Poor efficiency brewing 7 bbl on 20 bbl system

[Microbeergineer]

Greetings and Salutations!

I have a 20 bbl brew house that I need to occasionally brew 7 bbls on. This has decreased efficiency which, obviously, I need to improve. I’ve tried a number of things based on other threads, braukaiser, and the pdf from Van Havig to improve the process but I’m stumped by my failures. I know there's a dozen or more efficiency threads out there but I figured I’d try a post since the situation may be slightly more unique with the tiny mash on a big system. I hope someone can spot what I’m overlooking.

I always make single infusion mashes since there is no heat for the mash tun. I also do not have a lauter tun. Here are some stats:

BH efficiency: high 60’s to low 70’s% (full 20 bbl batch low 80’s% for comparison)
Conversion efficiency: mid 80’s%
Liquor to grist ratio: 2.6# / 1# (1.25 qt / #)
Mill gap: ~0.07 inches
Shakeout: ~70% above sieve #14
Mash time: 45 min rest, 15 min vorlauf, no stirring during this time but occasionally cutting the top with a paddle and filling gaps on the side if spotted
First runnings pH: between 5.3 and 5.4 typical (cooled sample)
Last runnings: between 1.018 (4.6 P) to 1.024 (6.1 P) (I presume this high last runnings is due to fast run off time)
Run off time: ~60 min
- I know slowing transfer to kettle may help but my flowmeter’s minimum flow is 5 gal/min. On larger batches I run off faster and aim for 75 min.
Sparge temp: 170 to 175 F

Cheers!

[Moutas]

My guess would be that you loose too much sugars in the last runnings.

If you have enough space to collect more wort try going a little deeper, like 3,5-4plato. Or if you dont have enough space, try making a little thicker mash/water ratio so you'll have a more concentrated wort during lautering.

Cheers

[Alpha King]

Your biggest challenge is mash bed depth. You are only filling the tun to approx 40% of it's average design parameters. A quality fabrication company will design a mash tun with an aspect ratio that will provide high levels of grain efficiency for a mid range beer wort ie:.5.6% pale ale. While your initial run off may be fine your sparge effectiveness suffers from the hot liquor moving through the shallow mash bed too quickly. This is indicated by the high gravity levels of your final runnings. You could see some improvements by slowing down your overall transfer time but it would have to be excessive - possibly 2 + hours. Since your final runnings have so much retained sugars you could transfer more wort (8.5 bbls ?) and go for a longer boil. You should also consider increasing your liquor to grain ratio to 3 : 1 for some slight improvement.
Cheers, Marc Martin (Northwest Brewery Advisors)

[dick murton]

1.018 (4.6 P) to 1.024 (6.1 P) is incredibly high for last runnings, which is part of the reason you are having such low extract efficiency. A couple of things you could do to improve things.

Firstly, you should be able to mash in a little thicker, say 2.25 litres liquor to 1 kg dry grist. You probably won't need to, but is the resultant wort attenuates more than it usually does, raise the mash temperature, 0.5 deg C should do it, though when I did 1/4 an 1/3 normal brews at one brewery, I had to raise the temperature a complete degree C. Critical to achieving consistent mash temperatures, assuming you mix liquor & grist in a mash mixer, is to preheat the mash tun. So the effect of this change in liquor to grist will be to make the sugar concentration in the mash somewhat stronger.

Secondly, run the vorlauf just long enough so you just transfer the water under the plates onto the top of the mash. As soon as the gravity starts to rise, you are mixing in strong wort and wasting the usefulness of the underplate volume. By recirculating for 15 minutes, you are diluting the first worts before you even start to run off. Sure, the clarity might not be quite as good, but I doubt very much if you will see any significant change in clarity (or flavour) of the final beer. By only transferring the underplate liquor, you are using this as the first sparge, and the gravity of first worts to the kettle will be considerably higher. You should be able to get something over 1.090 (22.5) from your very early runnings, before dilution by the underplate water starts to dilute. If you do decide to thicken the mash, you can sparge for longer.

The final change is simply to take longer to run off, as has already been suggested. If for example, you cut the runoff time further than the 75 minutes for a full brew, then you are not giving time for the sugars in the grain particles to migrate out into the sparge. Check what the flow actually looks like running into the kettle, and use this as the basis for controlling the flow manually.

Your extracts in the full 20 brl brew are also very low - expensively so. I would expect a runoff time around 150 minutes for a full brew, and because of the shallow bed on the 7 brl brew, still as much as 90 minutes. Assuming you are vorlaufing for 15 minutes on a full brew, I would cut that back so only the underplate volume (as above) is transferred onto the mash. With something like 5% ABV beer, 20 brls - I would be looking for 90% efficiency, ideally more.

[dick murton]

Great, the above drawings have got in the wrong order.

The first one, start of vorlauf is the middle one (I have used a drawing that indicates the vorlauf goes through the sparge system, which may or may not be possible depending on the method of sparge / vorlauf distribution)

The next stage is the bottom one, where the water under the plates has largely been transferred to the top of the mash, and has been replaced by an increasing strong mix of wort mixed with residual underplate water.

