Wobble Plate Air Compressor

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Toymaker

Well-Known Member
HMEM Supporting Member
Joined
Jan 26, 2022
Messages
483
Reaction score
241
Location
Thailand
This air compressor will be used in my Ambitious ORC Turbine project to supply 6 psi nominal air pressure at 70 L/min to a siphon-type fuel nozzle. There are several variations of wobble plate (aka, Swashplate) designs but the one I chose has 5 double end pistons operating inside 10 cylinders and produces a very smooth, steady output airflow. I know I could have used a far simpler vane pump design, but I really wanted to see if I could build a swashplate air compressor.

The compressor is nearly finished, I'm still waiting for two online orders to arrive: a 3:1 reduction belt & pulley set, and a rotary shaft seal, and I will add 10 little air filters, 5 on each end plate, cut from common dishwashing sponges.

Using my lathe to spin the compressor's drive shaft, I was able to chart air pressure outputs at different RPMs, with the fuel nozzle attached to the compressor.

1653962447325.png


A few specs: 85mm diameter x 120mm length. Drive shaft is 8mm diam.
Displacement: 6.86 cubic inches (112.4 cc) per revolution.
Pistons and "block" are 6061 aluminum.
Reed valve plates are 0.008" stainless.

The small screw in the top is used to partially fill the center section with oil to keep both swashplate and pistons lubricated.
1653962506990.png
1653962544099.png


The 5 large-ish round holes in both end caps will be used to hold small disc-shaped air filters.
1653962619973.png


End cap removed shows piston ends.
1653962656730.png


1653962726186.png

The smallest parts shown are made by brazing 6mm ball bearings onto a small flat washer. The washer slides flat along the swash plate while the ball bearing allows for constantly changing angles as the swash plate rotates. The bearing fits into a semi spherical machined surface in each piston and pushes against the centerline of the piston.
1653962770968.png
This pic was taken before I added the O-rings.

End cap and reed valve plates.
1653962823488.png
 
Last edited:
Cool. Some guys modify automotive air conditioner compressors for offroad use that uses the same kind of wobble plate.
 
Cool. Some guys modify automotive air conditioner compressors for offroad use that uses the same kind of wobble plate.

Thanks for the YouTube link Noitoen.
Yes, with a few modifications to the lubrication system, automotive air conditioner compressors make very nice air compressors; they can easily hit 200+ psi. Unfortunately, I couldn't find one that was small enough to meet my needs. The smallest commercial unit I was able to find is the Denso 10S15C (shown in pic below), which requires a fairly large, heavy electric motor to operate. The little compressor I built is nearly half the size of the 10S15C and will require a much smaller and lighter electric motor to power it.
1654000261867.png
 
Some pressure washers use a wobble plate type setup . I'd never seen one before I took one apart .
animal
 
Hi
Impressive.
You have the machine tools to make a Guided Rotor Compressor (GRC). These are used full size for high reliability gas compressors. The geometry is reasonably complex, too complex to cut by hand, but mechanically very simple and highly reliable. The should scale down to any size.
 
Hi
Impressive.
You have the machine tools to make a Guided Rotor Compressor (GRC). These are used full size for high reliability gas compressors. The geometry is reasonably complex, too complex to cut by hand, but mechanically very simple and highly reliable. The should scale down to any size.

The GRC is indeed a very clever design. I wonder why you don't see them used as air conditioner compressors? Used as an air compressor, seems like the many sliding surfaces would be a challenge to keep lubricated? Do you know how are they kept lubricated in the gas compressor industry?
 
The GRC is indeed a very clever design. I wonder why you don't see them used as air conditioner compressors? Used as an air compressor, seems like the many sliding surfaces would be a challenge to keep lubricated? Do you know how are they kept lubricated in the gas compressor industry?
Hi
This style is new enough that commercial use is limited by patents.
The rollers are the only contact points, so roller material determines lubrication requirements. Typically the rollers are some sort of plastic.
 
Nice build!
What was the compressor working against when you tested it on the lathe?
When I tested the centrifugal compressor I made for my radial engine, I restricted the output and measured airspeed using a cheap wind speed meter. Multiplying speed by the exit area gives flow rate, so I have a "map" of pressures and flows at a few speeds.
Pete.
 
