You are here: Home

Need Acrobat Reader for PDF documents?

Hot and cold food smoking

As far as I know, the only page I failed to save from the old site was the one from The Online Cookbook on hot-smoked salmon. In fact, my adventures with hot-smoking salmon came to an abrupt end when I smoked 70-worth of fish for my daughter Sarah's wedding in June 2013. Probably because I didn't clean out the Abu Roken smoker thoroughly after each of the many batches, my little meths burner managed to burn a big hole in the heavy-gauge steel bottom.

Always at a loss to buy presents for 'the man who has everything', Patricia suggested that I might like a new Abu smoker for my birthday in January 2014. We did some research on the web and ended up starting me on a whole new hobby!

I'll try to recap the important things I've learnt about hot-smoking, but I'll start with the new venture...

Getting into cold smoking

When you hot-smoke food, it smokes and cooks at the same time. After salt-curing for a couple of hours, my usual portions of salmon fillet cook and smoke in about 20 minutes.

Cold smoking - the process that produces what most of us know as 'smoked salmon' -  is quite different. For a start, the food isn't cooked at all, so the salt curing is needed for preservation as well as for flavour, and seems from the contraductrory information I've found on the web to go on for much longer. The food is then smoked at low temperature - the cooler the better, it seems, to the extent that people doing it in hot climates often put trays of ice in the smoker! - for many hours. Clearly this is a much more challenging process, and as I write this I am right at the very beginning of the learning curve. I've produced one 250-gram batch of smoked salmon, which I judged to be over-salted and under-smoked, but which I ate with some enjoyment last week.

The equipment

Patricia felt that the cost of the replacement hot smoker wasn't quite enough as a birthday present, so we agreed that she could add a smoke generator for cold smoking and I would build a housing for it. A bit more research suggested that the racks and drip-trays I'd need might be difficult to find without the housing, so I decided to buy a cardboard one! No, really! I said you cold-smoke at a low temperatrure, and to prove this you can do it in a cardboard box. Well, actually two cardboard boxes, one inside the other.

The product Patricia ordered is a ProQ Cold Smoke Generator, bought from MacsBBQ via Amazon. This is a remarkably ingenious piece of kit in which wood dust will smoulder gently in a square spiral made from steel rod and metal gauze. I dimly remember from second-form science that a flame will not travel through a fine gauze, and this device is a superb illustration of that. Here's a picture taken this morning when testing some smoking dust which I've had for a few years. If you look very carefully you can see a curl of smoke towards the right side of the picture.

The cold-smoke generator under

The cold-smoke generator under test

The dust is contained in a square-spiral channel of metal gauze. It is ignited by a tea-light placed in a special holder under the beginning of the spiral (top centre). The smoulder point then moves round the spiral very slowly - this picture was taken after about two-and-a-half hours.

The other piece of equipment, which I bought from a company called SousChef, is the ProQ Eco Smoker. Basically, it consists of two currugated-cardboard boxes, one of which fits closely inside the other.

The ProQ Eco Smoker

The ProQ Eco Smoker

The flap at the bottom allows the smoke generator to be slid onto the drip tray, which rests on the bottom of the box, once it has started to burn.

The inside of the Eco

The inside of the Eco Smoker

The upper drip tray and the three racks rest on ingenious push-ins cut into the corners of the inner box.

The observant user will notice that the right-hand flap is somewhat curly. This is due to water damage caused by a burst pipe, which caused a rain shower in our garage just after I'd done the first test of the smoker. Fortunately, corrugated cardboard is easily repaired with ordinary brown paper and PVA adhesive, and as I write the smoker is ready for use.

I am less concerned about this water damage than I could be because I am already planning to build a more permanent box, probably from MDF. This will allow better control of the air flow, which is pretty poor with the cardboard version - particularly with distortion from the water damage - because the box top doesn't fit very well.

The first test

This happened before the deluge. I used a nice tail fillet of salmon which I salt-cured for a couple of hours in the way I'd always done for hot-smoking: a couple of hours liberally sprinkled - 'smothered' might be more accurate - with sea salt. This was then washed and dried with kitchen roll and smoked for about six and a half hours, which is how long the bag of dust provided with the smoke generator lasted.

The resulting smoked salmon looked and felt pretty convincing, and was more than edible, but was obviously over-salted and under-smoked. I vac-packed 200 grams and gave it to stepson Alistair, who assured me that he liked his salmon nice and salty. I then ate the rest over a few days.

A slight delay

Then came the deluge. On the Sunday morning Patricia went out to the freezer in the garage and rushed back to report a flood. When I got out there, I found well over a inch of water on the concrete floor and water running down the wall. Worse, the water had obviously run over the tongue-and-groove pine flooring I'd laid to provide a usable attic for storage. I put the lights on, hurried up the ladder and discovered that the water was pouring out of the polyurethane foam insulation covering a mains water pipe that emerged from the house wall and ran along to feed a tap inside the front door of the garage - very handy for washing the car. There was a waterfall down the brick wall and a strong jet onto the attic flooring. I rushed down and closed the main stopcock under the bathroom floor and this stopped the flow - but left us with no water supply. Then I remembered that the excellent plumber who had installed our new boiler and shower when we moved in seven years before had put isolating valves in the hot and cold feed pipes, conveniently just inside the eaves loft door in our bedroom. I turned off the cold water there, which meant that only our en-suite lost its supply. On Monday our new plumber arrived and replaced a length of pipe, which had been corroded by the cement mortar where it went through the wall to the garage.

It took the ProQ Eco Smoker box a few days tro dry out in the conservatory, which Mother Nature graciously provided with a little sunshine at the end of January.

