For those with Hottop roaster it has either come about or will soon that you will need a replacement filter for the main, rear fan. A part of the original filter that comes with the roaster is made of a paper-like material and is not washable. If you don't replace it the filter will not pass enough air and the heat that builds up in the roast chamber will be sufficient to signal an overheat condition and prematurely eject the beans. If you remove the filter completely the heat loss will cause an under-roast condition. The only alternative was to purchase replacement filters which come in the original holder and cost about $10 each plus shipping. The factory states a life of about 40 roasts, so the cost per roast is enough to make us search out a replacement material.

      I wanted to find a material that was inexpensive, easy to handle, and preferably reusable. It had to at least be washable if not permanent in nature. It also had to be able to handle the heat of the exhaust. The furnace filters I found were not suitable because they were fibrous (like loose Fiberglas in nature) or not designed to take the heat. On one of my shopping excursions online I was looking at some affordable 120mm computer fan filters. Check this GOOGLE search as they are available for about $3 to $4 each plus shipping as of late 2009. As an example, at ModderSmart, two filters cost just a bit under $9 including shipping.

NOTE: Since we are using these filters in a manner very different from that which they were intended, be sure to examine them carefully before use because exposing them to the heat and filth from the roaster will most likely void any possible warranty they might have had.

      As you can see here, on the left is the stock filter as was supplied with the first run of Hottops. On the right is the filter holder with the stainless filter installed. In the center is the removed and now unused filter media retainer. The metal filter couldn't be a much better fit. I was a bit apprehensive about the extent of modifications I would have to make, but none were necessary. The inset shows the mesh material. It sort of reminded me of a range or stove filter. I blew through it and it seemed to afford some amount of air resistance and that pleased me. My biggest worry was that it would allow too much air flow, but that didn't seem the case in this highly-unscientific, preliminary test procedure.
The installation was as follows:

1) Remove rear filter assembly from roaster by pulling it up and out of the rear cover.
2) Push back the little molded-in clips that hold the filter medium retainer onto the filter holder. Work around until sufficient numbers of clips are released to allow the to sections to be separated. Put the medium retainer aside.
3) Carefully remove the original filter material and set it aside.
4) Place the replacement filter into the main frame.
5) Slide the filter assembly back into the roaster.

      That's it! I was happily surprised to find that the replacement filter was just about a perfect fit, and it is held by the original clips of the main filter carrier. The retainer would not fit without some modifications, but the filter seated so well I decided not to even investigate further. Even if yours does not fit correctly, the outer frame of the new filter is aluminum so it should be an easy matter to crush it a bit to make it wider if it is too loose, or even make some thin, aluminum shim material out of some imported sheet stock (in other words, get a beer can and go to work with a pair of scissors). If not, the four corner holes should allow the use of some sort of fastener. The filters each come with four computer-fan screws so I suppose those could be used, but I really don't think it would be necessary. Too tight should not be a problem. But again, the thin aluminum frame should allow some forming of the filter to get it to fit.

      Of course, all of that means nothing if it doesn't allow the roaster to operate properly. To find out I did the following test:

      My impression is that it does feel like it flows more air, or at least flows air more freely. It does catch the bits of chaff and such, but of course, will not filter out the micro-fine particulate matter that is found in smoke- all that brown and tan muck that fouls the stock filter material. Because of that, if you use a metal-mesh filter, it is important to stay out of the smoke as it has the potential of being a much greater irritant than if the cloth or paper filter is used.

      Examining the results of the three roasts (the warm-up roast and the two test roasts), it would be appear that they were identical. All were ejected with only a very few droplets of oil showing on the beans in the cooling tray (three or four drops seen per batch).

      I would not take too much stock in the temperature figures above. With the thermometer in the chaff tray, when the Hottop fan comes on it tends to pull the heat away from the thermometer. With the fan on high during second the indicated temperatures actually drops. The time from the beginning of first crack to the ejection point are quite close with the three batches:

      Add or subtract a few seconds for variables such as:
    - Operator error
    - Operator's perception of the level of activity in second crack
    - Bean Roaster's inaccuracy in temperature control
    - Environmental differences in temperature and humidity

      Based on that, and the small differences between the roast times, I think it would be fair to state that these times seem to indicate that the metal filter operates much like the cloth filter, or at least to say, the roaster seems to compensate for the added air flow.
      At this point I am going to leave the metal filter in place and continue to use it. If you choose to try it and find that it passes too much heat, use a layer or two of metal window screen or other fine mesh screening over the stainless steel filter to further reduce air flow through it. This added screen could be cut a bit oversize (by about 2mm) and would make a nice retainer for the mesh filter.

      The only drawback for me was that at the end of the roast I went to withdraw the filter carrier and I found that the mesh filter fell out of the carrier. It is possible for it to become lodged in the cavity making it necessary to remove the fan assembly to get the filter out. I recommend some sort of retaining system for the filter don't just snap it into the carrier and hope it stays there. You could use a piece of thin brass or copper wire, and tie the filter onto the holder to keep the assembly together.

      Two washable and reusable filters delivered for less than $17? Sounds like a deal to me! I will try to remember to post further observations in a month or so after I complete a few batches of my normal roast. And, as always, if you have any questions or comments, feel free to send them to me and I will do my best to respond.

      ADDENDUM (10/5/2003)- The above test roasts were done in my garage where I could better control conditions- mainly to keep the machine sheltered from our daily, gusty winds. This morning I roasted in my usual location, outdoors on the deck. I set the roaster on a picnic bench along a wall of the house that is mostly sheltered from the winds, but it is exposed to the sunlight so the smoke production can be clearly seen. It was evident this morning that the original cloth fiber filter does remove a lot of the particulate matter from the smoke making it less visible. With the mesh filter the smoke was prodigious. Near the end of the roast it appeared to be about 150-200% of what the machine normally produces. The roast progressed fairly normally, and I was able to get 300 grams into second crack with the use of one 'Plus,' and ejected manually before the second ejection signal sounded, so heat loss through the new filter did not seem to affect the roast to any great degree.

      ADDENDUM - [11/07/2003]- I had the opportunity to roast out of doors with the new filter in some relatively cold weather about two weeks ago (it was in the high 30s out). The roaster had a difficult time adjusting to the cold temperatures (at least that's what I assumed). After nearly the entire roast cycle passed, the beans were still green and the roaster only seemed lukewarm on the outside. I assumed that somehow it tried to compensate for the cold air and overheated itself, triggering the safety programming, and shut off the heating element. In any case, I ejected the beans, ran one roast without beans to preheat the roaster, then all was fine, and I was able to complete the next roast normally. For those who roast in cold climates, try preheating the machine by running a "Level 1" roast without beans before actually attempting to roast. It might help.

      ADDENDUM - [3/4/2010]- These permanent filters allow a lot of airflow. Becasue of that, if you are modifying the roast for a shorter roast time you need to modify the filter. You can try covering it partially with aluminum foil. I suggest covering the middle 2/3's or so leaving about 20mm at the top and 20mm at the bottom open and seeing how that works for you.