HomeRoast Digest


Topic: No max load gain from popper slot widening (23 msgs / 758 lines)
1) From: Douglas Strait
Motivated by various posts wrt increasing batch size by widening the 
slots of a hot air popper, I tried this on a 1400W Wearever pumper. I 
increased the width of 6 out of 16 slots by about 50%. This gives a 
total of about 20% increase in total air exit cross section. I found 
that this resulted in *no measurable increase* in the weight of beans 
that could be moved. All references to this modification where the 
popper was identified that I recall were specific to the West Bend P1. 
Has anyone had success with this modification on a Wearever 1400W 
pumper?
On a more successful note, I found that on the 1400W Pumper the batch 
size can be doubled by full wave rectification and filtering with 
220uF of the power to the fan.
Doug

2) From: Matthew Price
On 10/11/05, Douglas Strait  wrote:
<Snip>
My pumper already has a full wave rectifier across the fan.  This is
the one without the switch; is yours factory wired differently?  I
can't remember if the heater is also DC, but it must be since half of
it is in series with the motor.
Matthew

3) From: Gary Townsend
 Douglas Strait wrote:
<Snip>
 Doug,
I can assure you that this technique works very well on poppery 2 type
poppers, as well as Popcorn Pumpers...I have 2 of them that will all push 6
& 5/8 oz of greens...with a little bit of help from a wooden spoon. I adjus=
t
the load to *just* 5 oz, and use a glass chimney to stretch the roast time.
Which, BTW is about 14 minutes to FC+, and my roasts are very even and tast=
e
just fine.
 I have not yet widened a 1400w Pumper, yet, but I was looking at a way to
*open* up the airflow, after reading your post. I opened up all the slots o=
n
the poppery 2 types. The non-cast poppers are pretty easy to come by, and
fairly easy to widen. All you need is a 1/8" flat tip screwdriver blade, an=
d
you can easily bend the sheet metal. I would use my dremmel, but it died on
me 2 days ago, Sparkin' and arcing...So, I'll need to *upgrade* to a better
rotary tool, 1st. Maybe by this weekend...
I would attempt to open up all of the vents, and possibly look at opening u=
p
the sheetmetal that supports the fan motor. I have a few of those 1400w
Pumpers, and was never really happy with the load limit of 4 oz, especially
since it's beefy, cast chambers are very similar to the original poppery
1's.
 Roasting 5 to 6 oz in a $1.98 Pumper (1250) watt with a few 'no additional
cost' mods is fine for my needs. I can control the roast simply by adding
1/2oz of beans to the subsequent roast to say 5&1/2, and it will lower the
total roast time by about 2 minutes. If the roaster is not hitting 2nd crac=
k
at the 13 minute mark, I can induce 2nd by simply restricting the airflow
with a leather glove ( which I need to remove the glass chimney anyway,
before dumping the roast) placing a couple of fingers over the top of the
popper. Which causes the bean mass to drop back into the metal chamber.
 My advice is to keep trying... maybe widening all the slots will work out
just fine. Once that works with a stock fan, then any improvements to
boosting the fan should work that much better.
  *On a more successful note, I found that on the 1400W Pumper the batch
size can be doubled by full wave rectification and filtering with 220uF of
the power to the fan.*
**
  What's a "full wave rectification and filtering with 220uF" mean in
laymans terms ???
 Gary

4) From: Alchemist John
I also found no measurable increase in load as this is not the 
airflow bottleneck.  I opened the initial airflow holes (I actually 
removed the whole bottom plate.  That increased my load a good 40-50 
% if memory serves.
At 16:59 10/11/2005, you wrote:
<Snip>
John Nanci
AlChemist at large
Zen Roasting , Blending & Espresso pulling by Gestalthttp://www.chocolatealchemy.com/

5) From: Douglas Strait
Matthew,
The pumper that you describe is the 1250W model that uses a DC fan 
motor rated at 20VDC. The 1400W model uses a 120V series wound 
commutated motor. The 1400W model does have a switch on the front.
Doug

