samuellaw178 wrote:Mark, you seem to already have a grasp on what's going on. Sorta like 'been there done that' for you. Have you figured out what's happening there?

The TDS pick up whenever an immersion brew drains through the bed seems roughly predictable, I'd observed it, just not tried to quantify it before, but based on yours & Chris's figures (plus my own records of immersions that drain through the bed), it must be in the region of 11% prior to dilution in the vac pot.

The long and the short of it is that vac pot/siphon doesn't need any special adjustment to extraction yield calculation (drip mode, or immersion) in normal use. This isn't surprising as it was one of the methods in popular use when yield analysis was born.

Dick, I think Sam said later into the thread that 70-80g was retained in the lower bowl (I've been working on 80g), not 100g as first suggested.

Sam, whilst you may have a method for determining TDS within a broadly useful range, you're not going to get comparable yield figures without the VST software, the smart phone apps aren't expensive for what they do.

Lots of really helpful info in this thread! Thank you to all whom have contributed. The timing couldn't be better as I have recently been dwelling on my vac-pot technique. I used to be in the habit of excessive updosing, which would sometimes taste good, others it would be completely overbearing, as my TDS was through the roof, and the ext % was barely hitting 18.

Due to a combination of acquiring a refractometer & the information throughout this thread I have since changed my parameters to match that of basically how I would make an aeropress. The results have been incredible, and my brews are now hitting 21-22% at a TDS of 1.3-1.4.

Regarding the water left in the bottom-- my setup is rather unique as I am using a TCA-2 globe with a TCA-3 funnel. The stem is a bit longer than the stock piece, which enables more water to travel up, but still leaving enough to not boil dry. I measured it on a scale to be 12 grams out of a total 270 brew water that didn't make it up. Negligible, IMO considering efficiency of a siphon extraction.

I made back to back pots this morning within .02 TDS from one another of Gracenote's Guatemala Pulcal. Super sweet cup with a lovely mouthfeel & outstanding aromatics.

Peppersass wrote:Using inputs of 35g of coffee, 395g of beverage and TDS of 1.71%, the VST software to computes the extraction yield to be 24.28%. More over-extracted than you thought

Just to confirm does this use the immersion or pourover mode? I just want to wrap my head where I went wrong, or was it due to different assumptions used. 24% and 21% are very different I agree.

Peppersass wrote:If you updose to 44.7g and increase the total water to 590g (includes 100g retained in the lower globe) and adjust the grind so that your draw down time remains the same, and use the same 3 minute steep time, you should get an extraction yield of 20.02%. I calculated that using the same 1.71% TDS and 395g of pre-dilution beverage so that you'll get about 1.35% TDS after dilution.

Thank you for this. I will try it and report back.

MWJB wrote: Sam, whilst you may have a method for determining TDS within a broadly useful range, you're not going to get comparable yield figures without the VST software, the smart phone apps aren't expensive for what they do.

Thanks Mark. I sure would like to get the software or even the meter itself. But right now, it's not financially wise because we're saving up for my wife's education fee (Been working our arse off for that). We're trying to spend less whereever possible (except for the occasional 'asset' gain which I reckon could be sold off later shall the need arises).

After these exercises, I get to understand more why VST refractometer is useful and how it will make the process easier. It's expensive but the VST can help to understand the brewing process a little more clearly (especially if the method is new or the palate is not reliable for some reason). But I also think it's easy to fall into the trap that, if the user is not sure of the assumption used and use the number given blindly (sometimes even with the wrong protocols). Same caveats with all tools I believe.

ex trahere wrote:Regarding the water left in the bottom-- my setup is rather unique as I am using a TCA-2 globe with a TCA-3 funnel

That's an interesting way to use it. I have been thinking the same too, but am not sure if there's alternative that can fit on the Cona.

MWJB wrote:Dick, I think Sam said later into the thread that 70-80g was retained in the lower bowl (I've been working on 80g), not 100g as first suggested.

OK. If I use 75g for the water retained in the lower globe, I get 1.63% for the undiluted %TDS, 420g of undiluted beverage and 24.40% extraction yield. Not terribly different. The brew water is still off by 10g, though (should be 485g and I'm getting 495g.)

The revised recipe to get 20% extraction yield at 1.365% TDS is 44.7g of coffee and 592g of brew water.

samuellaw178 wrote:Just to confirm does this use the immersion or pourover mode?

