[clagwell]

Intro and Bottom Line

As many other people have already done I swapped my 10" HF impeller for a 12" Rikon replacement. Performance is subjectively better but I wanted to know how much improvement I actually got. There were several questions I wanted to answer and I've done a lot of testing that's resulted in quite a bit of data. I'm using the blog format to help organize the presentation of that data.

I won't make you wade through all of the up coming posts to get to the bottom line though. Here are the short answers:

1- How much better is the suction?

Roughly a 10% flow rate increase, 20% pressure, varying with the amount of restriction in the rest of the system (hoses, ducts, etc.).

2 - How much does the current increase?

That depends on flow rate. For typical operating conditions it will be in the 15 to 20% range.

3 - Will the motor burn up?

Well, running the fan all by itself, that is, no restrictions on either end, the motor can easily overheat even with the original impeller. Under normal operating conditions the temperature rise will be 15 to 20C higher with the Rikon impeller. That decreases insulation life by a factor of about 3 to 4. So, no it won't "burn up", but it certainly won't last as long.

4 - Does startup current increase?

The starting current is about 60A regardless of impeller choice but the larger rotational inertia of the larger fan causes the current to stay at that value for a longer time. To reliably start the dust collector with the Rikon impeller the circuit breaker's trip curve needs to allow 60A for about 4 seconds.

Actual answers are, of course, more complicated than that. I'll get into the details in later posts for anyone who's interested.

Oh, and about that original question? Was it worth it? I don't know. It's provided plenty of entertainment in the form of taking measurements, plotting results and fitting curves. I made the modification after using the HFDC for about eight years so to me it was an incremental improvement. Compared to similar dust collectors with similar performance there's a cost advantage, but part of that cost difference is due to the motor, which is marginal for a 12" fan. More on that in a later post.

 

[jamsomito]

Intro and Bottom Line

As many other people have already done I swapped my 10" HF impeller for a 12" Rikon replacement. Performance is subjectively better but I wanted to know how much improvement I actually got. There were several questions I wanted to answer and I've done a lot of testing that's resulted in quite a bit of data. I'm using the blog format to help organize the presentation of that data.

I won't make you wade through all of the up coming posts to get to the bottom line though. Here are the short answers:

1- How much better is the suction?

Roughly a 10% flow rate increase, 20% pressure, varying with the amount of restriction in the rest of the system (hoses, ducts, etc.).

2 - How much does the current increase?

That depends on flow rate. For typical operating conditions it will be in the 15 to 20% range.

3 - Will the motor burn up?

Well, running the fan all by itself, that is, no restrictions on either end, the motor can easily overheat even with the original impeller. Under normal operating conditions the temperature rise will be 15 to 20C higher with the Rikon impeller. That decreases insulation life by a factor of about 3 to 4. So, no it won't "burn up", but it certainly won't last as long.

4 - Does startup current increase?

The starting current is about 60A regardless of impeller choice but the larger rotational inertia of the larger fan causes the current to stay at that value for a longer time. To reliably start the dust collector with the Rikon impeller the circuit breaker's trip curve needs to allow 60A for about 4 seconds.

Actual answers are, of course, more complicated than that. I'll get into the details in later posts for anyone who's interested.

Oh, and about that original question? Was it worth it? I don't know. It's provided plenty of entertainment in the form of taking measurements, plotting results and fitting curves. I made the modification after using the HFDC for about eight years so to me it was an incremental improvement. Compared to similar dust collectors with similar performance there's a cost advantage, but part of that cost difference is due to the motor, which is marginal for a 12" fan. More on that in a later post.

Hey, my new rikon impeller is stilling on the cart of my HF DC as I'm reading this, cool

Nice summary, and very nice measurements / curves. Seems about what we expected, but I don't believe I've seen as much of a scientific summary before. Thanks for the insight, looking forward to your more detailed posts later. I'm interested to hear your test setup and your system configuration and your "subjective" thoughts on real world performance differences.

 

[clagwell]

Airflow Measurement Setup

Test Duct
My test duct is a 10' length of 6" PVC drain pipe (ASTM 2729). This is the type of duct often used for dust collection in home workshops and will most likely be repurposed for that in the future (after I plug all of the holes!).

At the midpoint of the pipe is a measurement port for a hot wire anemometer or pitot tube. There are three pressure ports located at 17" from the duct inlet, 57" from the duct inlet, and 22" from the fan. Performance curves used pressure data from the port closest to the fan.

The pressure ports are implemented as four tap piezometer rings.

Instruments
Another forum had discussed the Testo 405i hot wire anemometer and it's ability to record data on a phone via Bluetooth. That led me to their 510i manometer. It looked like the pair would save me a lot of time since I could record both flow rate and pressure and then dump the data to a spreadsheet. I took a chance and bought one of each.

Unfortunately the 405i turned out to be very wrong for use in a 6" duct and was demoted to thermometer.

I ended up using the 510i with a pitot tube for the flow measurements and had to use my old generic manometer for pressure. In retrospect it would have been better to buy a second 510i instead of the 405i.

