Problem with Digital Multimeters

[guidoilieff]

Hi. Im making my 4th amplifier, this time with a tda2040.


My problem is when I want to test AC voltage I get very low voltage. For example, when I test 220v from the outlet I get 15v. When I test a 12v transformer i get like 1 volt, and so on.

I tried with a dmm that a friend gave me I have the same readings!

I haven't tried amps. Any idea for an easy/accurate way to test this?


Also... Do you have any idea why this could happen so I don't make the same mistake?


thanks

[Rob Strand]

I do lot of testing and I've never seen that. 

For main frequency voltages you shouldn't see any weird stuff (eg. low or high voltages due to the high frequencies.)

Do you have any idea why this could happen so I don't make the same mistake?

My first impression is you are consistently doing something wrong.  Whatever that is isn't coming through in your question.

Sorry if these sound silly but:
-Are you setting the multimeter to the *AC* *voltage* range?     *DC* voltage will not work.
-Are you connecting both probes?
-Are you getting a good contact.
-Are the leads good?
- Try a different transformer.

You could generate a full-height sinewave 100Hz using your sound-card.
Remove your speakers & headphones.
Measure the output with a multimeter.  You should get around 1 to 2V rms.
It's not a very reliable method the advantage is you can change the frequency
and see if the voltage is relatively constant (say 5% variation at 1kHz).

A known transformer with say a 100ohm load is often more precise.  (It will usually be about 10% high, as the rated voltage is at full load.)

A sanity check to get a rough idea of the voltage would be to connect a rectifier to the AC  then use a DC measurement on the multimeter.

[Oh, flat batteries and sometimes blown fuses in the meter can cause weird problems.]

[guidoilieff]

I do lot of testing and I've never seen that. 

For main frequency voltages you shouldn't see any weird stuff (eg. low or high voltages due to the high frequencies.)

Do you have any idea why this could happen so I don't make the same mistake?

My first impression is you are consistently doing something wrong.  Whatever that is isn't coming through in your question.

Sorry if these sound silly but:
-Are you setting the multimeter to the *AC* *voltage* range?     *DC* voltage will not work.
-Are you connecting both probes?
-Are you getting a good contact.
-Are the leads good?
- Try a different transformer.

You could generate a full-height sinewave 100Hz using your sound-card.
Remove your speakers & headphones.
Measure the output with a multimeter.  You should get around 1 to 2V rms.
It's not a very reliable method the advantage is you can change the frequency
and see if the voltage is relatively constant (say 5% variation at 1kHz).

A known transformer with say a 100ohm load is often more precise.  (It will usually be about 10% high, as the rated voltage is at full load.)

A sanity check to get a rough idea of the voltage would be to connect a rectifier to the AC  then use a DC measurement on the multimeter.

[Oh, flat batteries and sometimes blown fuses in the meter can cause weird problems.]

Hi.

Dc readings are fine. If I use a rectifier voltages are ok. When I switch to ac and test the ac input I have this problem.
I measured AC with this dmm before and it was fine. Now I'm using two different dmm and show the same faulty readings. Im testing it in every power outlet on the walls, and I did it like 50 times by now. I tried even on the solder terminals of a chassis interlock connector with alligator clips to be sure its not a bad connection and moved the dial from ac to dc and back to ac again to be sure.



With the 100Hz test it reads .068v on the 2v scale of the dmm and like .05 on the 20v scale.


When the probes aren't touching anything the voltage floats a little bit, I guess that's ok because rf,emp or something.

[swever]

No expert here but could it be a dead fuse?

[guidoilieff]

No expert here but could it be a dead fuse?

It only has 1 fuse, so if that fuse is dead nothing in the dmm should work.


Also, I already tested it.


Thanks anyway

[diffeq]

No expert here but could it be a dead fuse?

Some DMMs are weird. Mine wouldn't measure current with a dead fuse, with all other features working.

[anotherjim]

The fuse in a DMM is normally just for the current ranges.
Test probes can have a fuse built in.

AC ranges above mV can demand more current from the circuit under test than DC ranges, this can make even a little contact resistance anywhere give an under-reading in those AC ranges.

Usually, if the AC reading is suspiciously low, it's actually reading stray induced current from EM radiation and something isn't making contact or the AC supply is actually switched off.

[duck_arse]

can you tell us the make and model number? meters these days don't need the lead moved from one jack to another for AC voltage measures, like some for current ranges, do they?

[guidoilieff]

can you tell us the make and model number? meters these days don't need the lead moved from one jack to another for AC voltage measures, like some for current ranges, do they?

This is the one https://www.hobby-hour.com/electronics/m890g.php

[guidoilieff]

The fuse in a DMM is normally just for the current ranges.
Test probes can have a fuse built in.

AC ranges above mV can demand more current from the circuit under test than DC ranges, this can make even a little contact resistance anywhere give an under-reading in those AC ranges.

Usually, if the AC reading is suspiciously low, it's actually reading stray induced current from EM radiation and something isn't making contact or the AC supply is actually switched off.

This is just a generic one. I'll be amazed if there's more than one fuse in the whole thing hehe.


