A (Epic?) journey in search of my perfect roaster

[Javaphile]

Re: A (Epic?) journey in search of my perfect roas

My pleasure JM. It has been a long and interesting journey so far and while I can now roast with it there is still further to go down this trail.

Despite all the interuptions and an issue with the roaster I did manage to get my big Christmas batch of 14kg or so done in time for the family gathering. It was a close thing though as I was roasting/packaging right up until I ran out the door and still arrived late.

Barrel number 3, the one with the broken bolt, gave me some problems during the big roast. Halfway through it suddenly started skipping gears. After turning everything off and letting it cool down enough to work on I managed to (*again*) tighten the worm gears set screw, once again tightening it to the point to where I was afraid something was going to break if I went any harder. I also put a nut on the bolt off the back of the barrel that holds it to the frame in back and tightened it down so that there was no more float in the gears. This allowed me to finish the session with no more skipping gears but will require looking into to find out what the problem is there.

With the big roasting session behind me Ive come to strongly consider if perhaps barrels 1 and 3 are set up for natural gas while barrel number 2 is set up for propane. The flame controls on both 1 and 3 are extremely touchy and go from not enough gas to keep the burner lit to a large enough flame to turn your beans to cinders in 2 or 3 minutes by moving the end of the lever perhaps 5mm. All this occuring at the 2.5 area on the scale of 0-6 by the levers. I cant believe they would have made this to have so touchy a flame adjustment over such a small area.

Ive not yet used the middle burner but if youll recall that was the one that was different from the other two and on the test burn produced smaller flames. Now that the pressure of having the big Christmas roast done is gone I plan on switching barrel 1 over to the number 2 slot and removing number 3. For my own personal consumption needs I can easily get by with just a single barrel running and if need be I can always toss barrel #3 back on for a session.

If as I suspect the middle barrel is set up for propane while the outter two are set up for natural gas I have a real delima on my hands. Ive already been told by 2 local shops that nobody will touch the roaster and that parts are not available for that type of a system anymore. I still have another couple of resources to check out for possibly getting the parts from before I give up on using it as-is.

Ideally I want to convert it over to a modern system that I can interface with my PC for control. That however takes some pretty major bucks (the one conversion I know of took $3,500) that Im afraid the piggy bank doesnt currently have and will take time to accumulate. Until then I could either run it on 1 barrel in its current location and use the others only in dire need and live with the fiddling fine adjustments on them. Or, I can move the roaster to where it can be hooked into the natural gas supply (there is no really convient location for this currently) and have 2 barrels that work and a third that very likely wont work at all or at best with only small quantities in it.

Option 2 is really not very appealing to me as it would require a pretty major rearrangement of my basement. Not the easiest thing to do when dealing with a Letterpress printshop and its tens of thousands of pounds of metal.

Fortunately I have time to consider what I want to do as all of it is a mute point until I have enough good barrels to actually run all three at the same time and before that is going to happen I need to source new gears for them.

Things that make you go hhhhmmmmmmm......

Java "The Journey continues" phile

[robusto]

Re: A (Epic?) journey in search of my perfect roas

Mmmmm, pity about option 2, because converting from propane to natural gas is easier than vice versa. The jets simply need to be drilled out to a larger size.

[Javaphile]

Re: A (Epic?) journey in search of my perfect roas

Mmmmm, pity about option 2, because converting from propane to natural gas is easier than vice versa. The jets simply need to be drilled out to a larger size.

That and some of the air injection vents plugged up as well it would appear. The one barrel that appears to be set up for propane has over twice as much air being injected into the gas flow as the other two barrels do.

Im hoping to purchase the complete gas supply part of the roaster from the gentleman in California who has converted his Gothot over to a modern temperature controlled system. Hopefully hell be willing to part with it and it will be set up for propane. If its set up for natural gas I guess Ill just have to live with that until such a time as I can afford to convert it to a modern system.

Not only would the propane be easier to set-up here but I have hope of taking my roaster on the road for roasting on location. Kind of hard to do if it runs natural gas. :-/

Getting proper roast temperatures on barrels 1 and 3 (the ones that appear to be set up for natural gas) was a real pain in the you know what. It was obviously doable but not something Id want to be fighting with for every roasting session.

Another option Im considering is to put an adjustable regulator on the gas supply and hope that by turning down the pressure I can achieve better flame control on the natural gas barrels. Something to consider if all other options fall apart.