6) From: Edward Spiegel
At 7:59 PM -0400 10/11/05, Douglas Strait wrote:
<Snip>
I've done it with two 1250 W Wearever Pumpers and have just about doubled the amount of coffee that can be roasted with a decent profile. Before widening, my pumpers could handle about 110 grams with a decent profile (5 to 7 minutes to first depending on various factors) now I can easily handle 170 grams and can even do 200 grams if I do a little handholding (occasionally agitating the pumper).
Note that I use about a 20 to 30 degree tilt which makes a huge difference.
I widened my slots no more than about 25%
Best,
Edward

7) From: Douglas Strait
John: I think you are talking about the West Bend Poppery, not the 
Wearever 1400W pumper. I have measured the pressure drop across the 
*inlet* slots of an unmodified 1400W Wearever pumper and found it to 
be negligible [<0.1" H2O] so there is not much to be gained by 
widening the inlet slots. By comparision, the deadhead pressure of 
this fan is on the order of 4.5 to 5.0" H2O.

8) From: Douglas Strait
This is a multi-part message in MIME format.
Gary,
Thanks for your observations. 
Full wave rectification in this case is the use of 4 diodes in a =
"Bridge" configuration to convert the input AC voltage to a pulsating DC =
voltage. The 220uF is the value of the capacitor used to smooth the =
pulsating DC to a somewhat steadier DC. With a sufficiently large value =
for the capacitor this technique can generate a DC voltage as high as =
1.414 times the value of the input AC voltage. With the 220uF that I am =
using I get somewhat less; about 159VDC from 120VAC input under the load =
of the fan motor. This increased voltage to the fan doubles the mass of =
beans that can be moved relative to 120VAC input to the fan.
So far, from the responses to date to my original query, it appears that =
no one is claiming that the 1400W Wear Ever pumpers respond favorably to =
widening the roast chamber slots.
Doug

9) From: Michael Dhabolt
Doug
My measurements of the pressure drop across the inlet vanes in the P1 are
similar to your results with the 1400 Pumper:
 >I have measured the pressure drop across the *inlet* slots of an
unmodified 1400W Wearever pumper and >found it to be negligible [<0.1" H2O]
so there is not much to be gained by widening the inlet slots.
 My measurements of the fan pressure necessary to generate an initial "very
slow" bean circulation in a P1 tilted 15 degrees and loaded with 460 gm of
the smallest densest beans (whatever high grown peaberry I had in the stash
at the time) were also in line with your measurements. mine was 4.9" H20.
 >By comparison, the deadhead pressure of this fan is on the order of 4.5 t=
o
5.0" H2O.
 What did make a difference in maximum load with the P1 was opening up the
inlet air passages into the fan housing. Quick and dirty way of
accomplishing it with the P1 is leaving the bottom plastic and tin off and
supporting the unit off the counter 1/2" with rubber "feet". Unfortunately
this also adds considerably to the ambient noise generated by the fan.
Opening up the original holes to 250% of the original inlet area and just a
little noise deadening material on the interior of the bottom portion of th=
e
original outer plastic housing gives the same flow increase with a
noticeable decrease in fan noise.
 Mike (just plain)

10) From: Alchemist John
OI, right you are.  Thanks for keeping me on my toes.
At 17:03 10/12/2005, you wrote:
<Snip>
John Nanci
AlChemist at large
Zen Roasting , Blending & Espresso pulling by Gestalthttp://www.chocolatealchemy.com/

