advertisement


full wave rectification AC voltage

hifienthusiast

Vintage by heart
Hi

I have built a dual 24V regulated power supply using two LM317 regulators. The transformer has two primary windings of 0-9V and 0-9V, so I connected the two windings in series to give an AC voltage of 18V, rectified by a full wave bridge (note: not half wave as used in Hicap) and filtered by a single 10000uF cap.

The regulated DC voltage from both 317 regulators is measured at about 23.5V, but I can measure an AC voltage of 50V from each of the regulators. Can someone give me a reason for this AC voltage?

I have a Flatcap2 power supply, but there is no AC voltage across the outputs.

Thanks in advance.

Regards
Hon
 
What are your regulator resistor values set to? I think you don't have anything like enough transformer volts to regulate to 24.

Paul
 
Hi there,

I have also built a PS using a toroid with 9+9 secondaries.

I joined them in series hoping to get 18V ac but FWIW the input to the bridge actually measures at 21.15 Vac .

From that, I expected to get approx 1.4 x Vac as a raw DC voltage but mine actually measures at 27.23 Vdc. (approx 1.29)

I feed that through a standard 317 setup to get an output of 24.29Vdc. There is no measurable AC on the 317.

My PS is powering a phono stage so there is no great load, but I'm not sure how the load affects real world measurements.

Exactly which terminals are you measuring this AC across?

I'm a bit confused about how you connected the secondaries (surely thats what you meant?) and still have two fullwave bridges working off it?

A picture is worth a thousand words! I'm sure we can help if you provide a diagram of how you have things connected.

Don't be shy.... I still have the lid of a 10000uf cap buried in my plasterboad ceiling after I wired it in with the wrong polarity!!...:D
 
The AC voltage I measured from the 317 regulator turned out to be a fault with the AC range on my multimeter. Using the AC range, it displayed 18V in AC across a 9V alkaline battery! I feel really stupid now. I will need to get a new multimeter.

Hon
 
Been there - done that. I thought I had a problem with my 260Z boards which turned out to be a fuse blown in the 'meter.

I still have the lid of a 10000uf cap buried in my plasterboad ceiling after I wired it in with the wrong polarity!!...
:eek: :D
 
The AC voltage I measured from the 317 regulator turned out to be a fault with the AC range on my multimeter. Using the AC range, it displayed 18V in AC across a 9V alkaline battery! I feel really stupid now. I will need to get a new multimeter.
Hon

If you or anyone else are in the market for a decent little DIYers multi meter i found and bought a great one from Argos of all places, for £25 the other day.

Quite a few months ago in the DIY pages/supplement of Hi-Fi World magazine there was a feature/recommendation for a Multi-Meter sold by Rapid Electronics, (Feb 2002 Page 102) it was said to be particularly suitable for those that play with valve amps. Anyhow based on this recommendation i bought one for home use, price £18.
I now find that i have an increasing use for a Multimeter at work too, these items always being useful to have in my work tool kit, especially for when the works Fluke is away on cal, or i have to go and fix/recover an Aircraft that has broken down at a different Airport, so i bought its bigger brother.

The MM sold by Argos really is a well featured little machine and covers most functions needed by the keen home DIYer. These include AC frequency (upto 20 kHz), Temperature -50 to +1000 deg C (allegedly), there is also a facility to test the values of capacitors that seems to work accurately with all of the caps I checked so far (given their tolerances). The meter also has all of the normal ranges you would expect AC/DC volts, resistance, Transistor/diode checker.
I cheked the frequency range on a (calibated) sine and square wave generator we have at work and it was spot on.

All of the Measuring instruments that i use at work are periodically calibrated, by an indipendant company. We need to do this to maintain our CAA authorisation and approval to operate. When testing this machine form Argos against the Calibrated machines at work I've yet to find anything to suggest that the 'cheap' machine is any less accurate. check it out here:
The well featured MM at Argos

Looking around the web and various other sources of MM sellers there seems to be a whole load of these machines , all the same units, but with different Badges on them. I now have two machines, The Rapid Electronics badged item as recommended by HFW and the higher facility count machine, badged as a 'Hilka' from Argos, it is clear that both units are made by the same people.
HTH.
 
Hi Richard and Sid

Thanks for your advice on multimeter. I have had my cheap meter bought in Hong Kong over 10 years ago. It has been a faithful workhorse until now. I used to mess with high voltage in valve amps a lot, may be I have blown the AC range of the meter. A few years ago, I had a much more expensive meter bought from RS, but it didn't last long after using it for the high voltage in valve amps, it completely shorted itself!