Traverse
Airflow velocity is not uniform in a duct so to get accurate numbers it's necessary to sample the velocity at various points within the duct and then use the average. In a round duct this is done by testing at several specific points along a diameter and then repeating at different angle(s). There are industry standards that specify the location of the points. A single pass across a diameter is called a traverse.

I'm a bit of a stickler about repeatability so I 3D printed an indexing jig for doing the traverse.

Restrictor
There are various techniques for restricting the duct inlet. One that you see often is a cone that is an adjustable distance from the entrance. It can be fine tuned if you are trying to set a particular pressure or flow. Another is to use various sizes of orifice. You're limited to a fixed number of restriction values but the repeatability of the restriction is excellent. I chose repeatability over fine tuning and 3D printed a number of orifice plates.

 

[Redoak49]

Airflow Measurement Setup

Test Duct
My test duct is a 10' length of 6" PVC drain pipe (ASTM 2729). This is the type of duct often used for dust collection in home workshops and will most likely be repurposed for that in the future (after I plug all of the holes!).

At the midpoint of the pipe is a measurement port for a hot wire anemometer or pitot tube. There are three pressure ports located at 17" from the duct inlet, 57" from the duct inlet, and 22" from the fan. Performance curves used pressure data from the port closest to the fan.

The pressure ports are implemented as four tap piezometer rings.

Instruments
Another forum had discussed the Testo 405i hot wire anemometer and it's ability to record data on a phone via Bluetooth. That led me to their 510i manometer. It looked like the pair would save me a lot of time since I could record both flow rate and pressure and then dump the data to a spreadsheet. I took a chance and bought one of each.

Unfortunately the 405i turned out to be very wrong for use in a 6" duct and was demoted to thermometer.

I ended up using the 510i with a pitot tube for the flow measurements and had to use my old generic manometer for pressure. In retrospect it would have been better to buy a second 510i instead of the 405i.

Traverse
Airflow velocity is not uniform in a duct so to get accurate numbers it's necessary to sample the velocity at various points within the duct and then use the average. In a round duct this is done by testing at several specific points along a diameter and then repeating at different angle(s). There are industry standards that specify the location of the points. A single pass across a diameter is called a traverse.

I'm a bit of a stickler about repeatability so I 3D printed an indexing jig for doing the traverse.

Restrictor
There are various techniques for restricting the duct inlet. One that you see often is a cone that is an adjustable distance from the entrance. It can be fine tuned if you are trying to set a particular pressure or flow. Another is to use various sizes of orifice. You're limited to a fixed number of restriction values but the repeatability of the restriction is excellent. I chose repeatability over fine tuning and 3D printed a number of orifice plates.

Very Good setup and will be interested in the results.

 

[a1Jim]

Airflow Measurement Setup

Test Duct
My test duct is a 10' length of 6" PVC drain pipe (ASTM 2729). This is the type of duct often used for dust collection in home workshops and will most likely be repurposed for that in the future (after I plug all of the holes!).

At the midpoint of the pipe is a measurement port for a hot wire anemometer or pitot tube. There are three pressure ports located at 17" from the duct inlet, 57" from the duct inlet, and 22" from the fan. Performance curves used pressure data from the port closest to the fan.

The pressure ports are implemented as four tap piezometer rings.

Instruments
Another forum had discussed the Testo 405i hot wire anemometer and it's ability to record data on a phone via Bluetooth. That led me to their 510i manometer. It looked like the pair would save me a lot of time since I could record both flow rate and pressure and then dump the data to a spreadsheet. I took a chance and bought one of each.

Unfortunately the 405i turned out to be very wrong for use in a 6" duct and was demoted to thermometer.

I ended up using the 510i with a pitot tube for the flow measurements and had to use my old generic manometer for pressure. In retrospect it would have been better to buy a second 510i instead of the 405i.

Traverse
Airflow velocity is not uniform in a duct so to get accurate numbers it's necessary to sample the velocity at various points within the duct and then use the average. In a round duct this is done by testing at several specific points along a diameter and then repeating at different angle(s). There are industry standards that specify the location of the points. A single pass across a diameter is called a traverse.

I'm a bit of a stickler about repeatability so I 3D printed an indexing jig for doing the traverse.

Restrictor
There are various techniques for restricting the duct inlet. One that you see often is a cone that is an adjustable distance from the entrance. It can be fine tuned if you are trying to set a particular pressure or flow. Another is to use various sizes of orifice. You're limited to a fixed number of restriction values but the repeatability of the restriction is excellent. I chose repeatability over fine tuning and 3D printed a number of orifice plates.

Wow very scientific my test is as basic as it gets "does it work for what I need it to do" It's great that you have the equipment and know-how to test this.

 

[clagwell]

Help! My motor's smoking!

Well, no, it's not really. I was just afraid that no one would look at a post with a boring title like "Temperature Rise".

Temperature rise may be boring but it's important to the life of a motor. In fact, it's critical to the lifetime of nearly everything electrical and electronic. Insulation weakens with age. It ages faster at higher temperatures. The hotter a motor runs the shorter it's life. The same for an extension cord.