11) From: Douglas Strait
This is a multi-part message in MIME format.
Michael
Thanks for sharing your observations. Based upon what I *think* I know =
about centrifical fans, I am completely baffled by your gain in flow by =
widening the inlet slots *if in fact* there was negligible pressure drop =
across them prior to widening. BTW, the fan pressure measured with the =
chamber fully loaded with beans is far from a deadheaded condition.  At =
120VAC nominal, a Wear Ever 1400W I checked measured 4.5" deadheaded but =
only 2.4" fully loaded to point of no bean movement and 1.8" empty. 
It is my observation that the fan in the Poppery I is a better and more =
efficient design than the one in the Wear Ever 1400W despite it's =
smaller diameter. Here is some misc data comparing the two fans for =
those who may be interested: 
Blade diameter: Poppery I is 4.0", Wear Ever is 4.4"; current draw with =
chamber empty: Poppery I is 0.54A, Wear Ever is 0.66A. These =
measurements were made on a Poppery with the 0.84A nameplate "Wah Ming" =
motor and a Wear Ever with the 0.64A nameplate "Johnson" motor. The =
current draw is at maximum with the chamber empty and decreases with =
bean load. Deadheaded, the Wear Ever measured 0.48A.
I made the Wear Ever fan pressure measurements by connecting the =
manometer to the small hole associated with the butter tray location. I =
have no pressure measurements on a Poppery. It appears that one would =
need to drill a hole in the fan housing on a Poppery to get a pressure =
measurement. 
Other data: Here is a table of fan pressure vs voltage with the chamber =
loaded to the maximum bean load that will barely move without tilting. =
Applied fan voltage is PWM DC and the motor has a catch diode installed =
across it's terminals.
Voltage  Head in "H2O  Load grams 
70             1.05                   27 
80             1.1                     58 
90             1.15                   87 
100           1.30                  115 
110           1.5                    145 
120           1.65                  174 
130           1.9                    211 
140           2.1                    246 
150           2.3                    278 
160           2.5                    345 
Pressure measurements made with a simple U-tube manometer and read as =
best I could to the nearest 0.05".  These data are from the 1400W Wear =
Ever that had 6 of 16 chamber slots widened by about 50%.
Doug

12) From: dsobcz716
I'm confused.  Following the thread, I'm not sure if you are talking about the slots in the roast chamber or the slots on the bottom the machine, or both.  I'm roasting with a poppery II and I would love to increase my batch size from the 95-100 grams I'm roasting now.
Dave

13) From:
Doug-
That fan motor probably has a full-wave pair of diodes soldered to its
terminals already. In fact, I'm surprised if it doesn't have 4 diodes in a
full wave bridge on the motor terminals.
Your addition of the cap. would sure put it in A/B!
Must be the ports already maximized the airflow for that little blower.
TIP: find a spare popper or two so you will have a spare blower assembly.
Think you'll ever find replacement brushes or other spares independently?
No!
You can always soup up a DC motor, at the expense of brush and commutator
life. If it's the same as mine, the motor has a Permanent Magnet field, and
you're putting power on the armature through the brushes.
Cheers -RayO, aka Opa!
Eschew Ersatz Grinders!

14) From: Michael Dhabolt
Doug,
 My comments re: the increase after enlarging the inlet area (pardon my poo=
r
description), were referring to the air inlet into the bakelite fan housing
- not the inlet to the roast chamber.
 Your comments:
 >I have no pressure measurements on a Poppery. It appears that one would
need to drill a >hole in the fan housing on a Poppery to get a pressure
measurement.
 Are also correct. I drilled into the radial volume in the aluminum housing
just prior to the inlet vanes. I didn't do nearly as complete an analysis a=
s
you. I was interested in quantifying the required pressure at the most
limiting roaster condition (for my roasting technique). I was varying the
roast chamber angle and bean load - trying to identify a given angle for th=
e
maximum bean load with the fan driven by 140V AC. For my personal roasting
technique it came out to 260 gm of the highgrown peaberry (most difficult
bean to circulate) at an angle of 15 degrees.
 At the time I was also experimenting with some other fans coupled to the P=
1
roast chamber casting. As Ray-O has highlighted, the fan would seem to be
the weak link and it would be nice to have an easily available replacement
for the motor identified. I have not resolved this issue - as of now.
 Your current draw numbers vs flow are to be expected in a fan or
centrifugal pump - highest current when doing the most work (highest flow,
generally (flow/pressure relationship)) - current draw is radically
decreased in a "loss of flow" status. As expected the resulting temperature
increase in the motor (and heating element) at the no flow fan status is
immediate and catastrophic. The nichrome element burns out first (if the
thermostat has been removed or disabled) followed within a couple of minute=
s
with the fan motor letting out the "magic smoke". What my experiments did
however, is convince me that the very low fan flow rate required to produce
my preferred slow circulation of bean mass at the beginning of the roast
(rather than true fluid bed lofting of the bean mass) is adequate to protec=
t
the fan from self destruction. The P1 fan motor continues to surprise me, i=
t
is considerably more robust and long lived than I originally expected.
 Mike (just plain)

