Calculating PWM

[Keosintal]

*Skip the first paragraph if you're uninterested in my introduction, thanks.

Hi, I'm new here.
I'm a professional automotive mechanic looking to further my knowledge in electrical and electronics field. Considering the ubiquity of electronic systems in modern cars, I believe a thorough understanding in electronics is essential to improve my ability in auto diagnostics and repair. I know very little about electronics and wish we can learn together.

I'm planning of buying the Fluke 88V specialty automotive DMM and have a few questions

1. One of the function on the 88V that catch my interest is PWM measurement. But I was pondering, is it possible for us to calculate PWM based on frequency and duty cycle measurement?

2. The 88V is not a true rms multimeter. I wonder how the lack of trms will (or will not) affect its PWM readings? I learned that trms only applies to AC, but I read somewhere in this forum that DC can also benefit from trms capability when measuring PWM and logic signal.

3. Many multimeters can measure temperature with a thermocouple probe. Question is, is the thermocouple an indispensable component to make measurements? Can't I just use my test leads?

Just to clear things up:
- I already searched and asked these questions elsewhere with disappointing result.
- Yes, I need a Fluke. I've had experience where other brands gave me weird readings.

Thanks everyone.

[tpowell1830]

*Skip the first paragraph if you're uninterested in my introduction, thanks.

1. One of the function on the 88V that catch my interest is PWM measurement. But I was pondering, is it possible for us to calculate PWM based on frequency and duty cycle measurement?

PWM is calculated with frequency and duty cycle as you mentioned. In ordered to measure the average voltage of a PWM signal, the meter needs the RMS feature, however Fluke is not the only meter with that feature.

This video is a reasonable explanation:

2. The 88V is not a true rms multimeter. I wonder how the lack of trms will (or will not) affect its PWM readings? I learned that trms only applies to AC, but I read somewhere in this forum that DC can also benefit from trms capability when measuring PWM and logic signal.

Just because the meter does not say true RMS, does not mean that it is not suitable for your needs, this meter does read RMS, but is not optimized.

3. Many multimeters can measure temperature with a thermocouple probe. Question is, is the thermocouple an indispensable component to make measurements? Can't I just use my test leads?

I don't really understand this question. Without the thermocouple device, you can't measure temperature with your leads.

Here is a brief explanation of what a thermocouple is:

Just to clear things up:
- I already searched and asked these questions elsewhere with disappointing result.
- Yes, I need a Fluke. I've had experience where other brands gave me weird readings.

Thanks everyone.

Also, vehicles nowadays require technicians to know a lot about electronic and electrical concepts.

This is a guy who is a mechanic and uses his meters and scopes regularly.

https://www.eevblog.com/forum/chat/automotive-test-equipment-use-of-picometer-as-oscope/msg1814510/#msg1814510

Hope this helps...

[Keosintal]

1. Thanks for confirming it. I wondered if we can calculate pulse width from frequency and duty cycle and search the web to confirm my theory but nothing pops up which makes me thinks I'm wrong. The video you recommended is really helpful and easy to understand.

2. Could you give me a deeper explanation on that? You said 88V does read rms but is not optimized, so how does it differ to an optimized true rms meter? Because I read somewhere in this forum that DC measurement (such as PWM and logic circuit) will also benefit from true rms capability. I'm confused how can a manufacturer as reputable as Fluke did not provide this ability for a meter that's supposed to do mostly 5v logic signal.

3. I apologize for the confusion that arised. Many meters such as 87v, 88v and 233 can take temperature readings. To take the temperature reading, an accessory called 80BK temperature probe is needed. What I mean in my question is can I take temp readings without this accessory, and use my test leads instead. I have watched the video you provided and it seems from my understanding that the thermocouple is what sense the temperature (from voltage difference), so I'm guessing that we need the thermocouple probe to use the temperature measurement function.

4. Thanks for recommending me another source to have a reference. Thank you for your time and explanation. I really appreciate it.