When operated at it's temperature rating the motor insulation will have lost no more than 50% of it's mechanical strength in 20,000 hours. There's a rule of thumb that a change of 10C in operating temperature results in a factor of two change in aging rate. That is, 10C hotter means half the lifetime.

A piece of wire has it's temperature rating marked on the wire along with it's diameter (gauge). Motors with a NEMA nameplate label their temperature rating as an insulation class with a letter designation. For example, a Class B motor is allowed 80C rise with a 40C ambient (40C is the default ambient temperature for ratings). Class F is 105C rise and Class A 60C. Nearly all US made new motors are Class F. Many of the Asian motors are Class A or B. When you see "HEAT 60C" on a Taiwan made motor that's Class A.

Heat in the motor is mostly produce by current flowing through a resistance. That heat varies with the square of the current value. It's produced in both the stator windings and rotor bars. It's the temperature of the Copper stator windings that's important here.

Let me repeat, it's the temperature of the windings not the outside temperature of the motor that counts. The winding temperature can be much higher than the casing, especially on a TEFC (totally enclosed fan cooled) motor.

The usual way of measuring the winding temperature is the Resistance Method from the NEMA MG 1 spec. It's not difficult but you can't do it with a common Ohmmeter and it can involve some personal risk. For that reason I am not going to describe the method.

Results

The Harbor Freight has no nameplate. The full load current and insulation class are unknown, so I started at a low current value and worked my way up. Each test takes an hour to reach a stable temperature and several to cool down. When I got to the third value of 15.8A the temperature rise was 85C so I stopped. Not knowing the temperature rating of the motor I really didn't want to run it for an hour above that.

The curve fit is simple square law. Above 16A it's an extrapolation. At some point above that, I don't know where, the curve will get steeper because the resistance of the windings increases with temperature. The motor might actually go into thermal runaway and the curve turn into a hockey stick. Above 16A the curve should be seen as a minimum, temperature rise will be at least what's shown.

 

[1965scooper]

Help! My motor's smoking!

Well, no, it's not really. I was just afraid that no one would look at a post with a boring title like "Temperature Rise".

Temperature rise may be boring but it's important to the life of a motor. In fact, it's critical to the lifetime of nearly everything electrical and electronic. Insulation weakens with age. It ages faster at higher temperatures. The hotter a motor runs the shorter it's life. The same for an extension cord.

When operated at it's temperature rating the motor insulation will have lost no more than 50% of it's mechanical strength in 20,000 hours. There's a rule of thumb that a change of 10C in operating temperature results in a factor of two change in aging rate. That is, 10C hotter means half the lifetime.

A piece of wire has it's temperature rating marked on the wire along with it's diameter (gauge). Motors with a NEMA nameplate label their temperature rating as an insulation class with a letter designation. For example, a Class B motor is allowed 80C rise with a 40C ambient (40C is the default ambient temperature for ratings). Class F is 105C rise and Class A 60C. Nearly all US made new motors are Class F. Many of the Asian motors are Class A or B. When you see "HEAT 60C" on a Taiwan made motor that's Class A.

Heat in the motor is mostly produce by current flowing through a resistance. That heat varies with the square of the current value. It's produced in both the stator windings and rotor bars. It's the temperature of the Copper stator windings that's important here.

Let me repeat, it's the temperature of the windings not the outside temperature of the motor that counts. The winding temperature can be much higher than the casing, especially on a TEFC (totally enclosed fan cooled) motor.

The usual way of measuring the winding temperature is the Resistance Method from the NEMA MG 1 spec. It's not difficult but you can't do it with a common Ohmmeter and it can involve some personal risk. For that reason I am not going to describe the method.

Results

The Harbor Freight has no nameplate. The full load current and insulation class are unknown, so I started at a low current value and worked my way up. Each test takes an hour to reach a stable temperature and several to cool down. When I got to the third value of 15.8A the temperature rise was 85C so I stopped. Not knowing the temperature rating of the motor I really didn't want to run it for an hour above that.

The curve fit is simple square law. Above 16A it's an extrapolation. At some point above that, I don't know where, the curve will get steeper because the resistance of the windings increases with temperature. The motor might actually go into thermal runaway and the curve turn into a hockey stick. Above 16A the curve should be seen as a minimum, temperature rise will be at least what's shown.

Clagwell, you've earned my respect and admiration for conducting this analysis and providing everyone with a cogent summary. Alas, I'm an aging English major. Information like yours only increases my math anxiety and will cause nightmares involving issues with junior high school algebra.

So, pardon me for not fully understanding. I'll just follow two rules:

1. When you buy cheap, you often end up buying twice.

2. When it gets hot, shut if off.

 

[clagwell]

Performance Curve Comparisons

Part 1 of this blog has a chart showing a comparison of the original impeller versus the Rikon. Here's a more detailed view of the performance of the upgraded unit. This is only an impeller and inlet size change, it's otherwise completely stock, including the bag filter. Future posts will look at other modifications.

Pressure is corrected to a standard air density of .075 pound per cubic foot to allow comparison with other published data.

I've added actual data points to the graph. These represent three separate runs each on a different day over a two week period. The solid line is a simple square law curve fit to those points.