15) From: jim-seaman
Not sure about the Poppery 2 or the Wearever model but the Poppery 1 works great with the changes Mike Dhabolt has suggested.  I mirrored his setup as closely as possible and have gotten great results.  I roasted 9.2 oz [262 gms] (7.2 oz of Sulawesi with 2 oz of Harar) just the other day.  I was pressed to get above 5 oz [142 gms] before Mike's recommended external slot modifications. (Nothing was done to the inner slots.)  [Thanks again to Mike Dhabolt!]
One thing to note is that the Poppery 1 has different 'vanes' at the bottom of the roasting chamber.  They are cast into the 'potmetal' of the blower housing.  The Poppery 2's that I've seen have taller but thinner slots that look to be 'punched and then bent' from the roasting chamber's wall.  What this tells me is: that I don't have any idea how a Poppery 2 would behave with the same mods.  I'm sure I'll try it some time in the future though...
I've got one spare poppery 1 that I'm saving for replacement parts and will be looking for more of the same or models with similar innards.
Anyone know which makes and models have ac motors and similar roasting chambers?
Thank you,
Jim
Woodinville, WA
Enjoying a Monkey blend Americano this morning!  
Not sure about the Poppery 2 or the Wearever model but the Poppery 1 works great with the changes Mike Dhabolt has suggested.  I mirrored his setup as closely as possible and have gotten great results.  I roasted 9.2 oz [262 gms] (7.2 oz of Sulawesi with 2 oz of Harar) just the other day.  I was pressed to get above 5 oz [142 gms] before Mike's recommended external slot modifications. (Nothing was done to the inner slots.)  [Thanks again to Mike Dhabolt!]
 
One thing to note is that the Poppery 1 has different 'vanes' at the bottom of the roasting chamber.  They are cast into the 'potmetal' of the blower housing.  The Poppery 2's that I've seen have taller but thinner slots that look to be 'punched and then bent' from the roasting chamber's wall.  What this tells me is: that I don't have any idea how a Poppery 2 would behave with the same mods.  I'm sure I'll try it some time in the future though...
 
I've got one spare poppery 1 that I'm saving for replacement parts and will be looking for more of the same or models with similar innards.
 
Anyone know which makes and models have ac motors and similar roasting chambers?
 
Thank you,
 
Jim
Woodinville, WA
Enjoying a Monkey blend Americano this morning!  

16) From: Douglas Strait
This is a multi-part message in MIME format.
Ray,
What you are describing applies to the West Bend Poppery II and the Wear =
Ever 1250W pumper. The 1400W Wear Ever [subject of my original query] as =
well as the original West Bend Poppery [called the P1 for convenience by =
many of us] both use a 120V series wound communtated motor without any =
diodes. 
Since my last posting I disassembled both a Wear Ever 1400W pumper and a =
West Bend P1 and measured or estimated the net cross sectional area of =
each restriction in their respective flow paths. Each have similar net =
oriface in the discharge to roast chamber slots [net about 0.64 sq in =
(16 slots 0.1"X0.4")] but beyond that they differ greatly. One key =
difference is that the flow path in and around the heating element of =
the Wear Ever is significantly restricted however for the P1 it is not. =
This difference appears to be the major limiting restriction for the =
Wear Ever and hence the lack of benefit seen when I widened the slots =
which discharge into the roast chamber. It would therefore appear that =
the only means of getting significant bean load increase on the 1400W =
Wear Ever is by increasing motor voltage. I also went back and =
reverified my original measurement that the pressure drop thru the =
plastic housing inlet air slots [23 slots about 0.4"X0.09"] is =
negligible. I measured about 0.05" H2O at 120VDC to fan with roast =
chamber empty [max air flow]. Another consequence of the restricted flow =
path in and around the heating element of the Wear Ever is that my =
choice of locations for pressure measurement [the butter tray port] is =
downstream of some flow restriction from the true fan discharge. Thus, =
except for deadhead pressure, my measurements are of relative value =
only. What you did on your P1 more truly measures fan discharge =
pressure.
Doug

