Jump to content
 

DMM shows different track voltage to other DMM's


Rick White
 Share

Recommended Posts

Starting to wire up a new switching layout that is under construction and so I use my DMM (multimeter) to roughly check track voltages after soldering up the initial couple of feeders to my dcc bus wires.

The DMM showed an AC voltage of 2.8v.  That can't be right so I connected my Fluke DMM and it reads 13.3v AC which is about what I was expecting.

The first DMM is a new autoranging, high spec Brymen and correctly measures the DC supply voltage to the DCC command station.

Anybody got any clues as to why the Brymen measures 2.8v when the actual voltage is 13.3v?

Edited by Rick White
Link to post
Share on other sites

If you connect both meters to the track via a bridge rectifier, they will then show the same voltage.... About 1.4V below track voltage, and perhaps a volt or so above the motor voltage ..... This result will be consistant because the rectified voltage will give a steady value ( dcc is a square wave )?.

The digital meters use sampling of the incoming signal to convert to the display... And especially for the 'cheaper' meters, this is based around a 50-60Hz sine wave as would be found when measuring mains voltage, or a lower voltage transformed directly from it.

 

I too, have found Fluke meters to give a very close result to that which would be measured off an oscilloscope.... All one can really say is that this reflects the quality and probably price difference, of the Fluke meter.... It may ( conjecture ) perform the equivalent of many modern oscilloscopes, and auto-time-scale its sampling if it detects a repeating natural frequency in the signal... Or it may not 8-)

 

For the 5gbp alternative.... Use an Automotive Voltmeter Module ( 0-30Vdc ) and connect it via a bridge rectifier... Perhaps also with a capacitor after the bridge..... Then it will also work on a gaugemaster's 'analogue' controller with pulsed outputs which can fall to zero between pulses if no back elf from a motor present.... Turning the display's internal processor off repeatedly!! .... Similar problem to dcc decoders placed on analogue track with pulse outputs...

  • Like 1
Link to post
Share on other sites

It could be to do with the Brymen's sampling rate when measuring AC.

Remember that a DCC signal is not AC, in that it is not a Sine Wave. It is a series of pulses at whatever line voltage you are using.

 

The best way to look at a DCC signal is with an oscilloscope, but even the handheld ones don't come cheap.

 

Regards,

 

John P

Link to post
Share on other sites

Thanks guys :

I previously had a cheap DMM and before it died it would measure in the same 13.3v area. And the Fluke is an old entry level model (but still a Fluke).

I bought the Brymen to replace the dead one after many recommendations that this brand and model is excellent.

I'm too cheap to afford an oscilloscope but I will try the bridge rectifier suggestion and measure the DC voltage.

 

Thanks again ... must be a function of the Brymen.

Link to post
Share on other sites

I bought one of those combo volts/amps panel meters to monitor my DCC track and when comparing its reading to my two DMMs they were all well different.

 

Much depends upon the sampling rate frequency of your meter as to the reading seen. Check your meter specs for frequency range on AC.

 

As already stated by others reading DCC voltage as AC only gives a rough reading which for comparative purposes is usually enough.

 

Google RRAMP meter which will read DCC properly, but again these are not pocket money items.

Link to post
Share on other sites

  • 2 years later...
59 minutes ago, Anthony566 said:

I know this is an old thread but has anyone tried one of these yet ??

 

https://www.dccconcepts.com/product/alpha-meter-for-dc-or-dcc/

 

They have been hinting at this for quite a while - great pity that they have made it a fixed item, screwed to the facia plate, as whilst this will tell you the output of the Command station that isnt a lot of use when you wnt to know if there is a voltage drop around the layout.

 

Brilliants idea, just not quite what is required.

 

EDIT - price is also steep at around the same or more as RRampmeter, which is far from cheap (and you can get RRAmpMeter which is portable  for less!)

Edited by WIMorrison
Link to post
Share on other sites

Measuring DCC voltages accurately that will show you volt drop around the layout isn't a very exact science using everyday multimeters as has been noted.

In this instance the query is about new wiring, not operational measurement that's being checked and there is no need to use the DCC system at all. I always avoid doing so in this type of situation just to make sure any error doesn't cause equipment problems.

 

if your meter is of  good quality,  don't energise the circuit at all. Simply use the resistance measurement lowest range with probes at (1) power source for rail 'A' and (2) rail 'A' where you want to measure. Repeat for othe rail 'B'  (and across the rails to make sure there are no shorts).

 Before use hold the probes together, they will show a small resistance and use this as a reference start value. As you get further from the source point, the meter lead will need extending, again measure the overall lead resistance first, then measure and subtract as noted above. The overall result will be higher the further from the source, but realistically should all be very very close to zero Ohms if all joints are tight.

Link to post
Share on other sites

The other way for build continuity and short checking is to wire a buzzer to a 9v battery with croc clips on the ends of the wires. Clip the wires to the rails or bus and if you cross a drooper to bus wire as you go then the buzzer sounds. No noise no problem, providing you test the circuit by putting a coin across the rails at setup.