----

I purchased the dust collector in 2010. At the same time I bought a pleated filter from Wynn Environmental and never installed the bag. After filling the filter almost immediately I realized I needed a way to get better separation as well as keep the dust in the bag after separation. I added a Thein type baffle to the bottom of the separator ring which helps with both of those issues. It also reduces flow. I wanted to know how much reduction I was getting so I made some measurements. I wasn't really trying to compare with the rest of the world, just with and without the baffle.

I later came across an article from the March 2008 issue of Wood Magazine. Given my somewhat crude DIY test instruments at the time I was surprised to see reasonable agreement between my data and the magazine article. As it turns out that was actually more luck than skill but, oh well.

As in the previous chart the X marks are my data. The solid line however is not a curve fit to my data but rather a copy of the plot from the magazine article, and that's what I'm going to use in the rest of this blog.

----

So here's the net result of the fan transplant:

Note that the original curve is for a 5" fan inlet. The Rikon mod uses 6" inlet.

----

That Wood Magazine article included curves for a number of single stage dust collectors. The next chart is an overlay of the HF/Rikon curve on the published chart of various 6" inlet units. The HF/Rikon curve is the dashed line.

The HF/Rikon curve has good pressure at low CFM but falls of rapidly at the high end. I suspect that is due to the 5" outlet on the fan housing. I think the other ones all have 6" outlets. I'll show some more data on this issue in a future post.

 

[Redoak49]

Performance Curve Comparisons

Part 1 of this blog has a chart showing a comparison of the original impeller versus the Rikon. Here's a more detailed view of the performance of the upgraded unit. This is only an impeller and inlet size change, it's otherwise completely stock, including the bag filter. Future posts will look at other modifications.

Pressure is corrected to a standard air density of .075 pound per cubic foot to allow comparison with other published data.

I've added actual data points to the graph. These represent three separate runs each on a different day over a two week period. The solid line is a simple square law curve fit to those points.

----

I purchased the dust collector in 2010. At the same time I bought a pleated filter from Wynn Environmental and never installed the bag. After filling the filter almost immediately I realized I needed a way to get better separation as well as keep the dust in the bag after separation. I added a Thein type baffle to the bottom of the separator ring which helps with both of those issues. It also reduces flow. I wanted to know how much reduction I was getting so I made some measurements. I wasn't really trying to compare with the rest of the world, just with and without the baffle.

I later came across an article from the March 2008 issue of Wood Magazine. Given my somewhat crude DIY test instruments at the time I was surprised to see reasonable agreement between my data and the magazine article. As it turns out that was actually more luck than skill but, oh well.

As in the previous chart the X marks are my data. The solid line however is not a curve fit to my data but rather a copy of the plot from the magazine article, and that's what I'm going to use in the rest of this blog.

----

So here's the net result of the fan transplant:

Note that the original curve is for a 5" fan inlet. The Rikon mod uses 6" inlet.

----

That Wood Magazine article included curves for a number of single stage dust collectors. The next chart is an overlay of the HF/Rikon curve on the published chart of various 6" inlet units. The HF/Rikon curve is the dashed line.

The HF/Rikon curve has good pressure at low CFM but falls of rapidly at the high end. I suspect that is due to the 5" outlet on the fan housing. I think the other ones all have 6" outlets. I'll show some more data on this issue in a future post.

Interesting data and one wonders on the return on investment both for dollars and time.

 

[clagwell]

De-Constructing the HFDC

This is the fun part where I remove parts of the dust collector and see how much the flow increases.

Baseline

The baseline configuration here is the stock DC modified with the Rikon impeller and a 6" inlet:

-

-

The performance curve was shown in the previous post but here it is again. Note that the CFM axis has been extended to 1200CFM. We'll need that later.

Remove the Filter Bag

Of course the first thing to do is remove the filter bag.

-

-

Internet wisdom says that the bag doesn't have enough area to get good flow and that's probably true for a well used one. This bag has only been used for testing, never with dusty air. It's porous enough that's there's almost no resistance to flow. The dashed Blue line is the performance curve without the bag. There's a difference but it's too small to show up on the graph.

Remove the Centrifugal Separator (aka "Bag Holder")

The next thing to go is the centrifugal separator (or "collector ring" or "bag holder"). I tried to keep the bend in the hose unchanged to isolate it's effect.

-

-

This makes a big difference. The graph shows 2.5" loss at 600CFM caused by the separator. It's definitely something that has to go if a pre-separater like a Thein baffle or cyclone is used. Using it to simply hold a filter is a waste of precious static pressure.

That 2.5" at 600CFM is probably enough to make up for the loss caused by a cyclone.

How Bad is the Hose?

Let's take it off and find out!

-

-

That's another 1" of loss at 600CFM.

Fan Outlet to 5" Round Transition

The last thing to remove is the plastic transition from the rectangular fan outlet to the 5" hose.

-

-

This of course is the performance of the fan by itself. Operating it this way with no restrictions will quickly smoke the motor. I had to be careful testing at these flow levels. I scrambled the order of the orifices and allowed for plenty of cooling time.