Back on task

While I waited for it to dry, I did a test with the oak sawdust which had been lurking in the shed for two or three years, a very small amount being used for hot smoking. This was a gift from my fishmonger, Darren Jelley, from whom I buy fish on Worksop market every Wednesday. When I first rescued my Abu hot-smoker from the shed, I  bought a bag of the official wood dust on Ebay, and I happened to mention this to Darren when buying some salmon from him. He said that there was a smokery next door to his premises in Grimsby and that he could get me some dust from there. He was as good as his word, and a couple of weeks later pulled a bag out from behind his van. It was about the size of a hunredweight sack of coal, and almost too heavy for me to manhandle across the car park. Darren's passing shot was 'Let me know when you need some more.' Not in this life, I thought!

I had read during my recent web-wandering that dust for cold smoking needs to be really dry, so I had put a roasting tin full of Darren's dust in the oven at 100C for 24 hours. At 9:15 yesterday I loaded the cold smoke generator with this, being very careful not to overdo it and provide a bridge between the arms of the spiral, and put a lighted tea-light in position. The dust quickly started smouldering, and went on very nicely after I had removed the tea-light until I checked it at 22:45 - twelve and a quarter hours later. All the dust appeared to have been consumed, but when I knocked the black residue out into my workshop bin there were still some bright sparks, so I had to put the bin outside for the night!

During the day I had been experimenting with some of my dried dust. The dust supplied with the smoke generator had seemed finer than mine, some almost like flour, so I had used our fine kitchen sieve to separate some of the dried dust into two fractions. I had then gone on to put some in the blender, but that had not managed the dust very well. However, our small food-processor (if you've read The Online Cookbook you'll know that we couldn't possibly have just one!) did a much better job, with the dust spiralling round the goblet violently and coming out much finer.

Having established e            arlier that the dust was fine just as it was, it was probably pretty silly of me to use a 50/50 mix of the dried and the processed dusts, but that is what I did this morning (31 January 2014). It proved quite a lot harder to light than the coarser stuff, and I was thinking of emptying and re-charging the generator when it finally caught. As I write, the smoker has been running for three hours and is looking good.

Finding useful information

I consider myself an expert web-user, skilled in finding the information I'm looking for. I have to say that my quest for hard information about salting and cold-smoking has been one of the most frustrating ever. Many people have committed their experience to websites, but very few have produced anything clear. The salt-curing and smoking times described vary wildly and I haven't found much unanimity about cures. I was so frustrated that ended up buying a book - not an e-book but one of those things with printed paper pages!

My cold smoker has stuff printed on the outside, amongst it a web address: This would be my source of hard information. Unfortunately the Guides section's link to cold smoking yielded just two lines:

Adding smoked flavour to foods such as cheese, salmon, salt, nuts, butter, etc without cookingCold Smoking is easy to do at home.

'Thanks a bunch!' I snarled when the page opened. I hit the Contact link and used the form to complain politely but bitterly. To my delighted amazement, I got a phone-call within 24 hours from a very friendly and helpful guy who directed me to their Links page and particularly to this link:

Smoky Jo's - Learn to smoke your own cheese, fish, meat and more! Unique and fun courses for anyone who loves smoked food and enjoys creating delicious meals at home.

He also told me a bit about Jo Hampson and her book Smoking Food at Home. By the end of the day I had ordered it from her Ebay shop and it arrived yesterday. I scanned, skimmed and searched, and found that it was certainly more informative and authoritative than any of the sites I'd found previously. I picked out a few pretty crucial bits of information and then dug the two hefty tail fillets of salmon I had bought from Darren on Wednesday out the fridge.

Getting serious

The first vital piece of information I picked out of Jo's book was a simple cure for salmon: two parts salt to one part brown sugar. I made a batch of this up using mostly Aldi fine sea salt and dark muscovado sugar.

The second was that you only cover the fish with something like 'a light covering of snow' rather than the thick, opaque covering I'm in the habit of using for a short cure before hot smoking (which was what I'd used for the first cold-smoked salmon, which explained the excess saltiness). So I did this, but unfortunately the sugar was in lumps, although I'd tried to squash them when mixing. Before I use the rest I'll put it through the food processor.

I used a lipped baking sheet which I sprinkled with the cure before laying the fillets skin-side down and sprinkling them with cure. That was at 15:00. I covered the fish lightly with clingfilm and put the package at the bottom of the fridge. At 22:45 I took the fish out of the fridge (7 hours 45 minutes curing time). There was a lot of liquid on the sheet but still some of the cure, now very wet.

I rinsed the fillets very thoroughly, rubbing with my hands to dislodge any residual cure. Then I blotted them off with plenty of kitchen roll and laid them on a cooling rack on top of the rinsed baking sheet. This went back in the fridge.

The fish was much darker in colour than it had been when fresh and was noticeably thinner and stiffer.

I couldn't find any tea-lights this morning, so I tried lighting the 50/50 coarse/fine dust in smoke generator with my blowtorch. I did this again at 09:55 and this time it was smoking fifteen minutes later. I have the impression that the coarse dust which I had used for the test yesterday burns more reliably than the mixture - probably because more air is trapped between the particles.

By 11:40 almost all of the first arm of the spiral was burnt black. By 12:20 the smoulder point (my own self-invented jargon!) was halfway along the second arm and by 13:45 the third arm had started burning. The first two add up to 240mm of a total 865mm. 240mm in 3 hours 55 minutes calculates roughly to a total burn time of about 14 hours 30 minutes. That would have the smoke stopping at around half past midnight.

It is now 17:00. We will see - although I am unlikely to stay up to see the finish if it is that late.

I have one nagging concern: I see less smoke escaping from the box than I remember during the first smoking, so I wonder if the 50/50 mix is working well. On the other hand, the dust used in the first smoking burned away in six and a half hours - less than half the anticipated duration of today's burn, so maybe the smoke will simply work more slowly but ultimately with the same effect.

At 1700 the smoulder point is halfway along the fifth arm - a total of 555mm in 7 hours. That gives a projected finish in a total of 10.9 hours - about 21:00.