Edited by RAF96
Link to post
Share on other sites

21 hours ago, steve W said:

if your meter is of  good quality,  don't energise the circuit at all. Simply use the resistance measurement lowest range with probes at (1) power source for rail 'A' and (2) rail 'A' where you want to measure. Repeat for othe rail 'B'  (and across the rails to make sure there are no shorts).

 

On a large layout (2) my be some distance from (1), certainly beyond the reach of the test leads. If you extend the leads then you are also measuring their resistance as well as the track!

 

Simply connect the meter to both wires at the power source (1) then short the two rails at the point where you want to check the resistance (2). You'll measure the total "out and back" resistance which is more meaningful since both legs contribute to any voltage drop.

 

  • Agree 3
Link to post
Share on other sites

35 minutes ago, Crosland said:

Simply connect the meter to both wires at the power source (1) then short the two rails at the point where you want to check the resistance (2). You'll measure the total "out and back" resistance which is more meaningful since both legs contribute to any voltage drop.

 

Yep, why didn't I think of that:rolleyes:

Link to post
Share on other sites

Note that a bridge rectifier will not improve a Dmms ability to correctly determine the DC level of a high frequency asymmetric pulse train like dcc , it’s actually very complicated to determine what it is.  The text book definition is the equivalent hearing value but that’s impossible to measure with a DMM 


a bridge rectifier converts alternating current into unipolar current , it still remains a high frequency pulse train. 
 

DMMs for various reasons make have quite a low pass filter often to about a couple of kilohertz again making comparisons difficult. 
 

simply put you can’t really make any assumptions about the value a DMM reads out in these circumstances 

Link to post
Share on other sites

1 hour ago, Junctionmad said:

Note that a bridge rectifier will not improve a Dmms ability to correctly determine the DC level of a high frequency asymmetric pulse train like dcc , it’s actually very complicated to determine what it is.  The text book definition is the equivalent hearing value but that’s impossible to measure with a DMM 


a bridge rectifier converts alternating current into unipolar current , it still remains a high frequency pulse train. 
 

DMMs for various reasons make have quite a low pass filter often to about a couple of kilohertz again making comparisons difficult. 
 

simply put you can’t really make any assumptions about the value a DMM reads out in these circumstances 

Hi,

 

A bridge rectifier may improve the DMMs ability if the DCC waveform voltage rise and fall times are fast.

 

DCC is symmetric both in voltage and pulse timing (except during the short Railcom messaging? - if turned on).

 

What I think may be better is a high speed bridge rectifier plus a small capacitor to smooth the rectified signal. The DMM should then display the DCC voltage minus the voltage drop across the rectifier.

 

Regards

 

Nick

  • Agree 1
Link to post
Share on other sites

11 hours ago, Junctionmad said:

a bridge rectifier converts alternating current into unipolar current , it still remains a high frequency pulse train. 

 

More to the point it rectifies the AC DCC signal to a DC voltage. DCC is near enough a square wave so we don't need to worry about RMS values. The rectified voltage (with a little smoothing) will be two diode drops less than the track voltage. 

Link to post
Share on other sites

There is no such thing as a “DC “ value of a square wave , there is a Peak , peak to peak , average or RMS value 

 

even rectifying it does nothing to resolve the DC component, adding a capacitor again doesn’t construct the original DC value of the signal , for example the ripple voltage can be very significant depending on the load 

 

hence the whole idea of trying to characterise  a DCC waveform by some notional DC number is nonsense , misleading and meaningless. 

Edited by Junctionmad
Link to post
Share on other sites

Quote

What I think may be better is a high speed bridge rectifier plus a small capacitor to smooth the rectified signal. The DMM should then display the DCC voltage minus the voltage drop across the rectifier.

 

Essentially that’s the peak value but it’s not the DC equivalent 

Link to post
Share on other sites

1 hour ago, Junctionmad said:

There is no such thing as a “DC “ value of a square wave , there is a Peak , peak to peak , average or RMS value 

 

even rectifying it does nothing to resolve the DC component, adding a capacitor again doesn’t construct the original DC value of the signal , for example the ripple voltage can be very significant depending on the load 

 

hence the whole idea of trying to characterise  a DCC waveform by some notional DC number is nonsense , misleading and meaningless. 

Hi,

 

The topic heading is track voltage so characterising all of a DCC waveform does not appear to be the OPs question.

 

As previously mentioned the DCC waveform is very close to a rectangular wave. I agree if there is ripple under load in the form of a medium impedance command station or electrical connection then a DMM on its own does not tell the whole story.

 

I'm not sure if significant ripple is allowed in the DCC spec for command stations - something to check.

 

A DMM/rectifier plus a suitable power resistor in parallel might be a better solution to indicate what voltage would appear on a DCC loco decoders internal reservoir capacitor.

 

Regards

 

Nick

Link to post
Share on other sites

Can I suggest that the actual ‘voltage’ is academic and what actually matters is the differential between what comes out of the command station and what is available to the track and a train at any given point on the track.

 

if you get ‘17.9v’ at the back of the command station (Z21 for me) and at the furthest point - using the same DMM - I get ‘17.88v’ then there is nothing wrong with the bus. 
 

if at some place on the track I got ‘16.5v’ or similarly low reading then I would have an issue that I need to resolve as that suggests a dropper is at fault.