One barrel will suffice for my own personal roasts, but the production with only one barrel running just isnt enough when I need to do up 10 or 20 pounds, which it appears I will have a call to do every 2 weeks. Plus a 5 pound batch Ill need every 3 weeks for my batch of Cold Process coffee.

It will be interesting to see where this all ends up. :

Java "And The Journey continues" phile

[Javaphile]

Re: A (Epic?) journey in search of my perfect roas

Well I had an interesting experience here last night while doing up 6 pounds of Nicaraguan Angelina Estate beans for a batch of CP.

I decided to play with the frequency setting on the microdrive controlling the motor as the drums were turning much faster than 40-50rpms with it set at 50Hz. At 50Hz the drums turn at about 100rpms. I started dropping the frequency and had to set it well below 30Hz before the rpms dropped to about 50. At that speed there appeared to be very little air flow through the system even with the flues opened all the way. A little experimenting resulted in my setting the microdrive at 40Hz which gave me a barrel rpm of about 60 and adequate air flow to clear the smoke and chaff from the barrels.

Dropping the freq from 50 too 40 also resulted in a much quieter operation. The roaster seems to prefere the slower speed with most of the noises and vibrations dissapearing. I have to wonder if the guy in California misspoke and ment he ran his at 40Hz rather than 40rpm.

Id just settled on 40Hz and was about halfway through the session when barrel #3 suddenly started spewing smoke from its fill hole. I opened the flue to increase the air flow through the drum but the smoke kept increasing in volume (the beans had just passed first crack). Wondering what was going on I grabbed the flashlight and took a look inside the barrel to find that the shaft at the rear of the drum where the air is drawn out was completely blocked with chaff!

Fortunately all the tools Id used to assemble the roaster were still sitting handy so I grabbed the longest screwdriver and tried to poke the chaff out from the front of the barrel. No go. The screwdriver wasnt long enough to reach the blockage through the fill hole.

By now the smell of parafin is starting to permeate the room along with copious amounts of smoke as the beans approach 2nd crack. I quickly kill the burner on #3 and dump the beans and flip the barrels door up all the way and try to reach the blockage again with the screwdriver, again with no luck. Even using a pair of pliars to hold the screwdriver its still not long enough to reach the plug of chaff. By now Im having visions of the chaff bursting into flames as the place starts smelling like a candle factory rather than a roastery.

Right in the middle of this (of course!) barrel #1 hits first crack and requires attention. While adjusting the flame height on #1 Im frantically looking around the room trying to find something longer that will allow me to reach the plug of chaff in #3 and clear it to no avail. The screwdriver is the longest thing I have handy, and its not long enough!

Casting about for a solution I realize that I can reach the plug from the rear of the barrel through the air channel by opening the flap on it. I carefully climb over the gas line and squeeze into the space in back of the roaster and start poking at the blockage with the screwdriver but as fast as I can loosen it it plugs right back up again due to the air draw! At this point the chaff is obviously on the verge of ignition and I dont have time to climb out from behind the roaster to try and find a larger tool (the long screwdriver Im using is a small slotted tip one) that will push the chaff out of the shaft and into the barrel against the air flow. As a last resort I give the plug a good stir with the screwdriver and then blow as hard as I can into the back of the shaft.

An explosion of chaff out of the front of the barrel signals success! Whew! Just in time! No fire today!!

At this point barrel #1 is in need of dumping as I can hear the first of the 2nd cracks starting. I clamber out from behind the roaster and dump #1 into the cooler and load another batch into it. Now Ive got a few minutes before #1 will need more attention and climb back behind the roaster and use the screwdriver to clear all the chaff out of the shaft on #3 where it clings to it.

Whew! That was a close call! A bit too close for comfort in fact! Some of the chaff ignited as it was blown out of the shaft!

The culprit appeared to be a bean that had worked its way into the shaft and while I was playing with the motors speed the low air flow allowed some chaff to get hung up on the bean which then quickly caught additional chaff and quickly plugged the shaft completely.

With a high enough air flow this shouldnt be a problem, but just in case I decided to make another of my high-tech tools. In this case a wire coat hanger bent to fit into the shaft that will quickly clear any blockage, from the *front of the barrel/roaster! ;D

The rest of the session was uneventful and I quickly finished up the 6 pounds of beans. Now I just have to wait for the beans to finish soaking and then I can taste my first batch of CP made from beans I roasted rather than ones from the speciality roaster.

