Volt Drop & Wire Size

[Admiral McAwesome]

Has anyone figured out a way to manually set the wire size in Revit 2011? I have found that Revit seems to be measuring from the fixture to the panel, and then back when calculating volt drop, so the wire sizes are usually incorrect when it comes time to upsize.

[RevitNinja]

The method of sizing is hard coded into the program. You can upsize by changing the circuit's rating parameter.

You can always add a custom project parameter to create a field for a manually added wire size.

[brad-TEC]

The distance is actually measured with parallel and perpendicular runs, from the furthest device, to each subsequent device, and finally to the panel. Fairly inaccurate, and not the way we (as I'm sure several others) do voltage dropping. When I posted a similar question last year, it was pretty unanimous across the board that no one used the wire sizes Revit was spitting out.

[Misteracad]

I saw a tutorial that noted the inaccuracy of wire lengths and sizing in Revit, and mentioned that there was an inherent flaw in the program. The suggested workaround was to DOUBLE the Max Voltage Drop values. The end result in the model should be lengths and sizes that more closely match what you would expect. However, calculating the true Voltage Drop would then have to be done manually.

Interestingly, I've yet to see anyone else identify the root of this problem, or offer a similar workaround. I can't help but wonder if Autodesk is going to get this fixed in an upcoming release. The info is not much good to you otherwise lol :roll:

[brad-TEC]

I saw a tutorial that noted the inaccuracy of wire lengths and sizing in Revit, and mentioned that there was an inherent flaw in the program. The suggested workaround was to DOUBLE the Max Voltage Drop values. The end result in the model should be lengths and sizes that more closely match what you would expect. However, calculating the true Voltage Drop would then have to be done manually.

Interestingly, I've yet to see anyone else identify the root of this problem, or offer a similar workaround. I can't help but wonder if Autodesk is going to get this fixed in an upcoming release. The info is not much good to you otherwise lol :roll:

When you say double the max voltage drop values, are you talking about the percentages? I could possibly see that making the sizes closer, but not the lengths. The way it determines the length is always going to be constant, and there really is no way (that I've found) to make them any closer to what we typically use (electrical center of the circuit). We have an in-house voltage drop calc which is very powerful, and I see us continuing to use it, even if Autodesk figures out a way to improve the built in functionality.

[Jimbob]

Hope I am reading this correctly but when we were going through training in 2010 I found (errors) problems in voltage drops and wire sizes, but actually they are right. I know in a lot of cases that the wire sizes shown in revit are something we would never design for or use but can be worked around by increasing the voltage drop settings to 10% instead of 3%. This will keep constant values throughout the project.

Go to manage > mep settings > wiring and see what your percentages are set for and increase by 10%

[Misteracad]

Yes, Brad and JimBob that is what I was talking about...the percentages. JimBob suggests using 10%, my source suggested doubling the OOTB setting of 3% to 6%. I guess you'll have to test it out and see what works best. Let us know how you make out! Cheers :beer:

[brad-TEC]

Interesting thought. I might look into it when I have some time, although I still don't think it'll give wire sizes that we are used to. As an example:

Single load of 1800VA, 100ft away from the source, with a 90% PF at 3% drop yields a wire size of #10 (2.77% total voltage drop).

10 general receptacles, spaced 10ft apart, with the furthest being 100ft away from the source, maintain PF and 3% drop yields a wire size of #12 (2.29% total voltage drop).

I have a feeling (might try it out later) that revit will still size the 2nd example, even with the setting of 6% with a #10 wire because it uses the entire distance, not the electrical center of the circuit. I could be completely wrong, though. Got a busy day ahead of me, but if I get some time, I will try out both examples and report back my findings.

[Jimbob]

Revity uses a simple math problem to solve voltage drop while my progam is more complex in it result, but delivers almost the same result.
The NEC reccomends the following, but of course it is not this simple when dealing with single phase circuits. The formula is written out in my program if anyone would like to try it.
Voltage drop was calculated using the formula found in 2005 NEC®, Handbook, Chapter 9, Example 2. Result is within ±5%. Zc = (R x cos Ѳ) + (Xl x sin Ѳ)
There will alwasy be variations in wire seperation and moisture differances.

http://www.powersofteng.com/eps2010/download.html
Its free.