17) From: Douglas Strait
This is a multi-part message in MIME format.
Mike,
Since my last posting I disassembled both a Wear Ever 1400W pumper and a =
West Bend P1 and measured or estimated the net cross sectional area of =
each restriction in their respective flow paths. Each have similar net =
oriface in the discharge to roast chamber slots [net about 0.64 sq in =
(16 slots 0.1"X0.4")] but beyond that they differ greatly. One key =
difference is that the flow path in and around the heating element of =
the Wear Ever is significantly restricted however for the P1 it is not. =
This difference appears to be the major limiting restriction for the =
Wear Ever and hence the lack of benefit seen when I widened the slots =
which discharge into the roast chamber. It would therefore appear that =
the only means of getting significant bean load increase on the 1400W =
Wear Ever is by increasing motor voltage. I also went back and =
reverified my original measurement that the pressure drop thru the =
plastic housing inlet air slots [23 slots about 0.4"X0.09"] is =
negligible. I measured about 0.05" H2O at 120VDC to fan with roast =
chamber empty [max air flow]. Another consequence of the restricted flow =
path in and around the heating element of the Wear Ever is that my =
choice of locations for pressure measurement [the butter tray port] is =
downstream of some flow restriction from the true fan discharge. Thus, =
except for deadhead pressure, my measurements are of relative value =
only. What you did on your P1 more truly measures fan discharge =
pressure.
Doug

18) From: Michael Dhabolt
Doug,
After giving your analysis (a lot of good data - keep it up) a lot of
thought and rethinking my own, I've come to the conclusion that the
increased head pressure from speeding up the fan is probably what benefited
my roast volume the most.
 I don't currently "loft" nearly the quantity of beans during the roast tha=
t
I used to. I do circulate the beans considerably faster during the middle
and especially the later stages of the roast than the minimal circulation
during the initial drying stage (<280 degrees F). During these later stages
of the roast I lower the fan speed far enough to just keep from charing and
tipping the beans. After the drying stage I could keep all the beans in the
air but the roast seems to have a fuller, softer more complete taste (that =
I
like) if I approach it from a circulation perspective rather than a fluid
bed approach. I'm sure that this approach would be futile without a PID or
variac to control heat to that necessary to match the circulation (or
fluidization) rate. And, what I am looking for in the roast is undoubtedly
not everyones preference.
 But then again - I have a tendency to talk myself into things based on
incomplete data - and yours is really useful and appreciated.
 Mike (just plain)

19) From: Alchemist John
At 07:48 10/14/2005, you wrote:
<Snip>
No, the P1 is pretty much a class by it self.  I actually tried to 
track down motors like that to nut much avail (at least for a reasonable price.
<Snip>
John Nanci
AlChemist at large
Zen Roasting , Blending & Espresso pulling by Gestalthttp://www.chocolatealchemy.com/