 

the ‘real’ voltage measured by some expensive meters could be any value, it doesn’t really matter as it will not change the performance of my layout or the reading on the meter.

 

for the record the Z21 outputs 17v ;)

Edited by WIMorrison
Link to post
Share on other sites

 

 

1 hour ago, WIMorrison said:

Can I suggest that the actual ‘voltage’ is academic and what actually matters is the differential between what comes out of the command station and what is available to the track and a train at any given point on the track.

 

if you get ‘17.9v’ at the back of the command station (Z21 for me) and at the furthest point - using the same DMM - I get ‘17.88v’ then there is nothing wrong with the bus. 
 

if at some place on the track I got ‘16.5v’ or similarly low reading then I would have an issue that I need to resolve as that suggests a dropper is at fault.

 

the ‘real’ voltage measured by some expensive meters could be any value, it doesn’t really matter as it will not change the performance of my layout or the reading on the meter.

 

for the record the Z21 outputs 17v ;)

 

I would be cautious about using voltage to check for resistance in a circuit.

A voltmeter has a very high resistance in order to not interfere with the characteristics of a circuit. Any high resistance joint in a model railway would be negligible in comparison.

 

Any volt drop would be caused by unwanted resistance. Since we were talking about multimeters, then they can measure resistance too.

 

What I would do is this:

 

On 17/01/2020 at 17:39, Crosland said:

 

Simply connect the meter to both wires at the power source (1) then short the two rails at the point where you want to check the resistance (2). You'll measure the total "out and back" resistance which is more meaningful since both legs contribute to any voltage drop.

 

 

Link to post
Share on other sites

7 hours ago, Junctionmad said:

There is no such thing as a “DC “ value of a square wave , there is a Peak , peak to peak , average or RMS value 

 

even rectifying it does nothing to resolve the DC component, adding a capacitor again doesn’t construct the original DC value of the signal , for example the ripple voltage can be very significant depending on the load 

 

hence the whole idea of trying to characterise  a DCC waveform by some notional DC number is nonsense , misleading and meaningless. 

 

7 hours ago, Junctionmad said:

Essentially that’s the peak value but it’s not the DC equivalent 

 

To have a DC component you need a 0V reference against which to compare. DCC measured at the track has no notion of where 0V is. If we have a nice clean DCC signal, with no zero stretching, then we can assume the DC component is zero, or even define it to so in our frame of reference. We can forget about the DC component.

 

When you rectify a sine wave for example, you get a series of positive going peaks. To work out the equivalent DC (e.g. equivalent heating effect) we do an RMS calculation. We square the signal, take the average and then the square root of the average. The result is peak/1.414 for a sine wave and only for a sine wave. I'm sure you know this :)

 

When you rectify a perfect square wave you get a steady DC voltage. If you calculate the RMS you get exactly the same voltage. This is the "DC equivalent" track voltage, for want of a better term. This is what you are measuring when you use a bridge rectifier and a DMM.

 

Ignoring diode and switching losses (which will be small with nothing but a DMM connected): if your command station has a 12V supply then the voltage between the rails is alternating between +12 and -12V. When you rectify this you get 12V DC (or -12V if you connect the meter the other way round).

 

Obviously things are more complex on a non-ideal running layout. Current demand will be a complex waveform due to all the PWM control in the decoders.

 

Link to post
Share on other sites

Quote

When you rectify a perfect square wave you get a steady DC voltage. If you calculate the RMS you get exactly the same voltage. This is the "DC equivalent" track voltage, for want of a better term. This is what you are measuring when you use a bridge rectifier and a DMM.

 

 

What I was stating is that , taken as it is , a DCC signal cannot be characterised by a DC number.  Run it through a rectifier and you have " converted " it to DC, but thats different 

Link to post
Share on other sites

Different but exactly what our locos run on, the decoder does the rectifying. So a useful measurement although just using aM on AC is fine for comparison purposes.

If you want to find high resistance joints then the best thing is to provide a load at the place you want to measure, something like a 10 ohm 5 watt resistor will pull enough current to show up the reduced voltage a loco may see.

Link to post
Share on other sites

@Grovenor I simply use a loco as the load on the track :)

 

If i notice a difference in performance, speed or pulling capacity of a loco in an area of track then I simply stick the meter across the track with the loco there and see if there is a voltage drop, if there is then I go hunting to find the voltage drop, if there isn't it generally means that the loco needs a service.

Link to post
Share on other sites

9 hours ago, Grovenor said:

If you want to find high resistance joints then the best thing is to provide a load at the place you want to measure, something like a 10 ohm 5 watt resistor will pull enough current to show up the reduced voltage a loco may see.

If you want to find resistance, then surely it is easier to use the resistance setting on your multimeter? Who has a 10 ohm 5 watt resistor in their electronics box going spare? Who even has an electronics box? (actually I have 2 ... until I find another I had forgotten about, but many don't).

 

I use the method described earlier by Crosland. We may not have a fat resistor but all have a spare piece of wire.

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...