I kept checking about every 90 minutes through the evening, and at 22:40 (13 hours after the burning had stabilised) I found that all the dust had burned away. The salmon was dry on the surface, quite firm and a very nice colour. I vacuum-packed one fillet for the freezer and clingfilmed the other for the fridge.

First taste - Saturday 1 February 2014

This morning () I cut a thin slice off the fillet in the fridge. The oak smoke flavour is good, but the fish is over-salty. The control with a long cure has to be the total quantity of cure applied. I'm surprised that Smoky Jo hasn't suggested a range of ratios. I'll have a close look at the book and maybe find something in the recipe section (later: I didn't). And next time the cure will have been broken down in the food processor to allow more control over the sprinkle.

I think I also need to adopt a trick recommended by many writers, which is - once the cure is finished and the fish is rinsed - to slice off a tiny sample and taste it. This should be quite safe if the curing has succeeded. If the salt level is too low, I could then sprinkle a little more cure mix onto the fish and give it anotjher couple of hours. If it is too high, I could give the fish a couple of hours in clean water and taste, adding more soak time for excess salt or a little more salting as above if too low.

I suppose there's no reason which I shouldn't soak and re-smoke a finished fillet if it really is too salty..

We had a generous portion each off one of the fillets at lunchtime today, on wheat crackers spread thickly with Philadelphia cream cheese. The saltiness was strong, but not enough to mask the smokiness or otherwise spoil the experience. The texture was excellent - quite dry, with none of the limp gelatinous quality you get in some commercial smoked salmon. I managed to cut fairly thin slices with my Dad's wonderful old ham knife, but next time I'll try Smoky Jo's trick of putting the fish in the freezer for 40 minutes to firm it for slicing.

Third smoking - Thursday 6 February

I bought snother nice big tail fillet of salmon yesterday from Darren and this morning I gave it a really good rinse and rub in fresh running water. It was then blotted thoroughly with kitchen roll.

While it was drying a little, I put the rest of the salmon cure in the small food-processor and gave it a reaL pasting, but the little balls of muscovado sugar only got smaller. They refused to disappear altogether. Mayb e demerara would have been better,,,

I sprinkled a thin layer of the cure on a lipped baking sheet, laid the salmon on this, skin-down, and then sprinkled the meat side sparingly with the cure. At least the session in the processor had improved the cure's sprinkling properties! I aimed for Jo's light sprinkling of snow, not trying to cover the colour of the fish.

After four hours quite a lot of liquid had been drawn out of the salmon but it still seemed quite flabby, so I set the timer for a further two hours.

At the end of the six hours there was still some salt left on the flesh side of the fillet. The fillet was quite stiff compared to how it was when raw. I decided to rinse the remaining cure off at this point and took a thin slice off the thick end to taste. The flesh was quite salty but nowhere near as salty as the first two batches.

Going for broke, I decided to put the fillet on a rack on a tray and air-dry it overnight in the fridge. This was done at 17:00, giving about 16 hours drying time before I can get the smoker on, which in turn will allow 12-14 hours smoking before bedtime tomorrow. I decided not to blot excess moisture off the fillet but to leave it to dry naturally in the hope that the much-disussed pellicle would form. This sticky coating is supposed to give the fish a gloss and to absorb more smoke.

Incidentally, I watched a YouTube video showing how to assemble and use the Eco Smoker and was told to put the bottom drip tray in upside-down, so I have done this for tomorrow's smoke. It certainly makes it easier to slide the lit smoke generatior in and out, and - on reflection - I can see that the pocket of air under the tray will provide some insulation between the burning dust and the cardboard bottom of the box.

Friday 7 February

The fillet was dry but tacky to the touch this morning, suggesting that it had formed a pellicle.

I put the fillet on a rack in the middle position of the Eco Smoker.

I filled the smoke generator with the coarse fraction of the sieved wood dust (see Back on task above) and lit it using a tea-light at 08:50. It was smoking gently at 09:00 but by 09:15 it had gone out. I re-lit it and it was going well at 09:35 and again at 09:50.

At 10:30 smoke was clearly visible emerging around the top of the Eco Smoker - more visible than on the second smoke with the 50/50 coarse/fine mix, which suggests that the coarse dust will burn away more quickly. That is logical as there will be more air between the grains. A check through the slot at the bottom of the box showed that the smoulder point had reached about halfway along the first leg of the spiral. A calculation based on this suggests that the burn will last 18 hours, but I find this highly improbably, especially as the dust seems to be burning faster than before!

At 12:35, with the second leg of the spiral just finished, the maths was looking a bit more sensible: 240mm in 200 minutes gives 865mm in 720 minutes - exactly 12 hours. So smoking should finish around 21:15 tonight. By 14:15 most of leg 3 had burned away, and doing the same arithmetic with my fairly rough estimate of just how much had gone I got a projected finish time of 22:30.

To be honest, I probably don't really need to check the generator this often, but if it did stop burning for any reason I'd like to get it going again as soon as possible.

Saturday 8 February

I checked again at 16:45 and 20:00, when all but the final leg of the spiral were burnt through. At 22:30, when I was quite keen to get myself and the dog off to bed, there was very little unburnt dust left, though there was a healthy red-hot cinder. Rather than leave this to burn out untended, I decided to empty the generator and take the fish into the house. After inhaling the aroma, which was much smokier than previously, I clingfilmed the fillet and left it in the fridge. This morning it feels quite firm and stiff, and I can't wait for a taste!

Sunday 9 February

I put the fillet in the freezer for 40 minutes and cut a couple of small slices just as a taster at lunchtime today. I thought the fish was less salty than the last batch but still quite strong, and that the smoky flavour was very mild even after 13 hours. Much more palatable than the previous fish, though.