[brad-TEC]

That is the same equation we use for our calculator. I've included a screen shot. I plan on downloading your just to check it out, as it looks pretty cool!

[Jimbob]

Yes please do, that is why it is there. I hope to improve and expand on it later. If you like it, please tell your friends.

FYI I tryed to using the side car calc in my program and saw maximum length to be 621 feet, otherwise it seems to be accurate.

[brad-TEC]

Single load of 1800VA, 100ft away from the source, with a 90% PF at 3% drop yields a wire size of #10 (2.77% total voltage drop).

10 general receptacles, spaced 10ft apart, with the furthest being 100ft away from the source, maintain PF and 3% drop yields a wire size of #12 (2.29% total voltage drop).

I have a feeling (might try it out later) that revit will still size the 2nd example, even with the setting of 6% with a #10 wire because it uses the entire distance, not the electrical center of the circuit. I could be completely wrong, though. Got a busy day ahead of me, but if I get some time, I will try out both examples and report back my findings.

Ok, so I set my percentages higher (6% branch, 4% feeder), and ran 2 different circuits, as mentioned above (PF was 100%, though). Created a voltage drop schedule, and added a calculated parameter for VD %. Both came back with identical loads (which they should have) as well as identical VD % (which they shouldn't have, at least for the way we do voltage dropping). Voltage drop was 5% for both, and with the doubled percentage allowances, both resulted in #12 wire. The problem is, for a single 15A load, 100ft away from a panel, you are going to want #10 wire. So it works for some circuits, but not for others...

[brad-TEC]

As a proof in point as to why it is wrong, lets break down the voltage drop at each device. For the single load (1800VA), 100ft away, the voltage drop will be 6 volts (5%). For 10 loads of 180VA, spaced equally apart, each device is going to drop as follows:

100ft - 0.6v
90ft - 0.5v
80ft - 0.5v
70ft - 0.4v
60ft - 0.4v
50ft - 0.3v
40ft - 0.2v
30ft - 0.2v
20ft - 0.1v
10ft - 0.1v

This results in an overall voltage drop of 3.3 volts, or 2.75%. The electrical center of the 10 device circuit (since they are equal loads) is 55ft. 1800VA at 55ft results in a voltage drop of 3.3 volts. That is the way the calculation needs to be determined, and if so, will result in the correct wire size for any given situation.

[Misteracad]

So in other words, the workaround I mentioned doesn't actually work? I'm not an electrical engineer by trade, but it would be good to know what to tell new Revit MEP users who have certain expectations about the program in its current form :?

[brad-TEC]

I wouldn't use it. ;)

With the explanation I gave, I think it is a very viable option for Autodesk to create a voltage drop system that actually works properly. Each component would be individually voltage dropped to the panel, and the individual voltage dropped summed together for the circuit to create a true voltage drop for the entire circuit. I don't see it being something hard to do, since its such a simple solution and all the information should already be there (component location relative to the panel, load, voltage, etc).

[mjdanowski]

We use Revit in conjunction with SKM to do our calculations. Revit sums up a fairly accurate load, and then SKM does the power flow, voltage drop and stupidity check with our sizing. Worst case voltage drop (full load) is done in excel as SKM gets annoying with it.

Revit is less transparent than Apple with their calculations, so I wouldn't trust it at all.

[Jimbob]

I think the base problem is that length of conductor needs to be EDITED and not something they think might be right. If the formula they use is correct, and I'm sure it is, then the only variabel that could be throwing it off is length of conductor. Reguardless of the excuss, it is serious and wrong.

[Jimbob]

Mathew, how do you export your data for the electrical SKM?

Is there a way to export electrical circuits to comma delimited or other that have load data? Is it best to do this from a schedule?

