017, 8:00 pm   #7
Guest
Guest
 
Posts: n/a
Default Re: Homebrew AC voltage references?

Modern posh op-amps have ridiculously low offsets and gain bandwidth, spend a fiver on a good one and forget trimming.

Rail to rail operation too.
 
Old 17th Jul 2017, 10:44 pm   #8
mhennessy
Dekatron
 
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
Default Re: Homebrew AC voltage references?

KISS

A good sine wave is essential, as you'll no-doubt want to test a mix of true-RMS and average-responding meters.

A simple Wien bridge oscillator will be fine. I have a Lindos LA1, and it's extremely stable, both in terms of frequency and level. The service manual, compete with schematics, is available here: http://www.lindos.co.uk/test_and_mea...s/discontinued

After that, calibrated attenuation and gain. That's the trickiest part of this, especially if you wish to go for high voltages.

Something like an LM3886 will do 30V RMS with no problem, with excellent linearity (far, far better than is needed, but I know it well and use them for other things, so "why not").

A discrete amplifier using high voltage transistors is easy enough - lots of hints in Douglas Self's power amp design books. Obviously, you don't need the current delivery of an audio amp, nor the linearity (1% THD is probably good enough), but I reckon you'd get 100V RMS with ease. The PSU and managing the safety aspects is probably harder.

If you don't need high voltages, you can power an NE5534 or 5532 from +/-22V, which ought to be good for 14V RMS. Use two to make a balanced output for double the voltage... I think the pro units use transformers, but those won't be simple off-the-shelf units.

Of course, you can add some output resistors, to approximate a current source, for the testing of the AC current ranges...

Why worry about DC offsets when you can capacitively couple

In terms of knowing the output, I'd suggest that instead of trying to build a precision rectifier (they are harder to build in practice than I expected), pick up a true-RMS chip. The classic AD536 might still be available, but there are lots of modern ICs that might be better. You just need the IC and 3 or 4 passive components - job done! Plus the output is the RMS value - no need to scale or calculate. And, you can use the output from this to control the amplitude of the Wein bridge while you're at it. That means the core oscillator is reduced to just 1 op-amp and one true-RMS IC. Neat!

I'd recommend adding the option to switch to square waves (1 op-amp). This is important for checking true-RMS meters, to see if they go up by root-2. Average-responding meters go up by rather more.

If I wasn't so busy, I'd give this a go.
mhennessy is offline  
Old 17th Jul 2017, 10:54 pm   #9
MrBungle
Dekatron
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
Default Re: Homebrew AC voltage references?

Interesting ideas. I just built the peak detector and it shows non linearity on DC already and 50-100uV drift with a hair dryer blast so I agree with your assertion about difficulty! Used CA3140 opamps in the end as that's what I had on hand.

I don't need to go high voltages for this. Main thing is approaching the FSD for each meter on one scale. 10v is more than enough peak.

Function generator, a Thandar TG210, is pretty good in the THD department with a little low pass filtering. I've got that set up already. The embarrassing bit with that is the Heathkit breadboard generator I have actually uses a bulb stabilised Wein oscillator with lower THD. Using my shoddy measurement methods (MAudio card plus PC) the former is around 0.5% and the latter 0.1%! There are some spurs if you load the Thandar generator hard.

I'm going to look at your RMS converter suggestion as I can get that calibrated and trimmed against another calibrated meter! Good tip thank you

For ref the DC offset worry was related to testing the peak hold with a DC value which I have and is known.
MrBungle is offline  
Old 17th Jul 2017, 11:34 pm   #10
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Homebrew AC voltage references?

Generating a sine wave from a decent DAC that has a fairly decent reference would be another option.

One of the reasons I bought a Keithley 2015THD was for the AF generator. This should be quite accurate too.

However, another option would be to use something like an Analog Discovery. I'd expect it to give repeatable results when generating a sine wave and it should give very low distortion. I'd expect it to be within 1% in terms of the DAC performance with a sine wave but that is just a guess. It will work from about 5V downwards and will work down into the mV region although I'm not sure how accurate it will be below 100mV. But it should be flat across the frequency range of a typical DVM.