I've read somewhere that oak is one of the mildest-tasting woods for smoking, although 'oak-smoked' seem to be the label on just about all smoked foods. It would be interesting to taste fish smoked with other woods. Patricia never stops talking nostalgically about the wonderful smell of apple logs burning on home fires in Normandy. Maybe when I prune our apple tree I can dry some of the wood and saw it with my circular saw. Or maybe I could just buy some apple dust. We'll be taking down a largish cherry tree grown from a stone soon - maybe that would be interesting, too...

Monday 10 February

I had a proper helping for lunch today. It actually seemed to cut more easily and controllably without a spell in the freezer. The colour is a stunning orange, there is no sign of free moisture or oil and the texture is fine and even, as if the fish has really been cooked by the cure. It looks as if any fat layer between the flakes has been cooked away leaving a lovely even texture.

The previous batch had seemed a bit gristly, but this was perfectly tender.

Most interestingly, it was far less salty than before, and less so - I thought - than the little taste I had yesterday. I wonder if that top slice had had the salt concentrated in it and over 24 hours the salt had spread itself more evenly through the flesh. Maybe a freshly smoked fillet needs to mature in the fridge for a couple of days to even out the flavour from the cure.

Eaten on a sliced sourdough roll, buttered and spread with Phildelphia soft cheese, the fish was quite delicious.

I intend to take a sample to the market for Darren to taste on Wednesday.

Wednesday 12 February

As it turned out, when I got to Darren's stall he told me he had been unable to get the curing salt he'd promised me from the Grimsby smoklehouse. He promised it next week. And I, in the rush to get out to the market, had forgotten to bring him a sample from my latest salmon! Better luck next time...

I bought a tail-end from a large fillet and cut enough to fry for our dinner from the little end. This left a much heftier piece of fish for the next smoking: 975 grams. At 13:30 I sprinkled a modest layer of the cure on a lipped baking sheet and put the fish skin-side down on this. Then I sprinkled cure thinly on the flesh side, rubbing it over with a fin ger to even out the layer. The tray was clingfilmed and put in the fridge with a timer set to six hours. At the end of this all the cure had disappeared, so I gave the fish another light sprinkle and two more hours in the fridge.

At the end of this time I rinsed the fish thoroughly and blotted it with kitchen roll. It was much darker in colour, quite stiff and now weighed 540 grame - 28 grams (1 ounce) less than before. I ran it under the cold tap to wet it and left it on a rack in the fridge to air-dry.

Thursday 13 February

At 09:40 I took the fillet from the fridge and weighed it. It had lost no weight at all. I filled the smoke generator with the coarse fraction of wood dust and it failed to light first time. I sprinkled a little of the fine fraction over the ignition area and it ignited OK second time round.

As the box, somewhat warped from its accidental shower a couple of weeks ago, had been leaking smoke all round the lid, I wrapped the top in wide clingfilm with a finger-sized hole poked in it over the hole in the box lid. There was still smoke issuing from the edges of the film, so this obviously wan't a brilliant idea - clingfilm doesn't cling very well to cardboard.

When I went back for a look at 10:15 the generator was burning well, but without much visible smoke escaping. Again at 11:00, burning was satisfactory but there still didn't seem to be much smoke at the top of the box.

Over coffee I was pondering, and began to wonder whether my wood dust was too dry, having been baked in the oven for several hours and kept in airtight containers ever since. So, at 11:50, I took a spray bottle of clean water out to the garage. The first thing that struck me was that there was quite a bit more smoke than before, but I went ahead and gave the generator a light misting anyway.

Maybe next time I'll leave a gap in the ignition area when filling with the coarse dust and mist this a little more before filling the ignition area with dry dust mix.

Nothing untoward to report until 23:00, when I decided to stop the burn and go to bed. I removed the smoke generator but left the fish in the box for tomorrow.

Friday 14 February

At 10:00 I half-filled the burner. At 10:15 there was a reasonable amount of smoke from the top of the box, but at 11:30 I had to stop the smoking to go shopping (I'm still not keen to leave the system burning when I'm asleep or out!). I left the generator on the garage step, but it raied quite heavily while we were shopping. I refilled it with Darren's Grimsby wood dust straight from the bag and re-lit it. At 15:15 I noted that it was still burning but not producting much smoke. At 15:45 I made the same observation - dust not dry enough or particle-size mix wrong? I microwaved a bowl of the dust on the simmer setting for next time. At 19:00 I stopped the smoking, The fish now weighed 520 grams - 48 grans down from the starting weight, which was around 8% weight-loss - nothing like the 20% recommended for preservation.

This piece iof fish looks beautiful. We'll probably taste it at Saturday lunch tomorrow...

Monday 17 February

In fact we didn't get round to tasting until lunchtime today. The fish is delicious: not over-salty and with a distinctly more smoky taste than before. The colour is rich and the flesh is tender. The first batch that is a real pleasure to eat. The proof is that Patricia ate quite a lot and declared it very good indeed.

There is a clear lesson in the curing process adopted for this batch: don't over-salt, but if all the salt disappears springkle some more and give it another couple of hours. And as far as smoking goes, it seems you can't oversmoke using the Eco Smoker.

Wednesday 19 February

I delivered the last 100 grams or so of last week's smoked salmon to Darren when I went to the market this morning. It was a bit of a wrench parting with the fish, which has been really delicious, but I want Darren's unbiased reaction - and anyway it was smoked with the wood dust he gave me and he has promised me some curing salt from the Grimsby smokehouse as well. He still hadn't managed to get the salt, but as I'm quite happy with my sea-salt and dark muscovado cure that wasn't a problem. I bought another big tail fillet - weight 778 grams - to use for frying and smoking.

On the way home I remembered that we wouldn't need salmon for frying this week, so I'll be smoking the whole piece. Oh dear!