Java "Another day, another lesson" phile

[robusto]

Re: A (Epic?) journey in search of my perfect roas

A near-tragic story with a happy ending! As well as the coat hanger, Id keep a fire extinguisher or blanket on stand-by too!

Glad to see youre ironic out the bugs.

Robusto

[Javaphile]

Re: A (Epic?) journey in search of my perfect roas

A near-tragic story with a happy ending! As well as the coat hanger, Id keep a fire extinguisher or blanket on stand-by too!

Glad to see youre ironic out the bugs.

Robusto

I think in one of the pics I posted earlier you can see not one but two fire putter-outters sitting near the roaster. A CO2 for if its small and a big honking dry chemical one for if its not.

Java "Ironic on!" phile

[Dimal]

Re: A (Epic?) journey in search of my perfect roas

Hi Java,

One could never claim that life was dull around your place

Mal.

[Javaphile]

Re: A (Epic?) journey in search of my perfect roas

Hi Java,

One could never claim that life was dull around your place

Mal.

Indeed Mal, indeed!! But then who wants to live a dull life? Not me! ;D

Java "Live to the fullest!" phile

[Javaphile]

Re: A (Epic?) journey in search of my perfect roas

Perusing through the forums I see I havent done an update on the Gothot for some time and as I just finished up a session this seems as good a time as any.

Im still running on only 2 of the 3 barrels as the Department of Finances has put the kabosh on paying a machine shop multiple hundreds of dollars for a single new gear, likewise any mods to the burner side of things.

The entire roaster is still barren of all instrumentation so all roasts are done the old fashioned way using wetware to judge the progression of the roast.

Ive settled on a batch size of 200g. 250g was too much with quite a few beans being sucked out of the barrel with the draft above 50%. At a 200g batch size over the course of a 3kg roasting session less than 10 beans end up in the chaff collector. At a 250g batch size 5% of the beans were ending up in the chaff collector! I had one batch of a Nicaraguan coffee that had really large chaff that required a really high airflow where 12% of the beans ended up in the chaff collector. Ugh!

Too date roughly 100kg of beans have been put through the roaster and its still chugging along happy as can be as long as the barrel shafts are kept well oiled.

The coating is holding up perfectly except on the left sides of both operating barrels where it has bubbled up and is flaking off. Only on the left sides, not at all on the right sides. Strange, very strange as the flame is symetrical on both sides of the barrel. It appears that the rotation of the barrel causes more heat to be drawn up the one side than the other. This is an issue to be explored later as it appears to be solely an appearance issue with no impact on either performance nor protection. Under the coating that is flaking there still appears to be a layer of it covering the metal so rust should not be a problem.

So over-all while the Gothot hasnt found its permanent home yet and is still in a temporary set-up and location running 2 of its 3 barrels Im very happy with it and still having fun playing with it. ;D

And yes, the room its in smells *wonderful! 8-)

I especially like going up there a few hours after a big session doing multiple varietals. The combined smell of 5 different beans in kilo sized lots is pure heaven! In fact, I think thats a perfect note to end this on. Time to run off and smell the (5 varietals, 6kg total mass) beans!

Java "Off to get a sniff of heaven" phile

[robusto]

Re: A (Epic?) journey in search of my perfect roas

Sounds like a very well-oiled machine, Javaphile. Too bad about that third barrel, but 2 out of three aint bad for a domestic set up, really.

Robusto

[Javaphile]

Re: A (Epic?) journey in search of my perfect roas

Sounds like a very well-oiled machine, Javaphile. Too bad about that third barrel, but 2 out of three aint bad for a domestic set up, really.

....tending towards light industrial even I think with the quantities Ive been pumping through it, an average of 25kg per month in the 4 months its been running.

It can get pretty hectic at times with just the two barrels going during long runs. With all 3 running I think anything not nailed down would get blown away by the whirlwind of keeping up with them!

At my current roasting quantities 2 barrels works out OK. If these amounts increase then it would be nice to have the third barrel running so-as to cut down on the total roasting time, but for now Im quite happy to have 2 up and running. It could easily have been just a single barrel what with the problems of the badly worn gear and broken bolt/easy-out on one of the two running barrels. Hopefully that barrel will continue to operate properly until such time as the Dept of Finance gives its stamp of approval for new gears.

