Orval Starter Culture

I'm planning an Orval clone and have had different recommendations on the best way to add the Brett. In his "Brewing Classic Styles" book, Jamil's recipe for an Orval clone calls for adding one package of Wyeast or White Labs Brettanomyes bruxellensis to the batch after the end of primary fermentation. However, a very capable homebrewer from my club said the best results come from pitching a starter cultured from the dregs of Orval bottles. Supposedly there are three distinct strains in Orval, and to get the most authentic character, it would seem that this would be the best route. So I am planning to split a 10 gallon batch and pitch the Wyeast/White Labs Brett in one half and the Orval dregs in the other. The only problem is that I need to create a starter for the Orval dregs -- pitching the dregs of the bottles directly into the pre-fermented 5 gallon batch would not be sufficient. Does anyone have experience doing this? I would assume that the dregs of Orval contain not just the wild yeast, but probably also some bacteria. Under these funky conditions, do you use the same techniques that you would use with an ordinary starter? How many bottles do you think it would take, and how much starter wort should I begin with?

If you search it, you will find it.

More or less, make a 1.030 DME solution, drink two Orvals, add the dregs to the DME solution. Wait for it to get going. Step it up to make a full strength starter. Pitch it into your fresh clone wort. Give it about 6 months before bottling so the Brett can do all it's work.

I'm planning on using Orval yeast in a beer, but not a clone. I haven't quite settled on what yet.

I have an Orval clone on top right now. Other than missing some spiciness it's close.

I did a primary fermentation for 2 weeks with wyeast 3787. It was extract and I think my gravities went from 1.060 to 1.014. I think let it sit on Brett L for 1 month in secondary. It halted at 1.010.

When I make my https://www.homebrewtalk.com/f72/drunk-owl-mango-55227/ I always just add the dregs directly to the secondary and I've never had a problem with it working. Although I know it's going to take the better part of a year until it gets good, I don't if using more would necessarily make it go any faster.