Caerphilly comeback

[mikekchar]

I have a lot of problems with Gavin's recipe.  It makes very little sense to me.  I'll jot them down here since I keep meaning to do that...

First, he's got 1 tsp (5 ml) of 200 IMCU/ml rennet for 10 liters of milk.  That's 1000 IMCU / 10 liters or 100 IMCU per liter.  Normal is about 40 IMCU per liter. He admits this is twice as much as he would normally add, but says nothing about *why* he's adding that much rennet.  I can't see any way this isn't just going to cause problems.

His cultures are weird too.  I mean, it's fine, but why MA4001 *and* M030?  Adding LL and LC twice...  And of course, it *isn't* M030 when he's using it, it's Mad Millie...  So, one sachet of Mad Millie, which is normally enough for 4 liters of milk (but it's Mad Millie who thinks that 60 IMCU is enough rennet for 4 liters of milk, so who knows...) 1/8th of a tsp of MA4001...  (rant: Volume measurements for something that's powdered... i.e. he want's completely random amounts of culture...)  Anyway, NEC says 1/2 tsp for 6-12 gallons (i.e. don't measure it this way because you'll get potentially half or twice as much as you wanted).  But 1/8th tsp is enough for 2-4 liters.  So his total culture is about 60-80% of normal.  He ripens for 30 minutes, so we're at a pretty high pH when adding then rennet.

Even at the high pH, with 100 IMCU per liter of rennet, this is going to set *quickly*.  I can't imagine it will flocculate more slowly than 10 minutes, which means that his 40 minute total set time is a 4x multiplier.  Of course it wasn't firm enough for him, so he waits another 10 minutes.  But what the heck was wrong with that break?  I mean, it's absolutely straight edges.  It's flowing clear whey.  What the heck is he looking for???  I mean intentionally adding twice as much rennet and then intentionally letting it sit until the curd rips.   Is there method to his madness?

Then he's stirring for 40 minutes while he raises the temp from 32 C to 33 C.  Yes.  One. Whole. Degree.  WTH?  But even if we just say he isn't cooking it (which is basically true), why not?  There are a couple of problems here.  First, it's going to potentially retain whey.  Granted, he's allowed the curd to firm to the point where the curd rips when cutting and so it will be draining whey faster.  So... maybe that's why???

I've got to stress that the drain is at 30 + 50 + 40 + 5 minutes = 125 minutes at 32-33 C, with 60-80% of the normal amount of culture.  He's going to be coming out at about a pH of 6.4 or 6.5, rather than the 6.1 or so you would normally want to aim for.

He drains into a slab for 10 minutes total and cheddars for 20 minutes.  30 minutes total.  So the whey hasn't had a chance to totally drain from the cheese.  We're at a pH of something well north of 6.  And... it's time to mill the curds???  His idea of "thumbnail size" is pretty strange, but to be fair he *has* to mill it that small or else he's going to be making something that resembles a crescenza. 

Time for the salt.  Again volume measurements for powders because we *like* adding random amounts of ingredients.  That's a whack of salt, though, no matter how you measure it. Caldwell actually adds a very small amount of salt at this stage so that the cheese will acidify slowly as it is pressing.  As he's adding the curds to the basket, you can see that they have all stuck together.  His tiny, baby thumbnail chunks have turned into finger sized chunks.  This is a very good indication that the pH is indeed greater than 6.

Stick it into the press and right away 5kg of weight, just to lock in those juices.  You can see a good 100 ml of whey sploosh out in the first 3 seconds.  The saving grace is that he *has* drained it for 30 minutes and added salt, so it's not crazy to add some weight.  Not that amount of weight, but a little would be fine.  He mostly gets away with it, though.  And, of course, more random amounts of salt when flipping each time.

Now, I have to admit to being a bit perplexed by the cracks in the cheese on the first flip.  At that pH and moisture level, I would really expect to instantly close the rind.  So while I'm being very flippant in my remarks, there is probably something going on there that I don't quite understand.  Possibly it is acidifying faster than I'm expecting.  Every time he talks about Caerphilly he's at pains to point out that it is known for being a very salty cheese.  This is completely incorrect.  I think he's confused because miners used to eat it to help with their salt levels in the mine.  But almost all cheese has a fairly high level of salt.  If you are eating 100 grams of cheese, you are getting quite a lot of salt.  But I wonder if his buckets of salt are what are drying out the curd to a certain extent.