So the whole piece was rinsed thoroughly and blotted dry. A thin layer of cure was then sprinkled on the lipped baking sheet and the salmon was laid on this skin-side down. The flesh side was sprinkled quite lightly with cure, which I rubbed over with my fingers before covering the fillet loosely with clingfilm and placing in the bottom of the fridge. At 12:25 the timer was set for six hours, the plan being to check for any residual salt after this time. There was none, so the fillet wase given a further light sprinkle and two more hours of curing. It was then rinsed and left wet on a rack in the bottom of the fridge.

I wouldn't be able to smoke the fish until Friday, so it would get a good long rest in the fridge after the cure.

Friday 21 February

The fish was left air-drying in the fridge until this morning, when I got up early to start the smoking. At 07:00 I filled the smoke generator with untreated Grimsby wood dust (no sieving or pre-drying) and put it in the oven set at 110C while I took George the dog for his daily morning walk. I also put the fish on its rack in the smoker. It was quite glossy - apparently the indication of a good pellicle.

At 08:20. having fed the dog, I lit the generator using a tea-light. Unfortunately the wick was rather short so the flam didn't reach the fuel. I re-lit is at 08:35 with a blowtorch and it was smoking well at 08:45. By 09:45 two thirds of the first leg of the spiral had gone and by 10:45 the second leg had started burning and by 11:15 was half burnt. At 15:10 the first four legs were gone and the smoke was still coming out at the top of the box. By 19:45 just the last three short legs were left unburnt and by 22:55 the generator was burnt out and stone cold, so the smoking probably ended around 22:00 - a total smoke of over 13 hours.

The fish was left in the smoker as the night was pretty chilly.

Saturday 22 February

The plan had been to re-start smoking early this morning, but we decided to go to Retford and I'm still not quite confident enough to leave the burning generator unattended. So I re-charged the generator with untreated Grimsby wood dust and put it in the oven. By the time we were ready to leave the oven r=had reached about 190C, so I turned it off to cool naturally. At 13:10 I re-lit the generator with the blowtorch, as the tea-lights I have been using are just too puny. The fire took first time and by 14:25 the first leg of the spiral had almost burnt away. Smoking was still going well at 15:25, two and a quarter hours after lighting. If I leave it to run until bedtime - around 23:00 - the fish will have had a total of over 23 hours.

Which is just what happened. At 22:50 I took the smoke generator out into the garden and emptied the still-smouldering remains of the dust out. Then I transferred the fish, which was very firm and a lovely dark orange colour, on its rack into the garage fridge for the night, leaving it unwrapped so that it could continue drying.

Sunday 23 February

This morning I brought the fillet into the house. It was dry to the touch and very still - I could hold it by the tail end and it hardly drooped at all. I weighed it and found to my amazement that it had gone from 778 grams to 520 - a reduction of 33%! Smokin' Jo says the weight should reduce by at least 20% for the fish to be properly preserved, so I guess this piece will be safe to eat.

This piece of fish has spent a lot longer uncovered in the fridge than the previous one, as well as having longer in the smoker, so the air-drying must have worked really well.

Patricia and I had some for lunch, and it was firm and dry, yet moist enough, with a nicely balanced flavour and certainly not over-smoked after almost 24 hours in the box. The last lot was very good, but this was definitely better. The depressing thing is that you carve the thinnest slices you can manage and eat what seems like a very modest portion, yet the fillet seems to shrink very rapidly!

A lifetime's supply of wood and salt!

Wednesday 26 February 2014

This morning my visit to Worksop market was more than usually fruitful. As well as a fine South Coast Dover sole (identified by its pale skin) for tonight's dinner and another salmon tail fillet for smoking, I got Darren's positive feedback on his sample from the week-before-last's salmon.

And I got a very large present - a bag of salt as used by the Grimsby smokehouse next door to Darren's premises. I say very large because it turned out to be a 25-kilogram bag. Darren never does anything by halves! He very kindly volunteered to put it in the car for me - provided I could get the car into the little blind alley behind his stall. Aided by Google Maps on the new phone and advice from a lady who way buying fish, Patricia and I managed the complicated feat of navigation and Darren carried the bag and put it in the boot.

We headed for home and unloaded the shopping. Duly fortified by two cups of strong Costa coffee (the roast and ground, sold in 200-gram bags at better supermarkets - very good indeed!), I lifted the bag out of the car boot and staggered into the hall, where I laid it as gently as possible on the hall table. A second leg got it to the kitchen table, where I managed to decipher the white text on the pale blue bag: 'Pure Dried Salt - The Salt Company, Nantwich, Cheshire'. There was also mention of an anti-caking agent called - rather ominously, I thought - sodium hexacyanoferrate. The 'cyano' bit always sounds a little dodgy, potassium cyanide being one of the most lethal of poisons, but it actually crops up in all sorts of perfectly innocent chemical compounds.

So to Google and a search for 'salt company nantwich', which threw up plenty of hits, the top being and yielding this: 'We can offer Uk mined and packed Salt from 750 gram tubs to 12.5kg bags. We can offer Israeli mined and packed Salt from 10kg to 25kg bags'. The fifth link was to was The Salt Company - FJ Need (foods). A quick wander round the site found - among other products - Pure Dried Salt and the following information: 'Pure dried salt is a refined natural sea salt. The salt is produced by solar evaporation of sea water in ponds located by the Red Sea coast. The salt manufacturing is done without chemicals thus obtaining the white crystal glaze. It is a high quality food grade salt. Production conforms to IS0 9001.2000.' Some of the company's salts are approved for use in organic food production, so I seem to have stumbled on a plentiful stock of really good salt for my curing experiments - not to mention all domestic use for years to come! Many thanks, Darren!

The use of the word 'mined' is a bit odd. You don't mine sea salt. However, Cheshire has been an important source of rock salt, which is mined, since Roman times.

We called in at the village Co-op on the way home from the market and I found a pack of bagels from the New York Bakery Co (made under licence in nearby Rotherham!). Straight from the packet they weren't brilliant, but lightly toasted and spread with butter, Philly cheese and last week's salmon? Pretty superb! The crumb became softer and the crispness was perfect.