The curve extrapolates to a free delivery rating of 1400CFM.

Flow rates above about 600CFM will cause reduced motor life. Above 800CFM can cause air quality issues (smoke).

 

[EarlS]

De-Constructing the HFDC

This is the fun part where I remove parts of the dust collector and see how much the flow increases.

Baseline

The baseline configuration here is the stock DC modified with the Rikon impeller and a 6" inlet:

-

-

The performance curve was shown in the previous post but here it is again. Note that the CFM axis has been extended to 1200CFM. We'll need that later.

Remove the Filter Bag

Of course the first thing to do is remove the filter bag.

-

-

Internet wisdom says that the bag doesn't have enough area to get good flow and that's probably true for a well used one. This bag has only been used for testing, never with dusty air. It's porous enough that's there's almost no resistance to flow. The dashed Blue line is the performance curve without the bag. There's a difference but it's too small to show up on the graph.

Remove the Centrifugal Separator (aka "Bag Holder")

The next thing to go is the centrifugal separator (or "collector ring" or "bag holder"). I tried to keep the bend in the hose unchanged to isolate it's effect.

-

-

This makes a big difference. The graph shows 2.5" loss at 600CFM caused by the separator. It's definitely something that has to go if a pre-separater like a Thein baffle or cyclone is used. Using it to simply hold a filter is a waste of precious static pressure.

That 2.5" at 600CFM is probably enough to make up for the loss caused by a cyclone.

How Bad is the Hose?

Let's take it off and find out!

-

-

That's another 1" of loss at 600CFM.

Fan Outlet to 5" Round Transition

The last thing to remove is the plastic transition from the rectangular fan outlet to the 5" hose.

-

-

This of course is the performance of the fan by itself. Operating it this way with no restrictions will quickly smoke the motor. I had to be careful testing at these flow levels. I scrambled the order of the orifices and allowed for plenty of cooling time.

The curve extrapolates to a free delivery rating of 1400CFM.

Flow rates above about 600CFM will cause reduced motor life. Above 800CFM can cause air quality issues (smoke).

No surprises from what I see. Just goes to show what the impact of the various parts of the system have on the pressure/flow.

I'm surprised you didn't over amp the motor and trip the breaker.

 

[MacNut11]

De-Constructing the HFDC

This is the fun part where I remove parts of the dust collector and see how much the flow increases.

Baseline

The baseline configuration here is the stock DC modified with the Rikon impeller and a 6" inlet:

-

-

The performance curve was shown in the previous post but here it is again. Note that the CFM axis has been extended to 1200CFM. We'll need that later.

Remove the Filter Bag

Of course the first thing to do is remove the filter bag.

-

-

Internet wisdom says that the bag doesn't have enough area to get good flow and that's probably true for a well used one. This bag has only been used for testing, never with dusty air. It's porous enough that's there's almost no resistance to flow. The dashed Blue line is the performance curve without the bag. There's a difference but it's too small to show up on the graph.

Remove the Centrifugal Separator (aka "Bag Holder")

The next thing to go is the centrifugal separator (or "collector ring" or "bag holder"). I tried to keep the bend in the hose unchanged to isolate it's effect.

-

-

This makes a big difference. The graph shows 2.5" loss at 600CFM caused by the separator. It's definitely something that has to go if a pre-separater like a Thein baffle or cyclone is used. Using it to simply hold a filter is a waste of precious static pressure.

That 2.5" at 600CFM is probably enough to make up for the loss caused by a cyclone.

How Bad is the Hose?

Let's take it off and find out!

-

-

That's another 1" of loss at 600CFM.

Fan Outlet to 5" Round Transition

The last thing to remove is the plastic transition from the rectangular fan outlet to the 5" hose.

-

-

This of course is the performance of the fan by itself. Operating it this way with no restrictions will quickly smoke the motor. I had to be careful testing at these flow levels. I scrambled the order of the orifices and allowed for plenty of cooling time.

The curve extrapolates to a free delivery rating of 1400CFM.

Flow rates above about 600CFM will cause reduced motor life. Above 800CFM can cause air quality issues (smoke).

I just scrolled through all 5 parts of this topic and I don't see a photo of the actual impellers. Did or could you take photos of them?

 

[Brawler]

De-Constructing the HFDC

This is the fun part where I remove parts of the dust collector and see how much the flow increases.

Baseline

The baseline configuration here is the stock DC modified with the Rikon impeller and a 6" inlet:

-

-

The performance curve was shown in the previous post but here it is again. Note that the CFM axis has been extended to 1200CFM. We'll need that later.

Remove the Filter Bag

Of course the first thing to do is remove the filter bag.

-

-

Internet wisdom says that the bag doesn't have enough area to get good flow and that's probably true for a well used one. This bag has only been used for testing, never with dusty air. It's porous enough that's there's almost no resistance to flow. The dashed Blue line is the performance curve without the bag. There's a difference but it's too small to show up on the graph.

Remove the Centrifugal Separator (aka "Bag Holder")

The next thing to go is the centrifugal separator (or "collector ring" or "bag holder"). I tried to keep the bend in the hose unchanged to isolate it's effect.