When he opens the cheese, it's a mess.  "There's a fair amount of cracking, but that's typical of Caerphilly.  I've found that before".  That's the initial over pressing.  Really, really classic.  The paste is *way* too soft and springy.  Whey Caerphilly *is* crumbly, it's not crumbly in the way that his is.  Again, that's from whey being trapped and refermenting.   I think the springiness actually comes from draining at such a high pH.  I was reading some comments from linuxboy about Beaufort and that the development of the protein really requires draining at a very high pH.

Anyway, as a Caerphilly, it's a total failure.  As a yummy cheese, it looks yummy.  I realise this is a fairly negative post, but I hope some people find it useful.

[bansidhe]

Hmmmm....  I did think that was a crazy amount of rennet.  I used slightly less but still a lot.  I also thought that was a lot of salt.. but That's what I used.  We'll see.  I am hoping that most left with the whey.  We'll see.

As for the culture, I used MA4002 and C101.  I decided on the C101 when I saw it was used in Wensleydale.  I didnt think it was a crazy amount of culture, since some recipes (cotswold) call for a full packet of C101 for 8 its.  Ive never had cotswold either but the make seemed a little similar.

As for the temp increase. Yeah, I thought that was insane.    I stirred for 10minutes then raised to 93-95 over ~30 minutes.

   My curds did not close right away.. it took the full amount of time.  We'll see how my two batches turn out in about three weeks.  After reading your comments Now, I guess I will make the Caerphilly in the 200 cheese book..  I need to see how that recipe alters the result.

As I said I have never had Caerphilly so I have no idea what it's supposed to taste, smell or look like. 

I decide to try Gavins recipe because his make for Bel Paese was my first successful cheese and it was awesome. :-)
I'll definitely be showing pics and sharing comments on the Caerphilly in a few weeks.    :-)

Since now I know, thanks to the information you presented, that Caerphilly is NOT a salty cheese I am a bit worried.
Darn.  I hope I haven't blown 4 pounds of cheese!

I'll try it again.  :-)

[mikekchar]

Gavin's recipes tend to be yummy, even if they are strange.  Bel Paese is a good example.  I think that's a really good recipe, actually.  But I don't think you are likely to make something similar to Bel Paese 

[OzzieCheese]

@mikekchar
You do bring up a few - well quite a few - good points and questions. I don't pretend to be in Gavin's mind, maybe you should ask him directly. He's a nice chap and will reply. My original Caerphilly was made using a recipe in "Making Artisan Cheese" by Tim Smith and if I remember it - was quite a few years ago that his timings were quite good and the cheese was pretty good if I remember.

[OzzieCheese]

My make sheet has the timing and quantities that I use but the bottom of the sheet has these.
Texturing:  This helps to drain the whey and increase the pH of the curd mass.  Drain for 15 minutes into a colander lined with Cheese cloth. While still in the colander maintain the 34 C temperature by covering the curd mass with a clean lid or tea towel.  After draining, return the drained curd mass to the pot, cut the mass into 2-3 slabs and stack on one another. Change the stack every 10 minutes. Caldwell's description is vague here so I'm going to slice and stack much like a Cheddar.
Temperature:  Caerphilly has a small cooking temperature rise about 3-4O C (6-8 OF).  Don't fret if it sneaks a bit past but the point is the slow rise and then keeping it there.
Pressing:  try to keep as warn as possible. Summer is ok but winter, going to try to press inside a double vat setup to keep it warm.
10 minutes 5kg (10lbs). Remove from the mould remove the cheesecloth turn cheese and re-dress- Work fast as the curd mass will be very fragile at this stage. The Secret is --- don't expect a cheddar 'cos it's not.
10 minutes 5kg (10lbs). As above.
25 minutes 7kg (15lb). Repeat as above. Keep as warm as possible.
Unmould and salt the outside of the cheese, redress and press again 15Kg (~35 lbs) for 16-24 Hours – Goal pH is 5.3 at this stage.
Caerphilly does have a slightly open texture and the press should just make the rind close and there should still be clear whey dribbling out.  Don't over press this one as the cheese cultures are still producing lactic acid. This is best eaten at 6 weeks it will have a nice tangy taste and after 8 weeks it mellows out.

and this is from quite a few years making it.