Maybe I can try to resolder all through hole components

[Rob Strand]

Maybe I can try to resolder all through hole components

While meter do get faults, two different meters with a similar problem points to a common external cause:
(- the meter settings)
- the meter leads
- the circuit
- the environment

If you have a Wall-wart/Plug-pack with an AC output measure the voltage at home then take it to work or a friends house and re-measure it there.

[guidoilieff]

Maybe I can try to resolder all through hole components

While meter do get faults, two different meters with a similar problem points to a common external cause:
(- the meter settings)
- the meter leads
- the circuit
- the environment

If you have a Wall-wart/Plug-pack with an AC output measure the voltage at home then take it to work or a friends house and re-measure it there.

I know, right? It can't be only the dmm

(- the meter settings)        I triple mega hyper double checked that
- the meter leads       I have spare probes and its the same deal.
- the circuit     Everything in my house works with 220v and everything is working. I can even test dc fine so I guess that means the ac is fine.
- the environment  I don't know what you mean by that


Is there a chip that's responsible for measuring ac? (I've read that cheap dmm transforms the ac to dc and measure that, so I guess when you test dc the dmm bypass the rectifier)

[Tony Forestiere]

Where in the world do you live? What type of socket/plug standard is used?
https://en.wikipedia.org/wiki/Mains_electricity_by_country

[Slowpoke101]

There is a link on the hobby-hour page that shows a circuit diagram. IC2A (TL062) seems to be of interest.
However you stated that another meter shows exactly the same fault (I think), which points towards a manufacturing production error - wrong part installed. Probably a resistor.

If the meter is under warranty - take it back as it is faulty. A low voltage reading on the AC mains could lead you to falsely believe that the voltage is low enough to be safe to work on without isolating the circuit. Not good.

[Rob Strand]

Is there a chip that's responsible for measuring ac? (I've read that cheap dmm transforms the ac to dc and measure that, so I guess when you test dc the dmm bypass the rectifier)

If you have an averaging meter (ie. *not* a true rms meter) then you will find the AC to DC converter is very much like the circuit surrounding IC1 in the circuit below.  Basically R15 through to the output at the cathode of D6/R20.  There two very common and similar circuits, this is one of them. 
https://www.eleccircuit.com/wp-content/uploads/2015/02/the-full-circuit-diagram-of-digital-multimeter.jpg
This one is a slight variant in that the circuit is around IC2A and the DC output comes off a smoothing cap,
http://www.frankshospitalworkshop.com/electronics/projects/voltage_reference/M890G.jpg

Opamps are almost always TL06x.   Occasionally LM358.

Some really dodgy meters might only have a diode.

You really should start with a meter that works and an AC that produces a known correct voltage.  From that point you can start isolating the problem ie. what works and what doesn't and take it from there.

The main issue I see in assuming the meter has a fault is you have two meters doing the same thing.  If they are not the same make and model it unlikely both meters have the same fault (although not impossible).    The last thing you want to do is assume the meter has a fault.  There's a risk you will break something when you pull the meter apart and if you start messing with it you might upset the calibration.  If you have no reference meter or known AC source you cannot re-calibrate it.

[duck_arse]

I found this:
https://www.eevblog.com/forum/testgear/dangerous-multimeter-mastech-m890g-m890g2-(aka-dc-electronics-dc03)/?PHPSESSID=36ia6cni50duvbsb4hchc94j03

[guidoilieff]

Where in the world do you live? What type of socket/plug standard is used?
https://en.wikipedia.org/wiki/Mains_electricity_by_country

I'm from argentina

Type I (Australian AS/NZS 3112); Argentinian version has reversed polarity compared to Chinese and Australian versions ( I don't know what they mean by polarity... phasing? idk. The thing is that I don't know any house that has a conductive rod driven into the ground because its expensive.)


My house has these outlets: So-called "universal socket" which meets no standard[5] but accepts a number of different plug types.(criticized as unsafe)

[guidoilieff]

I found this:
https://www.eevblog.com/forum/testgear/dangerous-multimeter-mastech-m890g-m890g2-(aka-dc-electronics-dc03)/?PHPSESSID=36ia6cni50duvbsb4hchc94j03

I'll check for a transistor...

A forum user posted:
"From my understanding, the NPN transistor "Q4" will blow up if in ohms mode with 240VAC"


I mean... I already opened it and nothing seems burnt but I have no idea what could it be.

[GGBB]

Your mains is 220Hz - that meter's rated range for AC voltage is: "Frequency Response: 40Hz - 200Hz".

[reddesert]

Where in the world do you live? What type of socket/plug standard is used?
https://en.wikipedia.org/wiki/Mains_electricity_by_country

I'm from argentina

Type I (Australian AS/NZS 3112); Argentinian version has reversed polarity compared to Chinese and Australian versions ( I don't know what they mean by polarity... phasing? idk. The thing is that I don't know any house that has a conductive rod driven into the ground because its expensive.)


My house has these outlets: So-called "universal socket" which meets no standard[5] but accepts a number of different plug types.(criticized as unsafe)

Polarity in this context means which pin is connected to the hot AC, vs. the neutral. See the "polarization" section of the wiki article.