-

-

This makes a big difference. The graph shows 2.5" loss at 600CFM caused by the separator. It's definitely something that has to go if a pre-separater like a Thein baffle or cyclone is used. Using it to simply hold a filter is a waste of precious static pressure.

That 2.5" at 600CFM is probably enough to make up for the loss caused by a cyclone.

How Bad is the Hose?

Let's take it off and find out!

-

-

That's another 1" of loss at 600CFM.

Fan Outlet to 5" Round Transition

The last thing to remove is the plastic transition from the rectangular fan outlet to the 5" hose.

-

-

This of course is the performance of the fan by itself. Operating it this way with no restrictions will quickly smoke the motor. I had to be careful testing at these flow levels. I scrambled the order of the orifices and allowed for plenty of cooling time.

The curve extrapolates to a free delivery rating of 1400CFM.

Flow rates above about 600CFM will cause reduced motor life. Above 800CFM can cause air quality issues (smoke).

Why would not having the hose on the output "reduce motor life"? I read somewhere (honestly I think it was on LJ) that the lack of back pressure is easier on the motor because the impeller is spinning in vacuum and doesn't have to push against that backpressure. I'm no expert, just got conflicting info. I work with a whole building full of engineers I think I need to ivestigate this. However thanks for the graphs, interesting measurements.

 

[clagwell]

De-Constructing the HFDC

This is the fun part where I remove parts of the dust collector and see how much the flow increases.

Baseline

The baseline configuration here is the stock DC modified with the Rikon impeller and a 6" inlet:

-

-

The performance curve was shown in the previous post but here it is again. Note that the CFM axis has been extended to 1200CFM. We'll need that later.

Remove the Filter Bag

Of course the first thing to do is remove the filter bag.

-

-

Internet wisdom says that the bag doesn't have enough area to get good flow and that's probably true for a well used one. This bag has only been used for testing, never with dusty air. It's porous enough that's there's almost no resistance to flow. The dashed Blue line is the performance curve without the bag. There's a difference but it's too small to show up on the graph.

Remove the Centrifugal Separator (aka "Bag Holder")

The next thing to go is the centrifugal separator (or "collector ring" or "bag holder"). I tried to keep the bend in the hose unchanged to isolate it's effect.

-

-

This makes a big difference. The graph shows 2.5" loss at 600CFM caused by the separator. It's definitely something that has to go if a pre-separater like a Thein baffle or cyclone is used. Using it to simply hold a filter is a waste of precious static pressure.

That 2.5" at 600CFM is probably enough to make up for the loss caused by a cyclone.

How Bad is the Hose?

Let's take it off and find out!

-

-

That's another 1" of loss at 600CFM.

Fan Outlet to 5" Round Transition

The last thing to remove is the plastic transition from the rectangular fan outlet to the 5" hose.

-

-

This of course is the performance of the fan by itself. Operating it this way with no restrictions will quickly smoke the motor. I had to be careful testing at these flow levels. I scrambled the order of the orifices and allowed for plenty of cooling time.

The curve extrapolates to a free delivery rating of 1400CFM.

Flow rates above about 600CFM will cause reduced motor life. Above 800CFM can cause air quality issues (smoke).

No surprises from what I see. Just goes to show what the impact of the various parts of the system have on the pressure/flow.

I m surprised you didn t over amp the motor and trip the breaker.

- EarlS

The highest current I measured was 20.51A. I made the traverse in about 5 minutes. That's not enough I-squared-t to trip a 20A breaker.

I measured the thermal time constant of the motor at about 15 minutes. My temperature rise extrapolation (see Part 3 of this blog) for 20A is 140C. 5 minutes is about one third of a time constant so I would expect a rise of somewhat less than 50C in that time. So, definitely over amped but not yet smoke.

I just scrolled through all 5 parts of this topic and I don t see a photo of the actual impellers. Did or could you take photos of them?

- MacNut11

The original fan is 10" diameter with 3" high blades forward curved at roughly 15 degrees.

-

-

The Rikon replacement is 12" or so in diameter also with 3" high blades. They are backward curved. I can't get to them right now to check the angle but I don't remember it being much different from the HF (opposite direction of course.)

-

 

[clagwell]

De-Constructing the HFDC

This is the fun part where I remove parts of the dust collector and see how much the flow increases.

Baseline

The baseline configuration here is the stock DC modified with the Rikon impeller and a 6" inlet:

-

-

The performance curve was shown in the previous post but here it is again. Note that the CFM axis has been extended to 1200CFM. We'll need that later.

Remove the Filter Bag

Of course the first thing to do is remove the filter bag.

-

-

Internet wisdom says that the bag doesn't have enough area to get good flow and that's probably true for a well used one. This bag has only been used for testing, never with dusty air. It's porous enough that's there's almost no resistance to flow. The dashed Blue line is the performance curve without the bag. There's a difference but it's too small to show up on the graph.

Remove the Centrifugal Separator (aka "Bag Holder")

The next thing to go is the centrifugal separator (or "collector ring" or "bag holder"). I tried to keep the bend in the hose unchanged to isolate it's effect.