[mikekchar]

I went back to look at your recipe again Mal, and I like it quite a lot (which doesn't surprise me one bit).  For me, I think the secret is that you want to cheddar at a higher than normal pH, mill and salt lightly, then press allowing it to acidify in the press.  This lets you do a "mostly cheddar" with a very much weaker press than you would need for a real cheddar.

Yeah, I should probably get in touch with Gavin.  I think I have a bit of OCD kicking in when I see recipes like that and I can't stop from saying something.  Contacting the author (which should be my first course of action) is hard for me to do :-(.  But if you're gunna complain, you shouldn't do it behind people's backs...

[bansidhe]

Interestingly, I found -  https://blog.cheesemaking.com/caerphilly-almost-instant-gratification/. which uses Gavin's older recipe.  That link is a NE cheesemaking type link.  This gives me a bit more confidence in Gavin's recipe if they would cite and use it. 

[broombank]

I find the expensive ph sticks with a 4-7 range made in Germany at around £20 ($26) for 100 to be very useful. (The cheap ones from eBay are useless) Under a good light the colour changes are reliable and allow me to make changes ( like shredding curd or salting ) at the correct ph level. I think I feel like you do that unless I invest a sizeable amount of money in a really good ph meter and can put up with the constant need for calibration,  I would rather rely on the sticks. I can try to calibrate my palate instead but its all a bit too vague.

[OzzieCheese]

I think there might be a bit of confusion in the rennet strength and I use the same rennet as Gavin, as we get it from The same place. The concentration is 200 IMCU. So it doesn't matter if you add 1 or 5 mls it is  same concentration it is diluted when you add it to the milk. So, 1 ml would be diluted in 10 litres and diluted, but it is still 200 IMCU. Which by the way is single strength rennet.

[mikekchar]

Yes, the rennet is rated by IMCU per ml.  So if you have 200 IMCU per ml and you have 1 ml, then you have 200 IMCU.  If you have 5 ml, then you have 1000 IMCU.

However, IMCU (International Milk Clotting Units) are defined by the amount of milk it clots at a certain temperature, pH in a certain amount of time.  The actual standard is a bit complicated (and there are several methods for defining it), but basically 40 IMCU will clot 1 liter of milk at 36 C, at a pH of 6.6 in 36 minutes (I think... I often forget the details and the actual standard is not available unless you pay stupid amounts of money).  "Clot" here means get to a "clean break", which is essentially a multiplier of 3.0.  Another way to put it is that 40 IMCU of rennet in 1 liter of milk at 36 C and 6.6 pH will normally have a flocculation time of 12 minutes.

If you have 10 liters of milk, then you will need 400 IMCU to have a 12 minute flocculation time at 36 C.  With 200 IMCU per ml rennet (the definition of "single strength" rennet), that's 2 ml of rennet.  The more rennet you add, the faster it will flocculate.  It's not linear, though.  So if you add 800 IMCU (4 ml of single strength rennet), it doesn't flocculate in 6 minutes, for example.  I'm not sure how fast it would be, but probably somewhere on the order of 8-10 minutes.   If you add less (say 200 IMCU), then it will be slower -- and it *is* pretty near twice the time in my experience.  About 25 minutes.  But again, it's not really a linear thing.  You have to experiment to see how it goes.  If you add a quarter (100 IMCU), which is what you would normally do for a semi-lactic cheese, you actually have to wait until the pH of the milk drops to about 5.3 before the rennet works fast enough to set a curd.  This normally takes several hours.

Generally speaking, as long as you have good milk, enough calcium and enough rennet, your curd will set.  The concentration of rennet (in terms of IMCU per liter of milk) only really affects the speed that the rennet will work (in my experience).  Of course, excess rennet will remain in the cheese and continue to break down the casein over time, which can add bitter peptides to the cheese.  So there is also a maximum amount that you want to add.  Adding more than that can make the cheese bitter.  And, of course, the speed the rennet is acting affects the texture of the cheese as it is cooking and draining as well.  Possibly this is why Gavin's recipe does not knit as well as I would expect even though the pH is very high (I've never used that concentration of rennet before, so it's hard for me to say).

