Some notes/ideas on cassia quality reiterating/expanding on Mick's suggestions. Not in any particular order. Also appreciate I have left out many other variables.
My assumptions:
The substance present in cassia bark necessary to make the Birmingham balti is in relatively low amounts, which accounts for the seemingly large quantity (55 g) used and the 1 h boil
Oil of cassia (mostly cinnamic aldehyde) is a likely candidate for this substance. Essential volatile oil. 1 - 2 %, dependent on cassia quality. Also used for making chocolate, liqueurs and in perfumery. Intense honey-like sweetness.
Not enough CA, or some other component of cassia present at low levels, results in failure/blandness (MKI base).
Difficulties
Boiling in water is relatively inefficient (compared to closed distillation); some loss of CA is inevitable during preparation of the stock.
Variability of cassia quality within/between brands. There are over 600 varieties of cassia, many of which differ in quality. Some varieties are scraped during harvesting. Found scraped and unscraped cassia in one bag of Yekta cassia (China) and not in another (Yekta, Iran). Lack of uniformity in many bags of cassia. Size (length) and thickness. Fudco brand was (usually) the most uniform I have come across, but no longer available at Sainsbury's.
Pressure cooking various cassia testers (within/between brands) gives stocks that differ in colour and taste/sweetness. However, colour is not necessarily an index of concentration. Taste testing also inconclusive (senses are swamped too quickly). The stock should be sweet though, enough to sweeten the batch of gravy, noticeably. Onion sweetness (lack of it) unimportant for the gravy. The 1 h veg boil will not melt the onions. Yet, the finished balti should be nectar sweet.
Solutions (more like ideas to reduce the likelihood of a negative outcome)
Some of this also already mentioned by Mick. For example, cassia thickness, increasing cassia amount used in the recipe. This is how I will do it until someone tells me otherwise.
If practical, don't buy cassia online. Use a large supermarket that stocks more than one brand and examine as many bags as you can before buying. Ignore what it is says on the label. Select the thickest and largest (length and width) bark you can find, unscraped. Also look for uniformity, semi circular appearance, outer coarseness etc. The rougher the better. Avoid bags full of small or tiny bits of cassia.
Consider increasing the amount of cassia to make the stock. Potentially the best fail-safe if experiencing problems. For the most recent batch (last week) the cassia I had was on the thin side. Upped the amount (110 g). I firstly put it all in a heavy plastic bag and gave it a good bashing (heavy rubber mallet). Also cut/broke up the other spices where possible. You could try doing this with a large mortar and pestle, but I think you will create a lot of dust. I was able to make this particular stock in a pressure cooker, with 1 pint of water as usual, and with very little reduction. A standard simmer/boil may need more water. This latest batch of MKI base is spot on. Finally, for now, the oil in this base needs to separate for the balti. This is an important part of it. The base is heavily spiced. It all comes together in the seasoned oil, which will also season the additional oil (and anything else) added at the pan cook stage. Mick's Birmingham balti. The ultimate BIR curry, when you get it right. I promise you.
Rob