DMM Frequency response
shan
I have some questions about measuring AC voltages.
I know that low-cost DMMs cannot measure accurately the AC (rms)
voltage for too high frequency normally below several hundred Hz even
for sinusoidal signal. But why? Is it due to the electronics inside or
is it a loading effect?
What above analog multimeters? Are they better/worse?
Besides the internal resistance ~ 10Mohm, what are the typical parallel
capacitance / series inductance values across a DMM inputs? Is it
normally neccessary to take these into account?
Thanks!
Charles Schuler
"shan" <shan....@gmail.com> wrote in message
news:1131880577....@g47g2000cwa.googlegroups.com...
If a DMM has an input impedance of 10 M Ohms and 20 pF, the 3 dB frequency
is 1/(2piRC) = 800 Hz. The capacitance in this case is a shunt effect to
ground after the 10 Megs of resistance ... so it forms a low-pass filter.
A 1 volt signal at 800 Hz will display as 0.7 volts.
Scopes are better for higher frequencies.
shan
-----\/\/\/\/----- Vdmm
Vin R |
C =
|
-------------------
If we put DC voltage across the inputs why not the DC voltage shows up
(in AC measurement mode)?
How scopes and other more sophisticated devices do? How to represent
the these as equivalent circuit?
Thanks very much!
Tim Williams
> If we put DC voltage across the inputs why not the DC voltage shows up
> (in AC measurement mode)?
Because it's made to ignore DC potential...
My RadioShank DMM (before it died...sigh) started dropping off at 3kHz,
based on my observations.
Tim
--
Deep Fryer: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
Charles Schuler
> So, the equivalent circuit is something like this?
>
> -----\/\/\/\/----- Vdmm
> Vin R |
> C =
> |
> -------------------
Yes.
> If we put DC voltage across the inputs why not the DC voltage shows up
> (in AC measurement mode)?
There is a series blocking capacitor.
> How scopes and other more sophisticated devices do?
It's selectable on a scope ... dc coupling (no blocking capacitor) ... or ac
coupling (capacitor). dc coupling is also called direct coupling.
Jasen Betts
> If a DMM has an input impedance of 10 M Ohms and 20 pF, the 3 dB frequency
> is 1/(2piRC) = 800 Hz. The capacitance in this case is a shunt effect to
> ground after the 10 Megs of resistance ... so it forms a low-pass filter.
>
> A 1 volt signal at 800 Hz will display as 0.7 volts.
>
> Scopes are better for higher frequencies.
couldn't they bypass the 10M with a suitable cap to reduce that ?
wether the DMM can sample the input fast enoug could be another problem..
Bye.
Jasen
shan dot at dot
news:4KKdnfM4UJu...@comcast.com:
>> So, the equivalent circuit is something like this?
>>
>> -----\/\/\/\/----- Vdmm
>> Vin R |
>> C =
>> |
>> -------------------
>
> Yes.
>
>> If we put DC voltage across the inputs why not the DC voltage shows
>> up (in AC measurement mode)?
>
> There is a series blocking capacitor.
>
If we add a series cap (denoted it as C1) at Vin, there should still be
DC voltage across C. Maybe C1>>C so that most voltage is on C1? By the
way what is the typical capacitance needed for DC blocking? What are the
main considerations?
>> How scopes and other more sophisticated devices do?
>
> It's selectable on a scope ... dc coupling (no blocking capacitor) ...
> or ac coupling (capacitor). dc coupling is also called direct
> coupling.
Even we can select dc/ac coupling, it only deals with the DC component,
how can it avoid low-pass phenomenen? It should not be simple otherwise
all cheap DMMs can do this, I suppose?
Thanks.
shan dot at dot
news:slrndnj4is...@clunker.homenet:
> couldn't they bypass the 10M with a suitable cap to reduce that ?
>
> wether the DMM can sample the input fast enoug could be another
> problem..
>
>
> Bye.
> Jasen
>
If they bypass 10M, I think the impedance may be too low.
Charles Schuler
"shan" <shan_(dot)_hello_(at)_gmail_(dot)_com> wrote in message
news:Xns9710D1...@216.168.3.30...
> "Charles Schuler" <charle...@comcast.net> wrote in
> news:4KKdnfM4UJu...@comcast.com:
>
>>> So, the equivalent circuit is something like this?
>>>
>>> -----\/\/\/\/----- Vdmm
>>> Vin R |
>>> C =
>>> |
>>> -------------------
>>
>> Yes.
>>
>>> If we put DC voltage across the inputs why not the DC voltage shows
>>> up (in AC measurement mode)?
>>
>> There is a series blocking capacitor.
>>
>
> If we add a series cap (denoted it as C1) at Vin, there should still be
> DC voltage across C. Maybe C1>>C so that most voltage is on C1? By the
> way what is the typical capacitance needed for DC blocking? What are the
> main considerations?
You are thinking about the distribution of a dc voltage across two
capacitors in series ... does not apply here. The capacitor is selected to
have a low reactance at the lowest frequency of interest. At dc, 0 Hz, its
reactance is infinite; hence the dc blocking function.