TL;DR: More rennet means faster coagulation.  40 IMCU per liter of milk (0.2 ml of single strength rennet per liter of milk) gives you a flocculation time of 12 minutes at the normal pH that milk is at after ripening.  This gives you a normal curd set in 36 minutes with good milk (i.e. raw milk).

As an aside: Why is single strength rennet set at "200 IMCU per ml"?  200 seems like a weird number to choose.  I *think* this is because 5 ml is 1 tsp.  That gives you 1000 IMCU.  1000 IMCU (1 tsp) is enough to coagulate 25 liters of milk, which is just over 5 imperial gallons.  My understanding is that the standard measurement technique for IMCU was developed independently by both British and German researchers and I think the standard was picked to accommodate both cultures.  Apparently it says in the standard why that value was chosen, but... stupid amounts of money.  I'm curious, but not *that* curious...

[momo]

Mal - I appreciate your description of your process, + recipe and photos. Could you give more detail about your stockpot setup - what size are your 2 pots, do you use something to raise up the inner pot, etc?  Thanks!

[bansidhe]

I don't know where to find non-homogenized pasteurized milk here.  How might things change for me when I make this cheese? Is there anythngn I should do differently to ensure a better results?

[mikekchar]

I've been experimenting with that recently and actually a few months ago I made a Caerphilly with homogenised milk (still haven't eaten it -- it's getting pretty well aged...)  The main thing is that with homogenised milk, the curds tend to "shatter".  Before you know it, you have essentially curd dust instead of curds.  They also retain moisture more.  So I've been experimenting with cutting 5 cm (2 in) columns in the curd at a multiplier of about 2.0-2.5, then waiting until one would normally cut, cutting it like normal, *then* for the next 15 minutes only stirring to keep the curds from matting.  The technique I do is simply, very  slowly transport the curds from the bottom of the pot to the top.  Then wait about 5 minutes and do it again.  After that 15 minutes, switch to stirring with your hand rather than a spoon.  Use your fingers to tease apart the curds, but otherwise be very gentle.  Again, you may want to stir for a minute, rest for a minute for almost the entire duration.

Once you drain, the curds act reasonably normally and you can cheddar as normal.  I find that the texture of the curd feels a bit "gritty".  I'm not sure how to explain it.  It's not as smooth and gel-like as using unhomogenised milk.  However, it doesn't seem to affect the final cheese in the end (I can't tell the difference in texture after aging it for a week or so).  Probably as the proteins break down during aging, it ends up pretty much the same.  I made my "Tomme that cracked" the same way and it was a phenomenal cheese (IMHO, though it was over-moist).

With that said, homogenised milk seems to hold on to fat better (it's wrapped around the proteins rather than just being caught in the spaces of the gel) and so the resultant cheese tends to have a higher yield with more fat.  I'm thinking that cheeses without much stirring will actually work better with homogenised milk.  However I made a camembert a while ago and I would have to call it a failure.  The curd never really came together the way I was expecting and the cheese had gaps, which was super weird.  I was using a completely different technique, though, and it might have been due to that.

[OzzieCheese]

Mal - I appreciate your description of your process, + recipe and photos. Could you give more detail about your stockpot setup - what size are your 2 pots, do you use something to raise up the inner pot, etc?  Thanks!

My pots are a 12litre stockpot sitting in 16 litre one of the same brand. The handles of the small pot sits on the rim of the larger one. It keeps the smaller one from hitting the bottom and therefore keeps a complete water jacket around the milk. I monitor the water temp in the larger one and found that accurate temperature control is very easy. During the initial heating, I try for a 2 degree C differential  between the the milk and the water and remove them from the heat just before the target temp is reached. The Thermal load in the water will be sufficient to continue to raise the milk temperature.
If you find the milk temperature rising too quickly you can lift the smaller one out and remove water from the large one and add cooler water to the larger pot to bring the thermal difference back down. After all you are not wanting to heat the milk as fast as possible but gently over time.