The final stage of vorlauf, which is where I suggest the vorlauf is stopped is shown in the top diagram, and strong wort is about to be recirculated to the top of the mash. At this point, runoff to the wort kettle should be started.

[Microbeergineer]

Thank you Moutas, Alpha King, and Dick Murton for the responses!

Seems like the general consensus is that my sparge and run-off are the culprits as indicated by the high last runnings. I was hoping for a different problem because my flowmeter does not read properly below 5 gal / min. I’ll have to get creative about making an in-kettle indicator for the “kettle full” volume and skip the flowmeter. I’ll try to slow the transfer first before collecting extra wort. I’d rather avoid boiling for extra time to save water and natural gas. As I’m sure everyone knows, resources are becoming expensive!

As for mash thickness, I was under the impression that a thinner mash (say 3:1) would increase brew house efficiency compared to a thicker mash (say 2.25:1). Is this thought inaccurate or am I missing finer details? Dick, I’m assuming that you’re suggesting to increase mash temperature slightly because the thicker mash offers better thermal protection for beta amylase which may increase the fermentability of the wort? Am I also correct assuming that 2.25 L / kg is the same as 2.25 lb / lb? I tried to convert between the two but I’m not sure if my math is correct.

(2.25 L water / kg grain)(3.785 gal water / L water)(8.156 lb water / gal water)(0.45 kg grain / lb grain) = 31.5 lb water / lb grain ???
(An online resource stated that water weighs 8.156 lb / gal at 160 F or 71 C)

My system’s vorlauf goes through a side arm that splashes wort against the side of the mash tun. I’m sure the same concept applies that I’m moving full strength wort to the top with the weak wort. (Awesome diagram, thank you.)

Next brew goals:

• Increase mash thickness
• Reduce vorlauf time from 15 min to 10
• Increase lauter / sparge time

[dick murton]

I would look at the wort concentration when vorlaufing, and use that and taste (unless you have a refractometer there will not be time to measure gravity) to determine when to stop the vorlauf and start runoff - at least for the first couple of runs, then of course the time taken to start getting stronger worts should be OK. I wouldn't simply rely on shortening the time the first time you reduce the vorlauf, though undoubtedly there will be some benefit in terms of reducing the sugars pumped back to the top of the mash.

2.25 litres (water) = 2.25 kg = 4.96 pounds - I wouldn't worry about the slight difference in density due to difference in temperature. And 1 kg (grist) = 2.21 pounds

So you want 4.96 / 2.21 = 2.244 pounds water per pound of grist - so yeah, call it 2.25 pounds water / pound grist

The reason for trying out a slightly thicker mash is simply to allow you to wash out more sugars by having an increased volume of sparge - nothing to do with conversion efficiency, all to do with the extra 35-40 litres (9.2 - 10.7 gallons) giving greater time / gravity differential to wash out the soluble sugars

Possible mash temperature increase to counteract the increase in enzyme stability / activity.

[Rob C]

I always pre heat my mash tun on the first brew and i started recording vorlauf temp as a better indicator of mash temp. On small batches the mash temp can drop much quicker than a full batch. To compensate for that loss in temp Ive sprayed extra hot liquor to further increase the temp before mashing in. Other than that I would make sure you have a layer of water on top of the mash bed during the sparge. Good Luck!

[Microbeergineer]

Alright, we’re getting somewhere!

TL;DR Efficiency has increased to 82%. This is roughly equivalent to my current proper sized 20 bbl batches. Still poor by most accounts but nice improvement with some additional knowledge to carry over to improve larger batches.

Now for the details!

My target was 2.25 # liquor to # grain but I ended up mashing in at 2.46 (Thank you Dick for the clarification to the conversion calculation). My new target first runnings was 1.098 (23.3 P) but I peaked at 1.079 (19.2 P) which was only a 78% mash conversion efficiency. Room for improvement here somehow.

The mash tun was preheated to 145 F (Strike at 166 F, Mash at 152 F) with grant temperatures lingering in the low 140s, upper 130s during transfer as shown in the chart below. In retrospect, this may have been because my 30 gal grant was always roughly 1/4 full. I reduced the grant level to a minimum closer to 2/3 of the way through transfer which may have shown a higher temperature due to less mixing of cooler wort lingering in the grant. All these temperatures were interesting because I sparged at 171 to 170 F.

I have a refractometer and I took several readings throughout the vorlauf. My goal was to start transferring to the kettle after 10 minutes but I got so carried away with readings that I started transfer after 20 min (Vorlauf started at 8:26 am and transfer started at 8:46 am highlighted in bold in the chart below). It was interesting to see how the gravity slowly increased and peaked at roughly 14 minutes into vorlauf. As mentioned above, the grant was more full than needed so adequate mixing may not have occurred for the most accurate readings.

So for 480 # grain, I mashed in with 146 gal and sparged with 198 gal. I used some bubble gum, a clamp, and a wedge built out of shims to create a measuring stick for a target volume indicator in my kettle. Transfer took 2 hours with a final runnings of 1.010 (2.5 P) which is the lowest it has ever been. Bonus was that my kettle gravity was 4 points higher than I was expecting. I’d call this brew a huge step in the proper direction thanks to everyone’s tips, tricks, and guidance.