It won't be good enough for a diehard volt nut but (in the real world) I'm not sure how important it is to be able to measure AC voltages very accurately in absolute terms with a DVM. I think the linearity of the meter is much more important because I think relative measurements often demand good performance from a meter. I use a decent step attenuator to check my various AC meters from LF through UHF on the dB scale in 1dB steps over many tens of dB range.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 17th Jul 2017, 11:50 pm   #11
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Homebrew AC voltage references?

The neat thing with the DAC/Discovery method is that you could automate it all if your DVM has a remote control feature. So you could write a few lines of code on a PC to control the Analog Discovery and the meter to check the meter across loads of voltage drive levels and across loads of different frequencies and it would probably do it all in a few minutes.
Obviously the Analog Discovery isn't a traceable calibration standard but the method above would at least give a rapid indication of meter linearity and frequency response for something like a 3.5 digit DVM that has a modest spec for accuracy on ACV (eg 1%).
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 18th Jul 2017, 7:53 am   #12
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Homebrew AC voltage references?

For a strange reason I bought an HP745A/746A AC voltage calibrator. The 745A alone goes from 10Hz to 110kHz and 0.1mV to 110V rms. It chucks out 50mA. That alone weighs 30kg

The 746A is an amplifier connected to the 745A using an umbilical. It uses ceramic bodied valves and internal 6kV power supply, and takes over to produce 1100V rms over the full frequency range at an entirely lethal 63mA rms. About 35kg. You definitely keep one hand in the pocket when using either of these.

Bought them from Helmut Singer in Germany for IIRC a couple of hundred quid. They arrived strapped onto a pallet. Moving them is a real hernia job at a total weight of 65kg.

Craig
Craig Sawyers is online now  
Old 18th Jul 2017, 9:52 am   #13
MrBungle
Dekatron
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
Default Re: Homebrew AC voltage references?

I think I'd keep both hands in my pockets when that was around
MrBungle is offline  
Old 18th Jul 2017, 7:09 pm   #14
chriswood1900
Octode
 
Join Date: Jan 2007
Location: Dorridge, West Midlands, UK.
Posts: 1,474
Default Re: Homebrew AC voltage references?

I considered building one some time ago but chickened out when I got offered a Bradley 125, 30hz to 3.2khz at up to 500V AC, like Craig's beast it is hernia inducing from the massive transformer and powered by 2 CV4060 (S11E12) valves, and has to be used with care. Although it is 1970s it still works remarkably well and is very stable with the reference Weston cell still in spec!
Chris

Last edited by chriswood1900; 18th Jul 2017 at 7:10 pm. Reason: Typo
chriswood1900 is online now  
Old 18th Jul 2017, 8:00 pm   #15
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Homebrew AC voltage references?

Quote:
Originally Posted by MrBungle View Post
I think I'd keep both hands in my pockets when that was around
Indeed. Particularly when the 746A fires up. There is a very large rectangular on light with a lightning bolt on it, and a set of fans that sounds like a 747 taking off.
Craig Sawyers is online now  
Old 18th Jul 2017, 9:16 pm   #16
The Philpott
Dekatron
 
The Philpott's Avatar
 
Join Date: Aug 2016
Location: Colchester, Essex, UK.
Posts: 4,081
Default Re: Homebrew AC voltage references?

....Your next challenge:

A battery powered reference with inverter that can supply 8v AC @ 50Hz, that you can take to swap-meets, charge 50p per calibration and pay for your fuel home!
The Philpott is offline  
Old 18th Jul 2017, 10:36 pm   #17
MrBungle
Dekatron
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
Default Re: Homebrew AC voltage references?

That's actually a pretty good idea. Combined AC/DC calibrator in one would be good.

Not going to get 5 minutes to play with this for a couple of weeks until my current contract is up. However then there's 2 months off before I start the next one
MrBungle is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 1:37 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.