[tpowell1830]

1. Thanks for confirming it. I wondered if we can calculate pulse width from frequency and duty cycle and search the web to confirm my theory but nothing pops up which makes me thinks I'm wrong. The video you recommended is really helpful and easy to understand.

No problem, glad to help.

2. Could you give me a deeper explanation on that? You said 88V does read rms but is not optimized, so how does it differ to an optimized true rms meter? Because I read somewhere in this forum that DC measurement (such as PWM and logic circuit) will also benefit from true rms capability. I'm confused how can a manufacturer as reputable as Fluke did not provide this ability for a meter that's supposed to do mostly 5v logic signal.

Here is a good explanation from Fluke. I hope it explains it better.

3. I apologize for the confusion that arised. Many meters such as 87v, 88v and 233 can take temperature readings. To take the temperature reading, an accessory called 80BK temperature probe is needed. What I mean in my question is can I take temp readings without this accessory, and use my test leads instead. I have watched the video you provided and it seems from my understanding that the thermocouple is what sense the temperature (from voltage difference), so I'm guessing that we need the thermocouple probe to use the temperature measurement function.

Yes, this is true, the thermocouple device is one  type of temperature measurement device. There are other types as well, but it is important that you use the correct sensor that matches your reading device (i.e. multimeter etc.).

4. Thanks for recommending me another source to have a reference. Thank you for your time and explanation. I really appreciate it.

Again, you are welcome and I hope you continue to do searches on your own to find answers that you seek. All of these were found with Google searches. This forum is a great place to get these answers as well, so feel free to ask more questions.

[tpowell1830]

Dave Jones, the owner of this forum also has a youtube channel and has evaluated meters with true RMS in a 2 part series.

[aromring]

Hi, you can build a simple circuit to measure frequency (46Hz - 450kHz) + duty cycle (1%-99%) using these guidelines:
https://www.instructables.com/id/How-to-Measure-High-Frequency-and-Duty-Cycle-Simul/

[Maxlor]

To take the temperature reading, an accessory called 80BK temperature probe is needed. What I mean in my question is can I take temp readings without this accessory, and use my test leads instead.

Yes, you need a type K thermocouple to take temperature readings*. It doesn't have to be the one from fluke, any type K probe that you can physically connect will do.

* Well, actually, there is one thing that you can do if you don't have a thermocouple: if you connect your test leads and short them, the multimeter will display the temperature of its internal temperature sensor, which will show ambient temperature or something close to it. It's occasionally useful. Just to be clear: if you do this, you'll get the temperature inside the multimeter case, it doesn't matter where you put your probes.

[John B]

I have a few Brymens (869, 829 etc) that measure duty cycle, frequency and can capture the max and min voltage up to a few tens of kHz at least. What more do you need to know about the signal?

[The_Eccentric]

Pulse width (the signal's duration) can be useful for some automotive diagnostic tasks.  If you can measure the signal's duty cycle and frequency you can determine the pulse width using this formula.  It can be found in the Fluke 87-V operators manual (and elsewhere of course).

Pulse width (in seconds) = % Duty Cycle divided by 100
                                                        Frequency

[MrAl]

*Skip the first paragraph if you're uninterested in my introduction, thanks.

Hi, I'm new here.
I'm a professional automotive mechanic looking to further my knowledge in electrical and electronics field. Considering the ubiquity of electronic systems in modern cars, I believe a thorough understanding in electronics is essential to improve my ability in auto diagnostics and repair. I know very little about electronics and wish we can learn together.

I'm planning of buying the Fluke 88V specialty automotive DMM and have a few questions

1. One of the function on the 88V that catch my interest is PWM measurement. But I was pondering, is it possible for us to calculate PWM based on frequency and duty cycle measurement?

2. The 88V is not a true rms multimeter. I wonder how the lack of trms will (or will not) affect its PWM readings? I learned that trms only applies to AC, but I read somewhere in this forum that DC can also benefit from trms capability when measuring PWM and logic signal.