Thursday 27 February 2014

The tail fillet I bought from Darren yesterday was a lot smaller than the previous one - just 398 grams. Probably lucky, because I wasn't my usual obsessively organised self and I made a mess of the curing!

At 09:25 I applied the cure and filmed and fridged the fish as before, setting the timer for six hours. I was busy when it went off, and although I intended to go back to it as soon as I had finished what I was doing, I forgot and didn't remember it until 22:10 - a 12-hour cure! All the salt and sugar had gone but I decided not to give the fish the usual 2 hours with a little more cure - not least because I didn't want to stay up for another two hours! I rinsed the fish and put it, wet and uncovered, on a rack in the garage fridge before retiring. The flesh was not as firm as last time but noticeably more so than when raw, and the colour had darkened from 'salmon pink' to a nice dark orange. The weight had fallen to 384 grams.

Saturday 1 March

It wasn't convenient to smoke on Friday, so I left the fish in the fridge until this morning. I pre-heated the smoke generator in a hot oven while I walked the dog and lit it at 10:00 with the blowtorch - tea-lights having provied rather ineffective. As usual there were a few false starts, but smoking was well under way within 20 minutes. The first day of Spring was conveniently frosty so there would not be any problems with temperatures high enough to promote spoilage. At 22:10 the last short leg of the spiral had been reached, so I left the smoker to burn out. Again it was a frosty night so I wasn't worried about leaving the fish in the smoker overnight.

Sunday 2 March

After the ignition hiccups yesterday I thought I'd be clever and pile the dust quite high over the ignition ramp. I lit this very easily with the blowtorch but the adjacent leg of the spiral caught fire too! I blew this out and sprayed it with water, but when I checked at 09:00 both legs of the spiral were going strong - the inner one smouldering at both ends so I was actually getting three times the normal quantity of smoke!

The generator burning in
                    three places at once!

I scraped the L-shaped area clean, making sure I got all the red-hod cinders out, and continued the smoking. By 13:00 the smoulder point had almost reached the break, so I filled this with dust. By 15:00 the 'repaired' section had burned away, and smoking continued normally until shortly after 19:00, giving a time of 10 hours 50 minutes. Added to yesterday's time, this made a total smoking time of 22 hours 50 minutes. The fish on its rack was transferred to the garage fridge.

This 'error' has given me an idea: if I want to give my fish a more intense smoking I could light the dust at both ends and wait for the smoulder points to meet in the middle. Or, as with the mistake, light two separate bits to get three or four smoulder points...

Monday 3 March

At 09:00 I brought the fish into the house, wrapped it in film and weighed it: 350 grams. It was quite stiff (though not as stiff as the previous fish) and a good dark orange colour, but the weight loss was only 12% - nowhere near the 20% needed for long-term preservation. This must have been due to the mess I made of the curing stage.

I haven't tried this fish yet, as we're are still eating the previous piece, which we both agree is excellent.

Wednesday 5 March 2014

Market day. I had an entertaining conversation with Darren while buying some salmon to smoke, some more to eat fried and a splendid Dover sole for tonight. He was interested in what I had discovered about the salt he got for me.

We tried the latest fish at lunchtime today. It was firm, fairly dry and very tasty, so my error with the curing doesn't seem to have had any ill effects. No need to bin it, then!

After lunch I split the smaller of the two salmon fillets I'd bought down the middle and skinned them, ready for frying. Then I washed the larger one, which weighed in at 468 grams, dried it and sat it on a thin bed of cure mixture on the usual lipped baking sheet. I sprinkled and rubbed it with cure, covered it with clingfilm and put it in the fridge for the first six-hour cure.

There was no obvious residue of salt on the fish when the timer went off, so I drained to liquid off and applied a light sprinkle of cure, leaving it filmed in the fridge for another two hours, as I had done with previous batches. At the end of this time, I rinside it thoroughly and, without drying it at all, put it on a rack over the tray in the outside fridge, which was very cold.

The fish will get a long air-drying period in the garage fridge because I won't be able to smoke it tomorrow. I'll have to start on Friday morning, so 36 hours to 'lose weight'...

Friday 7 March

At 07:30 I washed the generator - which is going to need a more thorough clean after this batch - filled it with Grimsby wood dust and put it in the oven, set to maximum (230C). It was on a lipped baking sheet with some spare dust in case any topping-up is needed. The oven was switched off when I went out to walk the dog, and at 09:00 the fish was put on the middle shelf of the smoker and the burner was lit with the blowtorch.

I then prepared a large cold-block, fresh from the freezer, wrapping it tightly in foil - in part to protect it against tarry fumes but mainly to slow heat transfer.

I've been reading a lot of very contradictory information about temperatures for cold smoking on the web. One argument is that the food being smoked needs to be kept at fridge temperature during smoking to avoid spoilage. Another says that cold-smoking should be done at around 80-90F (a US source, of course!). Who to believe? In the absence of any concensus, I decided to stick to really cold smoking, which is what I've been doing so far, thanks to the weather (we've had more frosts since the start of Spring than we've had through the entire winter!).

So I put the wrapped cold pack on the bottom shelf of the smoker.

At 09:15 the burner was going well, but when I checked at 11:00 it had gone out after burning most of the first leg of the spiral - weird. I re-lit it and at 12:00 there was plenty of smoke in the box.The outside temperature was well into double figures - it had been about 10C when I went out at 07:30 - and the sun was shining through the garage windows directly onto the smoker, so I put a large empty box in front for shade. Nevertheless, at 13:00 the surface of the fish was at 11C.