20) From: jim-seaman
Looks like I may have attributed the increased volume I'm roasting to the other changes I made as well.  Doh! :)  I added a lot of things to the mix before trying it out. (popper external air intake mods, pid, small variac to control fan speed, etc.)  Mike Dhabolt is probably right and the variac is largely responsible for the change.
Didn't get a chance to read the list for a few days and there's a lot to catch up on.
Ray, keep us posted on your larger air roaster project.  Sounds fascinating!
Tried my hand at blending espresso for the first time.  I used about 80% Sulawesi and 20% Harar Horse and it actually turned out pretty good!  Seems like I need at least one more coffee in there though...  (Maybe something with a brighter note...)
I've determined that I don't care for Sulawesi as drip but like it as espresso.  Its funny how different a sulawesi americano tastes compared to a pot of sulawesi drip.  I'm sure a lot of it depends on the drip machinery as well.
Hope everyone is enjoying their coffee as much as I am this morning.
Regards,
Jim
Woodinville, WA
Looks like I may have attributed the increased volume I'm roasting to the other changes I made as well.  Doh! :)  I added a lot of things to the mix before trying it out. (popper external air intake mods, pid, small variac to control fan speed, etc.)  Mike Dhabolt is probably right and the variac is largely responsible for the change.
 
Didn't get a chance to read the list for a few days and there's a lot to catch up on.
 
Ray, keep us posted on your larger air roaster project.  Sounds fascinating!
 
Tried my hand at blending espresso for the first time.  I used about 80% Sulawesi and 20% Harar Horse and it actually turned out pretty good!  Seems like I need at least one more coffee in there though...  (Maybe something with a brighter note...)
 
I've determined that I don't care for Sulawesi as drip but like it as espresso.  Its funny how different a sulawesi americano tastes compared to a pot of sulawesi drip.  I'm sure a lot of it depends on the drip machinery as well.
 
Hope everyone is enjoying their coffee as much as I am this morning.
 
Regards,
 
Jim
Woodinville, WA

21) From:
Doug-
I don't have a P1, but you can accomplish a boost in the motor's output as
long as it has brushes. Just hook 4 diodes to it in a full wave bridge
configuration and add a filter cap across the motor terminals. (Two anodes
to one motor terminal, two cathodes to the other.)
Just the f-w bridge alone should raise the rms value of the power you're
feeding to the motor and speed it up. As you add more and more capacitance,
the rms value will rise, almost to the sine wave peak value which is 1.4 x
the ac line rms value. "Look ma- no transformer required!"
This increase is not free, though. The brushes and and the commutator on th=
e
motor's armature will wear faster, and the line current load will rise a
bit.
Remember, your actual line voltage varies during the day, and these
variations have a surprisingly large effect on the heat output of the
nichrome heater element. Usually, the line voltage is highest in the wee
hours before sunrise when power usage is lowest.
Cheers -RayO, aka Opa!
One good shot deserves another-

22) From: Douglas Strait
This is a multi-part message in MIME format.
<Snip>
you're feeding to the motor<
Full wave rectification alone without filtering does not raise the RMS =
value of the resultant waveform in the absense of filtering. In =
practice, the RMS value is 1-2 volts lower due to the two diode forward =
voltage drops in the bridge.
As a comment to those who may be considering using the rectification and =
filtering technique that Ray and I are discussing, be aware that when =
the circuit is first powered up the discharged capacitor appears as a =
dead short to the diode bridge. The practical consequence of this is =
that even though the average current is less than one amp, the diode =
bridge needs to be fairly beefy to handle the inrush. For my popper, I'm =
using a packaged 4A bridge assembly rated for 150A 8.3msec surge.
Doug

23) From:
Uncle!
A mind is a terrible thing to waste.
Engage brain before releasing tongue...
I was thinking of the consequences of trying to use a single diode-
(1N4002?) anything to get dc.
The PM-field motors are so cheap and plentiful- that's how they save a buck=
.
But they require dc power, so the diodes. How else but a single half wave o=
r
4 diode f-w bridge, without adding a transformer?
I'm going to cobble up my Tesla coffee roaster using a Kirby- or whatever
blower motor, some thrift store hair dryer heaters, a Pyrex glass cylinder,
and maybe a couple of Cherry Bomb glass packs to quiet the intake and
exhaust of the blower...
Or something using a gas dryer burner along with its blower... I need
coffee!!!
Cheers -RayO, aka Opa!
--
The only reason for time is so that everything doesn't happen at once- - -
A. Einstein


HomeRoast Digest