I used the Immersion mode.

samuellaw178 wrote:But right now, it's not financially wise because we're saving up for my wife's education fee

If you have an Android smartphone, the VST software is currently on sale for $29.99. You've probably spent more than that for the coffee used in this experiment.

Peppersass wrote:The revised recipe to get 20% extraction yield at 1.365% TDS is 44.7g of coffee and 592g of brew water.

Thanks Dick. I've tried your suggestion this morning and got the TDS reading. Same parameters (Bean,grind,temp).Steep time, including drawdown, was about the same(~3 min total, 5 sec faster if anything). I got 1.51% as the final TDS.

Coffee Weight : 44.7g
Brew Water Weight: 592g
Final brew weight : 502.2g

Water Retention + coffee ground (44.7g) : 113.1g
Water Retention : 67.3g

Evaporation : 22.5g

How did it taste?

With those results, the extraction yield comes out to 21.44%. The cup should have tasted a little less bitter than before, but stronger. The latter may have made it seem worse.

I would think with more coffee at the same grind the draw down time would be a little longer.

Try grinding a bit coarser. That should lower the TDS and extraction yield, and may speed up the drawdown.

FWIW, here's how I prepare vac pot:

1. 18g coffee and 250g of brew water in a 3-cup Yama
2. Target temperature around 198F
3. After water rises into top chamber, dump in grounds and push down with a wire whisk to fully wet the grounds (no stirring)
4. Reduce heat to prevent temperature from rising, but not so much as to start draw down
5. Steep for 30-40 seconds
6. Remove heat
7. When draw down begins, stir two or three times. Not too much, just enough to get a small vortex going
8. Draw down time is typically in the range of 90 seconds

Three minutes total may be OK -- you have a lot more beverage going on than I do -- but try a shorter steep and/or shorter draw down.

Some people accelerate the draw down by wrapping the lower globe with a cool wet cloth after removing the heat source. You could try that instead of grinding coarser, or do both.

I find that stirring has a dramatic effect on the extraction yield, though it may be more pronounced in smaller vac pots. But you might try less stirring and see how that affects the extraction yield and TDS.

Peppersass wrote:How did it taste?

It did taste better, I would think the subtler notes came out better/stronger. It was an Ethiopian Harrar, and there was a grape-floral note that I didn't notice before. Though, I had the impression that the 'delicateness' became more apparent, even though the TDS shoot right up there. There is still a little bit of after-bitter(just very slightly). Could just be my roast(took it about City+). Or probably just my palate is hopelessly hopeless.

My drawdown is rather predictable and faster. After so many brews, my obervation is it will start drawing down in 15-20s after stopping the flame, and completes in about 40-45s. May be I will shorten the steeping time.

As for stirring, I do it 3 times. First stir is when I put in the coffee(more like N action to fold the coffee into water); second after the ground forms a top layer @ 45s(another N action); and last one after I cut off the alcohol burner(Real stir). I will try your two-stir and see how that fares.

I think you'll be better off with less stirring. I would make the change more drastic so you can see the difference. Try just pushing the grounds under the water to wet them at the beginning, then one stir just as the draw down starts.

15-20 second between removal of the heat and start of draw down seems pretty long to me. Mine usually starts dropping within 5-10 seconds at most. This could be another reason for the over-extraction. Coarser grind might help, or try the wet towel trick.

Something important we should keep in mind when comparing notes is that the more you fill the bottom chamber, the weaker the vacuum effect is on the drawdown.

With 600 g in my big stovetop Yama brewer, the water violently shoots up to the upper chamber at nowhere close to boiling and is drawn down through 41 g of coffee within 90 seconds of removing it from the heat source. You can hear a loud hissing at the end as air is pulled through the bed and bubbles below. This also has the side effect of producing more brewed coffee because the water retention is significantly lower than other immersion methods.

My new Hario will safely take a maximum of about 585 g of hot water, though the 5 cup mark is more like 570 g (~590 mL). It's more difficult to control the temperature of the water in the upper chamber because there isn't enough pressure to push it to the top until it's a lot hotter. The drawdown is so weak it takes over 3 1/2 minutes with the same coffee and method as the Yama, and never really completely finishes. When I remove the upper chamber, the grounds are not dry and there is still brewed coffee that escapes. Using it at half capacity, it behaves much the same as the Yama.

This is kind of contrary to the common wisdom of only using vacuum pots at their full rated capacity. You have to weigh the extremely minor effect of dilution against much greater flexibility with brew times and methods.