I'm hoping that full curing and 36 hours' air-drying will be enough to preserve the fish properly, with smoking mainly to add flavour. I don't think I've ever got into anything where the information on the web is so garbled and contradictory - it's really fristrating. Even Harold McGee's On Food and Cooking- my bible for food-related science - is strangely brief on salt-curing and smoking. I'll just have to busk it! I think my system can handle Listeria and Salmonella, but I must admit to pangs of anxiety when I read mentions of botulism...

The burn progressed without problems. By 17:00 more than half the spiral had burnt through, and at bedtime there was only the last very short leg left. I decided to abort today's smoking at this point, putting the fish, uncovered, in the outside fridge and clearing the generator out.

Saturday 8 March

I turned the oven on to pre-heat to 230C as soon as I got up, washed the generator and charged it with dust. The oven had reached maximum temperature when I was ready to walk the dog so I turned it off. When I got back there was a (to me, anyway!) pleasant scent of hot oak in the kitchen. At 08:50 I took the generator out to the garage and lit it with the blowtorch before putting the fish in. I decided the cold pack was unlikely to help and left it out. At 08:55 the burn looked rather puny so I gave the dust another blast with the blowtorch. Ten minutes later smoke was visible above the box. When I checked at 09:25 and then at 09:55 everything looked good, with the first leg of the spiral burnt through at 10:30. I checked every hour or so through the day and found no cause for concern, with ⅓ of the fifth leg burnt by 16:00. At 20:50 the last leg had started burning, and when I went back to check at 22:00 the generator was burnt out and stone cold. I moved the fish onto an open rack in the garage fridge and left it overnight.

Sunday 9 March

This morning I weighed the fish: 394 grams, reduced from 468 grams. That is a slightly disappointing weight-loss of 15.8%, which suggests that the fillet was under-cured. The timing was okay, so  the only other variable - the amount of cure - must be the problem. The trouble is that the professional information is pretty limited - 'like a light sprinkling of snow' doesn't exactly make for repeatability!

Perhaps the answer is to weight the container of cure, apply a slightly heavier sprinkle and re-weigh to establish how much has been used. This could be used to produce a repeatable (or modifiable) ratio with the raw weight of the fish...

Anyway, I halved the fillet lengthways, following the natural centre-line, and filmed the two halves. These are now sitting in the kitchen fridge. The sensible plan might be to vac-pack, label and freeze one piece and eat the other.

Monday 10 March

And that is what I did. We ate some of the salmon for lunch, and I had slight doubts about the flavour - something metallic, I thought. We'll see...

We ate the last of the previous batch at teatime, chopped up and stirred into some of Patricia's amazing, creamy scrambled eggs.

Monday 24 March 20

Just finished the last of the very first batch, which was cured for only two hours with a thick covering of salt - the process I have always used for hot-smoked salmon. The fish was flabby compared with all subsequent batches, and lacked any distinct smokiness. However, it was edible and didn't go to waste!

I've just checked: there is a half-fillet dated 10 March in the freezer - quite a small one so I will have to get some more fish on Wednesday for smoking at the end of the week.

Hit-and-miss curing

It really would help to have some better data about the ideal ratio of cure to fish. It seems to me that it should be possible to calculate the optimum quantity of salt for a given weight of fish, yet I can't find anything more precise that the 'light covering of snow' formula! This is really frustrating for someone who likes to be systematic. After all, the fish isn't cheap, and the investment in time and effort really shouldn't be gambled.

Maybe I should start by doing the covering-of-snow bit from a known weight of cure so I can calculate how much of the salt/sugar mix is actually used...

Wednesday 26 March

I bought a nice big (760 grams) salmon tail fillet from Darren this morning. After washing it thoroughly, I split it in two along the centre-line.

I had very little of the muscovado cure left, so I added 200 grams of fine sea salt (Sainsbury's - not Darren's yet) and 100 grams of white caster sugar, tumbling the jar to mix the ingredients thoroughly (I hope - two white ingredients give no visual clues. Maybe that's why people use brown sugar!).

I sprinkled the skin sides with 20 grams of the cure and rubbed it over as evenly as possible before turning the fillets skin-side down. This, I thought, was a lot more precise (and economical) than sprinkling a roughly-salmon-shaped pattern of cure on the sheet and putting the salmon on top. I then sprinkled 40 grams of cure on the flesh sides and again rubbed it in to even out the coating.

This gave a ratio of one part of cure to 15 parts of fish. This is me trying to do something that is actually repeatable!

After six hours in the fridge under clingfilm all the granular cure had disappeared, so I sprinkled and rubbed a further 20 grams on and left this for another 2 hours. By the end of this some of the mix was still visible on the fish. I'll do the rest of the maths later!

At 21:30 the fillets, which were fairly stiff, had been rinsed and left wet, then placed on a rack over a baking sheet. This was loosely covered with clingfilm and put in the outside fridge for the night.

Thursday 27 March

As usual, as soon as I got downstairs, I filled the smoke generator, using straight-from-the-bag Grimsby wood dust, and put it in the oven, leaving it on until the thermostat light went out and then turning it off. After walking and feeding th dog, at 08:15, I lit the generator and put the two half-fillets on the middle rack of the smoker. At 09:30 the smoulder had died, so I re-lit the dust. Then I remembered that we had to go out for the second half of the morning, so at 10:15 I raked the smoulder away from the rest of the dust and prodded it to kill the fire.

Smoking proper started at 14:15, and at 14:40 and then 15:40 combustion and smoke output were steady. I killed the fire at bedtime and will do a whole-day smoke tomorrow.

Friday 28 March

The smoke generator was caked with black residue after yesterday's rather disorganised smoking. I decided that this might be impairing the flow of smoke to the smoulder point, so the first job at 7 o'clock this morning was to put it to soak in very hot water with plenty of washing-up liquid. Then I pre-heated the oven to 230C and put in a tray of wood dust.  Finally, I scrubbed the generator thoroughly with a stiff brush, rinsed it and popped it in the oven to dry. By the time the dog and I were ready for our daily two-mile walk, the thermostat light was off, so I turned the oven off.