-

-

This makes a big difference. The graph shows 2.5" loss at 600CFM caused by the separator. It's definitely something that has to go if a pre-separater like a Thein baffle or cyclone is used. Using it to simply hold a filter is a waste of precious static pressure.

That 2.5" at 600CFM is probably enough to make up for the loss caused by a cyclone.

How Bad is the Hose?

Let's take it off and find out!

-

-

That's another 1" of loss at 600CFM.

Fan Outlet to 5" Round Transition

The last thing to remove is the plastic transition from the rectangular fan outlet to the 5" hose.

-

-

This of course is the performance of the fan by itself. Operating it this way with no restrictions will quickly smoke the motor. I had to be careful testing at these flow levels. I scrambled the order of the orifices and allowed for plenty of cooling time.

The curve extrapolates to a free delivery rating of 1400CFM.

Flow rates above about 600CFM will cause reduced motor life. Above 800CFM can cause air quality issues (smoke).

Why would not having the hose on the output "reduce motor life"? I read somewhere (honestly I think it was on LJ) that the lack of back pressure is easier on the motor because the impeller is spinning in vacuum and doesn t have to push against that backpressure. I m no expert, just got conflicting info. I work with a whole building full of engineers I think I need to ivestigate this. However thanks for the graphs, interesting measurements.

- Brawler

Ok, first of all, there's nothing remotely close to a vacuum in a dust collector. A vacuum is about 400" of water. The most negative pressure here is 10". Nowhere close to a vacuum.

The horsepower requirements of a fan increase with increasing flow rates. Minimum power at zero flow and maximum power at maximum flow.

There's a chart in Part 1 showing motor current as a function of flow rate. Higher current means a hotter motor (see Part 3). The higher the motor temperature the shorter it's life.

 

[Redoak49]

De-Constructing the HFDC

This is the fun part where I remove parts of the dust collector and see how much the flow increases.

Baseline

The baseline configuration here is the stock DC modified with the Rikon impeller and a 6" inlet:

-

-

The performance curve was shown in the previous post but here it is again. Note that the CFM axis has been extended to 1200CFM. We'll need that later.

Remove the Filter Bag

Of course the first thing to do is remove the filter bag.

-

-

Internet wisdom says that the bag doesn't have enough area to get good flow and that's probably true for a well used one. This bag has only been used for testing, never with dusty air. It's porous enough that's there's almost no resistance to flow. The dashed Blue line is the performance curve without the bag. There's a difference but it's too small to show up on the graph.

Remove the Centrifugal Separator (aka "Bag Holder")

The next thing to go is the centrifugal separator (or "collector ring" or "bag holder"). I tried to keep the bend in the hose unchanged to isolate it's effect.

-

-

This makes a big difference. The graph shows 2.5" loss at 600CFM caused by the separator. It's definitely something that has to go if a pre-separater like a Thein baffle or cyclone is used. Using it to simply hold a filter is a waste of precious static pressure.

That 2.5" at 600CFM is probably enough to make up for the loss caused by a cyclone.

How Bad is the Hose?

Let's take it off and find out!

-

-

That's another 1" of loss at 600CFM.

Fan Outlet to 5" Round Transition

The last thing to remove is the plastic transition from the rectangular fan outlet to the 5" hose.

-

-

This of course is the performance of the fan by itself. Operating it this way with no restrictions will quickly smoke the motor. I had to be careful testing at these flow levels. I scrambled the order of the orifices and allowed for plenty of cooling time.

The curve extrapolates to a free delivery rating of 1400CFM.

Flow rates above about 600CFM will cause reduced motor life. Above 800CFM can cause air quality issues (smoke).

The logical question is what is the max cfm that you can run without damaging the motor.

AND the big question is given the max cfm that you can run is the upgrade worthwhile?

A great job with this and should answer many question about the true cfm of the HF dust collector.

 

[clagwell]

De-Constructing the HFDC

This is the fun part where I remove parts of the dust collector and see how much the flow increases.

Baseline

The baseline configuration here is the stock DC modified with the Rikon impeller and a 6" inlet:

-

-

The performance curve was shown in the previous post but here it is again. Note that the CFM axis has been extended to 1200CFM. We'll need that later.

Remove the Filter Bag

Of course the first thing to do is remove the filter bag.

-

-

Internet wisdom says that the bag doesn't have enough area to get good flow and that's probably true for a well used one. This bag has only been used for testing, never with dusty air. It's porous enough that's there's almost no resistance to flow. The dashed Blue line is the performance curve without the bag. There's a difference but it's too small to show up on the graph.

Remove the Centrifugal Separator (aka "Bag Holder")

The next thing to go is the centrifugal separator (or "collector ring" or "bag holder"). I tried to keep the bend in the hose unchanged to isolate it's effect.

-

-

This makes a big difference. The graph shows 2.5" loss at 600CFM caused by the separator. It's definitely something that has to go if a pre-separater like a Thein baffle or cyclone is used. Using it to simply hold a filter is a waste of precious static pressure.

That 2.5" at 600CFM is probably enough to make up for the loss caused by a cyclone.