[brad-TEC]

I think the base problem is that length of conductor needs to be EDITED and not something they think might be right. If the formula they use is correct, and I'm sure it is, then the only variabel that could be throwing it off is length of conductor. Reguardless of the excuss, it is serious and wrong.

Although the length of the conductor has the ability to be slightly off (it doesn't take into account the fact that the wire will go up into the ceiling and then back down to the panel), I don't feel that is the root of the problem. As stated above, the main issue is that they are using the entire length of the run as the distance by which the calculation is done. This is fine for single loads on a run, but once you have multiple loads on a single run, it is overkill (as I showed in my example above, where the total voltage drop is almost twice what it should be).

There are other items that, if they continue to use the generic formula, as stated in the 2010 Revit MEP User's Guide, are going to cause issues:


Revit MEP uses the following formula to calculate voltage drop in wiring runs:

VD=(L*R*I)/1000

Where:

VD = Voltage Drop
L = One-way length of circuit in feet
R = Conductor resistance in Ohms per thousand feet, from the wire size impedance tables for the specified wire type.
I = Load current in amperes


I'm not even sure where they came up with the values listed in the wire size impedance table (http://docs.autodesk.com/RVTMPJ/2010/ENU/Revit%20MEP%202010%20Users%20Guide/RME/index.html?url=WS1a9193826455f5ff2b5a256211d5f2b3e4a-7a46.htm,topicNumber=d0e35180), but it doesn'e seem to match anything I've seen before. Besides that fact, they are using a specific value based on an ambient temperature, when they also need to be taking into account the power factor on the circuit, as that is going to affect the final wire size as well.

As Jimbob stated above, they really need to use the following formula to determine the effective impedance:

Ze = R x PF + Xl x SIN[ARCCOS(PF)]

and then determine the voltage drop for each component individually using the formula listed above, including the ability to add (or define in settings) the up and down distance for the run. When the circuit is defined, do the calculation for each component individually, then add up the voltage drop of each component to come up with the overall voltage drop for the entire circuit. Doing it this way will result in the most accurate wire sizing.

[brad-TEC]

Just to show how impedance can affect wire size, when I developed our in-house voltage drop spreadsheet, I also included an impedance graph which plots out where the NEC minimum wire size is, as well as the calculated wire size. As an example, I have a 208v/3ph circuit, at 45 amps. The circuit length is 165ft, voltage drop is 3%, and my power factor is 90%. This results in a 1-step upsise, from #8 to #6 (see attachment). As shown on the graph, it is very close to the curve. If the power factor is changed to 95%, a #4 wire size is required. #6 is still very close (it ends up being like 3.01%), but as the chart shows, there are some larger wire sizes that will be affected much easier with a change in PF.

[Jimbob]

I used to be under the impression that you don't have to double the length either BUT you do for single phase circuits. Only a three phase doesn't have to be doubled when using that formula. I you see the examples shown in the NEC handbook as I mention in my program, the example shows a differant formula when solving for single phase. I choose to use the "engineering" formula because I have all the information available to me, along with being easier to deal with and done correctly, can be very accurate, which is what I want. Please try my program and see if you get similar results when using single phase wiring. The formula is written out for you.
As another test, you may try the excel sheet on Mikle Holts web site that is also very accurate. I checked all my calcs using this and other methods.
http://www.mikeholt.com/documents/freestuff/other/VoltageDropCalculator.xls

The simple math formula can be accurate if all the info is entered correctly.

[brad-TEC]

I'm not sure I would say that the distance is necessarily doubled for single phase loads, its just that the multiplier is set to 2, whereas for 3-phase loads its set to square root of 3.

I still have yet to look at yours (can't DL and install it at work, and keep forgetting at home), but I looked at the one on Mike Holt's site and I found some things I don't like about it. First off, the fact that it will recommend a wire size with an ampacity LESS THAN the load entered isn't good (it returns an error, so that's at least helpful). Also, the fact that it is using the resistance from Table 8 (DC resistance) instead of Table 9 (AC resistance) is kind of wierd.

With that said, it really doesn't matter how other VD calcs work, as that's not the point here. The point is to try to pass on to Autodesk what improvements they need to make to the program to make voltage drop and wire size more accurate out of the box.

[Jimbob]

You are right, though, with such a crude formula, you will get a crude result, but hey, there are actually a lot more things that needs huge improvment than just vd, things which I will not elaborate on, till I see how 2012 will turn out.

[brad-TEC]

Not sure when you started using Revit, but 2011 is leaps and bounds better (on the Electrical side) than it was in 2010, and its hard to even compare to Revit Systems 2 (ie, Revit MEP 2006/7...?). Although I haven't had a chance to look at the beta for 2012, reading up on the Beta release notes, not much changes for Electrical. There are a few things I'm looking forward to trying out, though.

[Jimbob]

Yea, we'll see. I have no ideas what they are doing either.

[sparkie001]

I'm not sure I would say that the distance is necessarily doubled for single phase loads, its just that the multiplier is set to 2, whereas for 3-phase loads its set to square root of 3.

If memory serves me correctly, the single phase drop must be multiplied by 2 for unbalanced loads where there is current in the hot and neutral and the wire length involved is therefore double the run length.

Three phase voltage drop is sqrt(3) x the single phase drop calculated.

[brad-TEC]

If memory serves me correctly, the single phase drop must be multiplied by 2 for unbalanced loads where there is current in the hot and neutral and the wire length involved is therefore double the run length.

Three phase voltage drop is sqrt(3) x the single phase drop calculated.

Yes and no. The calculation is as follows:

Voltage Drop = Distance * Phase Constant * (Ze/1000) * Current

Where Ze is the effective impedance (as calculated above somewhere for various power factors, or from Table 9 of the NEC for a general number based on 85% PF) and where Phase Constant equals 2 for single phase and 1.732 (sqrt3) for 3 phase. So yes, you are multiplying by 2 for single phase, but you are not multiplying the single phase drop by sqrt3 (the way I read what you are saying is you effectively multiple by 3.464).

I guess I just never pictured it as doubling the distance, I always just thought of it as multiplying by a phase constant...

[Jimbob]

Where Ze is the effective impedance (as calculated above somewhere for various power factors, or from Table 9 of the NEC for a general number based on 85% PF)

As I have done more research, there are actually few circuit which will use a 85% power factor, in this case a bad low efficent motor, so the power factor side of table 9 is not applicable to most circuits. The formula I use in my program is not based on power factors. I use the left side of table 9 in first two columns only.

Again that program can be found at: There is a conduit fill and voltage drop formula as a side bar.
http://www.powersofteng.com/eps2010/download.html

There are also two very useful pdf's there which have some very useful for design information on them.

[mjdanowski]

Mathew, how do you export your data for the electrical SKM?

Is there a way to export electrical circuits to comma delimited or other that have load data? Is it best to do this from a schedule?

We just do it manually. I don't think there is a way to do it automatically, and if there was I would be wary to trust it.

[Jimbob]

We just do it manually. I don't think there is a way to do it automatically, and if there was I would be wary to trust it.

I am pondering Revit API and EPS with 2 way communications. Revit seroiusly lacks in overall electrical continuity and work flow I think, although everthing else is great. Integrating a great program like EPS into revit would be the ultimate, provided I could force values back into revit and lock calculations.
If you try eps please let me know what you think.

[Jimbob]

I have created a voltage curve chart. Thought I would post it so everyone could see the surpising results of power factors I saw.

.1 power factor has less voltage drops than a more efficient 1. I was very surprised. The slopes are linier.

[brad-TEC]

To maybe get a better view of what you are seeing, have a look at my attachments. This is impedance curves for various wire sizes, and plotted impedance points for entered information. You can see how the impedance of the wire varies with power factor, and the curves are different for each wire size. The curves shown are based on copper wire in steel conduit.

The first curve is a 208v/3ph, 180A circuit, 95%PF, with a 3% voltage drop allowance and a length of 275ft. The NEC required is the minimum size wire for the ampacity of the circuit, and the calculated is the required upsize of the wire in order to have a 3% drop. In this case, the NEC required is 3/0, but the calculated is 250kcmil.

In the next example, all things stay the same, except the length of the circuit is decreased to 150ft. You can now see that the calculated wire size is only 1/0 for the 3% drop, but the minimum size is still 3/0 per the NEC. This results in a voltage drop of only 2.05% for the 3/0 wire.

[Jimbob]

Your chart is difficult to read. Not sure where those height bubbles or for that matter what the height is. Voltage drops are linier.

What did you use to generate this chart and what formula did you use?

[brad-TEC]

They aren't really "heights". If you scroll all the way to the bottom, you will see power factor percentages. The "bubbles" are the calculated impedance (Zeff) of the wire for the given conditions. The Zeff is calculated by the equation listed in Table 9 of the NEC...

Zeff = R x PF +Xl sin(arccos(PF))

where R is the resistance of the wire and Xl is the reactance of the wire, all based on values in Table 9 of the NEC.

[amanda353909]

I know this is an old thread. Just wondering if there has been any changes. Does anyone know if Autodesk is working on the wire size and voltage drop issue? It seems like most companies either do it manually or with a different program. It is often difficult to encourage the use of Revit for Electrical when there are other programs that include the calculations.

[mhartmann]

no changes, the wire length of a circuit that revit is coming up with is completely inaccurate, so it cannot be used for tagging on documentation or in voltage drop calcs. i've tried to walk them thru it, or convince them to let us override the circuit length and wire size. so, until it works right or they let us override, we continue to do all wire sizing and voltage drop calcs manually, blah...

[Misteracad]

Not sure who is still following this thread, but something related came up on the LinkedIn Revit MEP group 7 months ago. I was just on there again just now and thought I would share...

http://www.linkedin.com/groups/Is-anyone-using-panel-schedules-1550697.S.146686996?qid=26aee0b2-a985-477c-9249-7060d8c05471&trk=group_followed_item_list-0-b-cmr&goback=.gmp_2558687.amf_2558687_37669188.gmp_1550697

Specifically I would look at Martin Schmid's responses. Cheers :beer:

[MEP1]

from mine experience and from the previous post, I understood, that VD (voltage drop) seems to be quietly "ununderstandable" in terms of calculation origin.
With a simple test based on 2 identical panels (230V), I found the reason that the cable setting (Ampere, dimension, diameter, used in calculation) represents the most important scenery to be taken into consideration.
Unfortunately, by the preset cable (revit default file, USA standard, es. THWN), the drop voltage has a sense, but by simple personalized cable, there is no way out to figure with drop voltage calculation different from zero.
What's gonna be the wrong/mistake/error?
waiting your response, I thank you.

[JoelLondenberg]

The method Revit uses for calculating wire size vs voltage drop is consistent and matches the method used by all of the Electrical engineers that I've personally analyzed it with.

The short description of calculations:
wire size is driven by max voltage drop
voltage drop is calculated by using the distance (cumulative x,y,z from panel to farthest load) and the load (the largest of either the breaker size or connected load)
it also assumes that the entire load for the entire circuit all goes to the farthest load

More detailed descriptions are available here: http://revitoped.blogspot.com/2012/06/revit-mep-electrical-calculations.html

[David Robison]

no changes, the wire length of a circuit that revit is coming up with is completely inaccurate, so it cannot be used for tagging on documentation or in voltage drop calcs. i've tried to walk them thru it, or convince them to let us override the circuit length and wire size. so, until it works right or they let us override, we continue to do all wire sizing and voltage drop calcs manually, blah...

Revit 2018 added the "Circuit Path" tool to allow you to set the circuit length. It's a step in the right direction, but you still don't have control over the wire size.

While you wait for them to implement that missing piece, take a look at our electrical add-in for Revit. We size your feeder and branch circuit wires for you, but also give you the ability to override the sizes when necessary.

You can learn more and download a free trial on our website:
http://www.designmaster.biz/revit/voltage-drop.html