As an aside and point of interest for others looking at oils for roasters, I ended up using common 3-in-1 household oil for the shafts. I had several high temp oils I had originally planned on using but in the end I went with the 3-in-1 due to the nasty smells of the high temp oils.

While the high temp oils required the usage of less oil for good lubrication, they produced some really nasty smells that I just couldnt stand and the 3-in-1 proved to be able to withstand the temps involved and didnt reek. The 3-in-1 is a thinner oil and is used up faster than the high temp oils requiring more frequent applications. With each application being only a few drops, the larger amounts used and the much lower price of the 3-in-1, cost is not a factor.

With the nasty smells the high temp oils produced I flushed the bearing surfaces with lots and lots of the 3-in-1 to get rid of all of the high temp oils. They smelled so bad and so strong I have little doubt that the smell would have been passed onto the roasting beans.

The 3-in-1 has worked well so far and I see no reason why it wont continue too. At normal roasting temps it produces almost no smell and even while sitting between roasts you have to get with-in a few inches of the area before you can detect the oil. The high temp oils made the whole 2nd floor of the house reek. Yuck!! :P

Java "Just an ordinary common oil kind of guy" phile

[Bullitt]

Re: A (Epic?) journey in search of my perfect roas

OK Andy, how did you manage to do it, I might have missed something unless there was a PM not specifically mentioned.
But ringing someone accross the globe and in a Coffee House (please dont condemn me if thats an in-appropriate desciption, have never seen or been to a Starbucks, AFAIK there aint any in Tasmania)

OH and when is the rest coming Java.....great story....phile???

[Javaphile]

Re: A (Epic?) journey in search of my perfect roas

OK Andy, how did you manage to do it, I might have missed something unless there was a PM not specifically mentioned.
But ringing someone accross the globe and in a Coffee House (please dont condemn me if thats an in-appropriate desciption, have never seen or been to a Starbucks, AFAIK there aint any in Tasmania)

Hey Bullitt,

While Starfu*ks may technically qualify as a Coffee House I wouldnt call what they serve coffee. :P But then Im a Coffee Snob so what can I say. ;D

As to how Andy was able to call me, I had just made a post updating the story of the journey (this may have been in the moderators forum and hence not publicly viewable) so he knew I was in Larado in a Starfu*cks. He googled it and discovered that there were only 2 in the whole town, so he chose one at random and called them and asked if there was a guy with a laptop there. 8-) There wasnt anybody there at all. LOL So he called the other one and Bingo! There I was! LOL ;D

When they told me I had a call I assumed it was the garage calling me about the van as who else would be calling me there?!? When I answered the phone I was all set for the strong Mexican accent the owner of the garage had and so for the first few seconds my mind just heard garbage. Hahaha It took a few seconds for the old grey matter to realize what was going on and switch modes from Translate.Mexican too Translate.Aussie and then it finally hit me who it was that was calling! LOL An Oz accent was the *last thing I expected to hear sitting in a small town on the Texas-Mexico border! ROFL

We talked for some minutes and then I started getting lots of dirty looks from the workers (the phone with its short cord was sitting right in their ordering area and the lunch rush was on) so we said our good-byes and I once again thanked Andy for making my day.

As a side note, once I told the workers that the call had come from Australia they got very sheepish about their cutting us off. :-[

OH and when is the rest coming Java.....great story....phile???

As to the rest of the story it will probably happen in bits and pieces as the Ministry of Finance authorizes the expenditure of funds for the gears and machine shop time for removing the broken bolt. Until then Ill just have to limp along on 2 barrels.

Im still experimenting with roasting times and profiles trying to come up with the ideal ones and as yet have really come to no firm conclusions but its looking like a 12 minute roast is working better than an 8-10 minute one. Im also running a few batches out to as long as the 15 minute mark to see if that makes a difference as well.

I will post more stories as people have questions or as things happen here.

Java "Google is your friend!" phile

[Javaphile]

Re: A (Epic?) journey in search of my perfect roas

A quick update as Ive had several questions about the status of my Gothot.

Im still only roasting in 2 barrels as the Ministry of Finance has not yet approved the outlay needed for the new gears. Hopefully this will happen in a couple of weeks if things work out as planned.