How Bad is the Hose?

Let's take it off and find out!

-

-

That's another 1" of loss at 600CFM.

Fan Outlet to 5" Round Transition

The last thing to remove is the plastic transition from the rectangular fan outlet to the 5" hose.

-

-

This of course is the performance of the fan by itself. Operating it this way with no restrictions will quickly smoke the motor. I had to be careful testing at these flow levels. I scrambled the order of the orifices and allowed for plenty of cooling time.

The curve extrapolates to a free delivery rating of 1400CFM.

Flow rates above about 600CFM will cause reduced motor life. Above 800CFM can cause air quality issues (smoke).

The logical question is what is the max cfm that you can run without damaging the motor.

AND the big question is given the max cfm that you can run is the upgrade worthwhile?

A great job with this and should answer many question about the true cfm of the HF dust collector.

- Redoak49

Thanks Redoak49. I hadn't been able to find performance data on the "Rikon Upgrade" so I thought I'd share what I've learned.

I'll have a lot more to say about the motor and the value of the upgrade in a future post.

 

[dbhost]

De-Constructing the HFDC

This is the fun part where I remove parts of the dust collector and see how much the flow increases.

Baseline

The baseline configuration here is the stock DC modified with the Rikon impeller and a 6" inlet:

-

-

The performance curve was shown in the previous post but here it is again. Note that the CFM axis has been extended to 1200CFM. We'll need that later.

Remove the Filter Bag

Of course the first thing to do is remove the filter bag.

-

-

Internet wisdom says that the bag doesn't have enough area to get good flow and that's probably true for a well used one. This bag has only been used for testing, never with dusty air. It's porous enough that's there's almost no resistance to flow. The dashed Blue line is the performance curve without the bag. There's a difference but it's too small to show up on the graph.

Remove the Centrifugal Separator (aka "Bag Holder")

The next thing to go is the centrifugal separator (or "collector ring" or "bag holder"). I tried to keep the bend in the hose unchanged to isolate it's effect.

-

-

This makes a big difference. The graph shows 2.5" loss at 600CFM caused by the separator. It's definitely something that has to go if a pre-separater like a Thein baffle or cyclone is used. Using it to simply hold a filter is a waste of precious static pressure.

That 2.5" at 600CFM is probably enough to make up for the loss caused by a cyclone.

How Bad is the Hose?

Let's take it off and find out!

-

-

That's another 1" of loss at 600CFM.

Fan Outlet to 5" Round Transition

The last thing to remove is the plastic transition from the rectangular fan outlet to the 5" hose.

-

-

This of course is the performance of the fan by itself. Operating it this way with no restrictions will quickly smoke the motor. I had to be careful testing at these flow levels. I scrambled the order of the orifices and allowed for plenty of cooling time.

The curve extrapolates to a free delivery rating of 1400CFM.

Flow rates above about 600CFM will cause reduced motor life. Above 800CFM can cause air quality issues (smoke).

Great info! However last time I looked at how to read a tape measure, the OE HF impeller is 9.75"

 

[Madebywilly]

De-Constructing the HFDC

This is the fun part where I remove parts of the dust collector and see how much the flow increases.

Baseline

The baseline configuration here is the stock DC modified with the Rikon impeller and a 6" inlet:

-

-

The performance curve was shown in the previous post but here it is again. Note that the CFM axis has been extended to 1200CFM. We'll need that later.

Remove the Filter Bag

Of course the first thing to do is remove the filter bag.

-

-

Internet wisdom says that the bag doesn't have enough area to get good flow and that's probably true for a well used one. This bag has only been used for testing, never with dusty air. It's porous enough that's there's almost no resistance to flow. The dashed Blue line is the performance curve without the bag. There's a difference but it's too small to show up on the graph.

Remove the Centrifugal Separator (aka "Bag Holder")

The next thing to go is the centrifugal separator (or "collector ring" or "bag holder"). I tried to keep the bend in the hose unchanged to isolate it's effect.

-

-

This makes a big difference. The graph shows 2.5" loss at 600CFM caused by the separator. It's definitely something that has to go if a pre-separater like a Thein baffle or cyclone is used. Using it to simply hold a filter is a waste of precious static pressure.

That 2.5" at 600CFM is probably enough to make up for the loss caused by a cyclone.

How Bad is the Hose?

Let's take it off and find out!

-

-

That's another 1" of loss at 600CFM.

Fan Outlet to 5" Round Transition

The last thing to remove is the plastic transition from the rectangular fan outlet to the 5" hose.

-

-

This of course is the performance of the fan by itself. Operating it this way with no restrictions will quickly smoke the motor. I had to be careful testing at these flow levels. I scrambled the order of the orifices and allowed for plenty of cooling time.

The curve extrapolates to a free delivery rating of 1400CFM.

Flow rates above about 600CFM will cause reduced motor life. Above 800CFM can cause air quality issues (smoke).

Does anyone have their original impeller that they're willing to part with? I'm located in Wake Forest, NC. Thanks in advance.

 

[papley]

I’m curious if there is any insight on the effect of the reversed blade direction of the new impeller.