I am going to Hong Kong in Dec to see my family, I will have a look there for multimeters. The Argos one certainly looks great.

Regards
Hon
 
I have built a dual 24V regulated power supply using two LM317 regulators. The transformer has two primary windings of 0-9V and 0-9V, so I connected the two windings in series to give an AC voltage of 18V, rectified by a full wave bridge (note: not half wave as used in Hicap) and filtered by a single 10000uF cap.

The Hicap does use full-wave rectification, but in the form of 'dual half wave' using a centre-tapped transformer.

For your transformer you will get a raw supply at the full rated transformer load of: -

(18V x 1.4) - (2 x 0.7V) = 23.8V approx (1)

The 2x 0.7V is the voltage drop in the two diodes of the bridge rectifier - here you see a potential advantage for the dual half wave circuit - only one diode (and its non-linearities) in series with supply.

Now at lower current drains this voltage will rise by an amount determined by the transformer regualtion figure, let's assume a 10% regulation figure and the raw voltage will be: -

23.8V + 10% = 26.1V approx(2)

You need to measure the raw DC under load to determine the accuracy of the estimates in (1) and (2).

All regulators have a drop-out voltage - this is the minimum input - output differential for correct operation. This isn't a fixed figure either, it varies with current, temperature and manufacturing spread.

Assuming you wish to run something like an old preamp or CD player o/p stage you're in the region of 100mA maximum so using this conservative figure you can look to the manufacturers data sheet.

LM317 Data Sheet

Look at the graph on page 6 labelled 'Dropout Voltage', from this you can see 5 curves, plotting input-output differential against temperature.

We obviously need to look somewhere between the 20mA and 200mA load curves from this you can see at 25 deg C you need a differential of 1.5V minimum for a 20mA load, rising to approx 1.7V (3) for a 200mA load.

Good design practice calls for us to be conservative in our design practice - use next highest load line to that you are operating under and factor in the temperature extremes.

In your case you have a raw DC supply that you can't change without changing transformer, so we can work backwards to the acceptable regulated voltage you can achieve for correct operation, the normal route would be to design for a target regulated voltage, then choose an appropriate transformer.

Based on (1) and (3) we can see that the maximum regulated voltage one can achieve is: -

23.8V - 1.7V = 22.1V

using (2) and (3) it would be: -

26.1 - 1.7V = 24.4V

I suspect the truth lies between the two, but the above will allow you to take measurements and determine the correct values.

I doubt you will get 24V though.

Andy.
 
Quality post, as usual, Andy.

Sid, I have one of those Rapid Electronics meters, too - good spot re Argos though - and good price!
 
What is the precision when measuring ohms, say when measuring 5 ohms? My cheapo MM is 5+/-1ohm which is pants.

If it could measure to 0.1 ohm , then I'll go out and get one!
 
My Rapid is +-0.8% of the reading, +- 3 digits which equates to about 5 ohms. However, I think the probes have a resistance of 0.7 or thereabouts so being able to measure 0.1 ohm is impossible.

Pardon the question but why would you need to?

Mick
 
Speakers. Ever had to sort resistors to get a matched pair? Plenty of single digit ohm resistors here. I can measure the resistors with high accuracy using a sound card and a test jig, but sometimes its nice to have the facility to do a quick and dirty check...

0.7 ohms seems quite a high resistance for test probes...
 
Hi,

I just bought a cracking little meter from maplins it a 3 ½ digit and it has ranges that go down to 200uA, 200mV and HFE restance diodes etc so it will do for setting up my 135 clones (4.4mV across the 0.1Rs) price only £2.99 unbelievable! I have been looking at Flukes etc for £70+ that do the same thing



pete, Mad Bad and Dangerous to Know
 
What is the precision when measuring ohms, say when measuring 5 ohms? My cheapo MM is 5+/-1ohm which is pants.

Richard,

Trawling the depths of my brain from many moons ago, a Wheatstone bridge is what you are looking for.............

A google search will reveal any number of references that will get you going.

In essence you can use bog standard known resistors to find an unknown value.

What is less well publicised is that the accuracy of the result is well beyond the nominal accuracy of the other resistors in the bridge, its a kind of cumalitive effect in the positive sense.

If I was really flash I would give you the mathematical proof............ but it would probably take me days!!!!:D

A real world use would be a strain gauge on a structure where the resistance changes as the strain changes. Because the bridge is so so sensitive, minute changes in resistance can be detected.

Cheers

pm
 


advertisement


Back
Top