Getting new cooling trays made up has also become a higher priority as the fan on my Bean Bucket Cooler died in the middle of a roasting session here a couple of weeks ago and so all the cooling is now done in the roasters cooling tray. Which with only one of them present makes for some interesting timing issues. Use of the roasters cooling trays is preferred here now as were in the full heat of summer (38C for the next week predicted) and the heat from the cooling beans is vented outside rather than being put into the room as with the BBC.

The MicroDrive has continued to work flawlessly powering the roaster, converting the 120v 60Hz Single Phase too 240v 2 Phase giving what seems to be the appropriate revolutions when set at 40Hz.

A batch size of 200g is working very nicely with all the chaff being cleared from the roasting chamber with only an occassional bean being sucked out.

The roaster has now had almost 200kg of green beans put through it since coming on-line last December and is still performing perfectly.

Propane consumption has been very low and Im still using the original 15lb tank I bought for it with maybe a quarter of it left.

The Powder Coating has stood up perfectly in all areas with the exception of the lower left side of the outter cover over the drums which has peeled off on both drums. This is the side on which the majority of the heat from the burners is drawn up around the drum due to its direction of rotation. This is also the side of the drums that the beans spend the majority of their time on due to the angle of the interior vanes and the drums rotation. The right side of the covers show no sign of peeling. With a 2,000F rating for the Ceramic Coating that was applied Im a bit surprised that its peeling as the drum certainly never comes anywhere close to approaching that kind of a temperature. I suspect that the problem is from the application of the coating and not because the cover is getting too hot.

Ive found that the InfraRed thermometer works very well for measuring the temp of the empty drum giving a very consistant reading both in the heat up and the cool down stage. Using it to read the temp of the roasting beans though is no where near as reliable. Between the motion of the beans, the hot air being circulated through the drum, the presence of chaff, and the smoke produced by the beans the temp shown by the IR varies too much to rely on it as a precise measurement of the bean temp. With the drum empty the reading is rock solid to the tenth of a degree. With beans in the drum however the reading will vary over about 2C early on in the roast, with the varience hitting over 5C in the latter stages and in some cases (usually with the beans that produce large sized chaff) as much as 10C.

The roasting time that seems to work best is in the 10-12 minute range with the draft open to about 3/4s. If I use less draft the chaff has a tendency to get stuck in the outlet of the barrel, especially on the larger chaff beans like the Nicaraguans. On the smaller bean sizes, notably peaberrys and a lot of Ethiopians, I can close the draft down to about 2/3rds. Once the summer is over here and I move back to espressos from my Cold Process I plan on doing more experimentation with roasting times to get a better idea of the difference in the cup between a 10 minute and a 15 minute roast.

As Im still operating on only the two barrels Ive yet to run any experiments on the difference between the middle burner and the two outter ones. I still suspect that the outter barrels gas systems are set up for natural gas while the middle one is designed for propane. Controlling the heat once 1st crack is reached is very problematic. The burners basicly have to be turned down to the point that they are barely staying lit with there being very little adjustability at that point. The flames will go from barely lit to 1cm high with nothing in-between. Above 1cm flame height there is full control, but below 1cm, nothing. If the flame height is turned up at all the beans will go straight from 1st into 2nd crack with no pause in-between. Not good. Once finances permit Im still planning on getting a whole new gas system for it which will allow me to fine tune my roasts better. The roaster is certainly usable as-is but it would be really nice to have more control over the flame height at the lower levels.

So over-all Im very happy with it and still having fun playing with it. It still has things to be done to it but hey, whats life with-out a few projects? Boring! ;D

I think that about covers it. Did I miss anything?

Java "Tries not to lead a boring life" phile

[Javaphile]

Re: A (Epic?) journey in search of my perfect roas

Somebody in another thread asked for pictures of my roasting set-up in action and I said Id take some during my next roasting session, which I just finished, so here ya go! Seeing as I dont recall which thread they were asked for in Ill post them here.

This pic shows the beans for the Cold Process blend all weighed out and ready to go.

Starting from the upper left the beans are: 400g of Peruvian La Florida, 800g of El Salvador Maxtapulat Estate Organic, and 1600g of Ethiopian Yirgacheffe Grade 1.

These were roasted as SOs in 200g batches and blended post roast. They will rest for a couple of days before being ground and soaked.