When I got back, after rubbing the dog down and feeding him his breakfast (some things take precendence even over my culinary experiments!) I charged the generator with the warm and slightly-browned wood dust and fired up the smoker using a tea-light underneath and the blowtorch above. For once it caught first time and by 10:00 the first leg of the spiral was burnt.

Note: I need a good brush for cleaning the gauze of the generator. I used to have a brass-wire suede brush once - wonder if it's in the shoe-polish box...

10:30 That brush dates back to when I used to wear brushed-pigskin shoes called Hush Puppies. According to Wikipedia, these were launched in the USA in 1958. They were pretty trendy here in the early sixties. I probably got my first pair when I was aged about 20, around 1963. I only mention this because I've just found the abovementioned brush nestling in the polish box, so tomorrow I'll be cleaning my smoke generator with a brush I've had for half a century. The brass wire should be fairly gentle on the chromium-plated gauze.

By 15:30 the first four legs were burnt. All day the production of smoke seems to have been pretty even, with wisps constantly visible above the carton lid. I estimate the that dust will last until late in the evening, and I will extinguish it if it has not burnt out by bedtime, which will mean a 14-hour run today to add to the 8 yesterday - 22 in total. The fish will be left uncovered in the garage fridge until tomorrow - probably mid-afternoon before I vac-pack one half for the freezer and clingfilm the other for immediate consumption.

Monday 31 March

The smoking got quite disrupted because the generator went out a couple of times. In the end I decided that enough was enough and transferred the fish to the garage fridge, where it rested, uncovered, until Sunday morning. The weight was 640 grams, a 16% drop from the original 765.

As planned, I vac-packed one piece for the freezer and put the other, clingfilmed, in the fridge. I had some for lunch today and found it very good, though still lacking in a really definite smoky flavour. The oak dust from Grimsby is obviously rather mild, so I may try buying some small quantities of other woods, in spite of the 'artisan' prices! However, the texture was excellent - fairly dry and firm but still succulent - and the saltiness was fine.

The total amount of cure used was 80 grams: 20 on the skin side and 40 on the flesh side for the 6-hour period and another 20 on the flesh side for the extra two hours. This is a ratio of 9.5 parts of fish to one part of cure. In my next batch, I'll try using 60 grams on the flesh side for the six hours, extending to eight if the cure is not fully dissolved.

Wednesday 2 April 2014

I bought a 575-gram tail fillet from Darren this morning. This weight was simply Darren's best guess at cutting another 700-gram piece - he'll never be a millionaire!

I rinsed the fillet thoroughly in cold water and blotted it off on both sides with a single thickness of kitchen roll. This didn't dry it fully - I wanted enough surface moisture to make the cure stick.

The calculation was 575/765x20 for the skin side and 575/765x60 for the flesh side, which gave me nice round figures of 15 and 45 grams for today's fillet.

I put the fillet skin-side up on a board and sprinkled the 15 grams of cure over this as evenly as possible, managing to lose hardly any off the edges. A light rub with one finger evened out the coating nicely. I then lifted the fillet by the tail end, turned it over onto a lipped baking sheet and tipped the small amount of cure left on the board onto the fish. There was almost none on the sheet.

Then I sprinkled the 75 grams of cure as evenly as possible onto the flesh side. The colour of the salmon made it easy to judge the evenness of the coating. I tried to get as near the edges as I could without any overspill and to make the coating thicker at the thick end of the fillet. Again, a rub with a finger smoothed the coating nicely.

A sheet of clingfilm laid lightly over the tray finished the assembly, which went on the bottom shelf of the fridge with the timer set to six hours.

At the end of this time cure was still visible on the surface of the fish, so I left it in the fridge for another two hours, after which it didn't look a lot different. Is there some limit to the amount of cure a given piece of fish can absorb?

I rinsed and weighed the fish: it had gone from 575 grams to 528 - a disappointing loss of only 8%. I will weigh again after smoking.

At 19:30 the fish was placed, uncovered, in the bottom of the fridge overnight

This was a far more controlled - and therefore repeatable - process than I had managed with any of the previous batches.

Thursday 3 April

At 07:00 I put a tray of Grimsby wood dust in the oven set at 190C, turning the oven off when the therostat light went out. At 08:50, after walking and feeding the dog, I charged the smoke generator with the warm dust and lit it at both ends in an attempt to double the smoke output. The normal ignition point started smouldering more easily than the end-point, which needed a couple of re-lights, and at 09:50 both ends were smouldering well.

When I checked at 13:00 both ends were burning well, with substantially more smoke visible around the edges of the carton lid, but when I went back at 15:30 the dust had all burnt out. I refilled the generator and lit it again. At 16:30 both ends were smoking well.

Smoking was still going well at 21:00 but when I checked at 22.30 the dust had burnt away and the generator was cold. The fish was quite still and a beautiful dark orange colour. I put it, uncovered, on its rack in the garage fridge for the night.

Friday 4 April

This morning, from  528 grams, the fish had shrunk to 492 grams - a total weight-loss from fresh of 83 grams or 14.5%. I trimmed about two centimetres from the pointed end and then cut the fillet in two down the centre-line. The skin was very tough to cut - far more so than previously.

I cut each of the offcuts into three little slices, skin and all, mainly to add to the dog's remains of the fried fresh salmon from last night's dinner. However I cut the meat off one piece to taste: it was very salty and not noticeably smokier than previous batches, but this was from the thinnest part of the fillet.

I vacuum-packed the larger piece for the freezer and clingfilmed the other for immediate consumption. I still have a very small amount of the previous batch to finish, so it will be interesting to compare the flavours.

I now have half-fillets from three different batches in the freezer.

Of course I have fallen into the oldest trap in science by changing two variables in a single batch: the quantity of cure and the smoking process...

More later...