3. Many multimeters can measure temperature with a thermocouple probe. Question is, is the thermocouple an indispensable component to make measurements? Can't I just use my test leads?

Just to clear things up:
- I already searched and asked these questions elsewhere with disappointing result.
- Yes, I need a Fluke. I've had experience where other brands gave me weird readings.

Thanks everyone.

Hi,

Are you talking about a square wave PWM or other?
The square wave or rectangular wave PWM is easy to understand, if it is not a square wave then it is a bit more difficult.  Even a sine wave can have a non 50 percent duty cycle (positive half vs negative half cycles).

For a square wave PWM the average voltage is simply the pulse time high divided by the total time period, times the peak voltage.
For a non square wave it is the integral over the period divided by the period:
(1/T)*integrate(f(t),t,0,Tp)
where
f(t) is the wave to be measured,
t is time,
0 is the start time of one period,
Tp is the period time.

You do have to be careful though sometimes you have to integrate over each half cycle independently then add the results in recognition of the fact that areas are usually taken to be always positive.  In cases of 50 percent duty cycle you can usually get away with integrating over just one half cycle then multiplying the result by two.

Example of a regular sine wave 50 percent duty cycle:
f(t)=Vp*sin(2*pi*f*t)

and with Vp=10 and f=100 we end up with:
f(t)=10*sin(200*pi*t)

and using floats:
f(t)=10.0*sin(628.3185307179587*t)

integrate over one half cycle which for 100Hz is 0.005 seconds:
a=integrate(f(t),t,0,0.005)

we get:
a=0.031830988618379

and then divide by the period which is the half cycle 0.005 seconds, we get:
Vavg=6.366197723675813

Note because we integrated over the half cycle using the half cycle period we get the average over one half cycle but since the other half is the same the entire average is as shown.
Note this is considered the average value of a sine wave but the mathematical average is zero because one half cycle cancels the other half cycle.  In measurement theory the result either 6.366 volts or 0.000 volts is interpreted as dictated by the application.  Usually power line measurements use the 6.366 volt measurement interpretation and this is what many regular AC volt meters would read.  RMS volt meters would read 7.071 volts RMS because that is a different measurement and different calculation:
Vrms=sqrt( avg(f(t)^2) )
where
avg=(1/T)*integrate(F(t),t,0,Tp)
as before but f(t) is first squared so F(t)=f(t)^2.
The main difference is we first square the wave before taking the average, then take the average, then later take the square root.  Hence the mnemonic RMS which is Root of the Mean of the Square, where Mean is another word for Average.
Also as before, for the regular sine wave we could do this over just one half cycle but because the wave is first squared we could also do it over an entire cycle without worries about the interpretation.

[Zero999]

Power = V²/R

So a 2V power source will give 2²/1 = 4W into a 1 Ohm resistor.

Now if the duty cycle is cut to 50%, only power is going into the resistor half of the time, so the power dissipation is now divided by 2, or multiplied by 0.5, so P = 4×0.5 = 2W.

Calculate the steady DC voltage this is equivalent to, which is the same as the RMS voltage.
V = √(P×R) = √2 = 1.414V

Put the above together and we get.

V_RMS = √(duty×V_HIGH²)

duty = 50%
V_HIGH = 2V

V_RMS = √(0.5×2²) =√(0.5×4) = √2 = 1.414V

EDIT:

This can be simplified to:

V_RMS = V_HIGH×√duty

V_RMS = 2×√0.5 = 2×0.717 = 1.414V

[jonpaul]

We have bought Fluke 87 V, similar to automotive 88V.
Fluke meters 87 get  highest recommendation a professional tool, made in USA, robust, accurate and gret ergonomics. 10 yrs wty.

The T meas can use any type K TC but the connector should be standard 4 mm banana plus as on the meter.

An adapter is available for Type K mini blade connectors, $2..15 from Fluke or China.

Fluke has a  TC to match the metes but all of our old Keithley, Omega and DIY TC work fine.

Jon

[JustMeHere]

For #3, Dave answers how to do this: