Calculating measure total without visual level filter (not slicer)

[sja133]

I thought I had this, but it's not working for some reason and thought maybe the forum might be able to help. 

I'm developing a logistics measure for Rush Orders with no charge as a % of Total order by customer. I am currently displaying a table visual which has a few visual filters on, one of which is "Charge Description" with the filter on "Rush Order No Charge" (I need this filter on for a few other measures to work).

I created the below measure 

% RushOrders no Chg =

var RO = CALCULATE([# Orders Shipped],Data[Charge_Desc] = "RUSH ORDER NO CHARGE")

var TotOrd = CALCULATE([# Orders Shipped],ALLEXCEPT(Data,Data[CustNumber])

return Divide(RO,TotOrd,0)

When I test the individual variables, RO works as intended with/without the visual level filter mentioned above. However TotOrd seems to show RO's result when I do the above formula and turn on the filter Rush Order No Charge. I would have thought specifying the Charge_Desc column within the ALL function would have removed the effect of the visual level filter, but it seems to not be working. 

Summary of the results I'm getting: 

Actual number of Orders 168 (i.e. this is what TotOrd should be showing always)

Charge_Desc filter off:
RO shows 19
TotOrd shows 168

Charge_Desc filter on:
RO shows 19

TotOrd shows 19

Appreciate your ideas on this.

[sja133]

I have also tried the below, but still no luck

TotOrd = CALCULATE([# Orders Shipped],ALL(Data[Charge_Desc])

[amitchandak]

@sja133 , try these

TotOrd = CALCULATE([# Orders Shipped],removefilters(Data[Charge_Desc]))
TotOrd = CALCULATE([# Orders Shipped],ALL(Data))

TotOrd = CALCULATE([# Orders Shipped],ALLSELECTED(Data))

[sja133]

Thanks for your suggestions @amitchandak , unfortunately these end up giving me the same result when i turn on the visual level filter. 

[amitchandak]

@sja133 , Can you share sample data and sample output in table format? Or a sample pbix after removing sensitive data.

[sja133 · Accepted solution]

@amitchandak i was prepping the data for this question, but i trialled the ALLEXCEPT formula again and seemed to have gotten an acceptable solution. thank you for trying to help 😄