Nice build!
What was the compressor working against when you tested it on the lathe?
When I tested the centrifugal compressor I made for my radial engine, I restricted the output and measured airspeed using a cheap wind speed meter. Multiplying speed by the exit area gives flow rate, so I have a "map" of pressures and flows at a few speeds.
Pete.

During initial testing on my lathe the compressor output line is connected to the actual siphon-type fuel nozzle I'm using in my burner. The fuel nozzle (below) was purchased from China thru eBay and the only air related specification given is pressure. I derived the need for 70 L/min from my previous use of a commercial fish-tank air pump which used a single diaphragm and is spec'd at 50 L/min at 6 psi. That pump could barely keep up and the output air flow "pulsed" with each pressure stroke from the diaphragm, which transmitted a noticeable, and troublesome, pulse into the burner.

I've only tested my little wobble plate compressor up to 9 psi at 888 rpm.

Total Length: 67 mm
Nozzle Bore Diameter: 2.0mm
Working Pressure: 4 ~ 6 PSI
Fuel: 14 liters/hour

1654128285766.png
 
Ah, I see!
I had read that it was a compressor for a turbine and I had thought gas turbine, not steam...
Presumably your burner operates in near atmospheric firebox pressure.

I'm using a small cordless leaf blower to add additional air into the burner (stainless steel cans in the pic) but I doubt the blower adds any significant pressure to the burner,...maybe one psi if I'm very lucky. So, yes, pretty much atmospheric pressure inside the burner.
1654166072444.png
 
... That pump could barely keep up and the output air flow "pulsed" with each pressure stroke from the diaphragm, which transmitted a noticeable, and troublesome, pulse into the burner.

There is a theory that pulsed fuel flow creates balls of fuel that increase turbulence and the flame front surface area. Without evidence, I don't think it would be a good approach. I note that real gas turbines and steam boilers use constant fuel flow to maintain a steady flame.

Pulsed fuel flow may help to increase the fuel flow throttle range. Pulsing the fuel flow would allow a high fuel pressure to maintain a good sprayer pattern while reducing average fuel flow. I think it would increase the risk of flame-out and subsequent re-ignition from the hot parts. Something worth avoiding.
 
At least with my burner assembly, the pulsing limited the low end burn rate, and had no real impact on the high burn rate. I didn't understand why, when using the diaphragm air pump, I couldn't adjust the burner to as low a setting as I could when I was using my shop air compressor as the air source,... then I reviewed the video I had taken, frame-by-frame; in one frame the flame front was large and burning brightly, while in the next frame the flame front would nearly be gone,...next few frames the flame front would expand greatly, and then suddenly be nearly gone in the subsequent frame. That pulsing pattern repeated with the pulsing air flow from the diaphragm compressor and was completely absent when using my shop air compressor, which was delivering a steady flow of air from the tank.

Designing a burner like the one you describe that achieves better average fuel burn via pulsing flows, is an intriguing idea, but is well beyond my abilities.
 
Last edited:
At least with my burner assembly, the pulsing limited the low end burn rate, and had no real impact on the high burn rate. I didn't understand why, when using the diaphragm air pump, I couldn't adjust the burner to as low a setting as I could when I was using my shop air compressor as the air source,... then I reviewed the video I had taken, frame-by-frame; in one frame the flame front was large and burning brightly, while in the next frame the flame front would nearly be gone,...next few frames the flame front would expand greatly, and then suddenly be nearly gone in the subsequent frame. That pulsing pattern repeated with the pulsing air flow from the diaphragm compressor and was completely absent when using my shop air compressor, which was delivering a steady flow of air from the tank.

Designing a burner like the one you describe that achieves better average fuel burn via pulsing flows, is an intriguing idea, but is well beyond my abilities.
I have had some experience in utilizing atomizing burners with steam and air. The air compressors for these systems are specially designed for an even air flow. They are also expensive. You can take out some of the pulse by using a storage tank the bigger the better. You can also add a regulator to even out the line pressure but it has to be sized correctly. The burners will have a turn down ratio which is difficult to go below. Has to do with nozzle velocities and is specific to its heat output. Below 1/3 its tricky but 1/4 is doable with careful attention to flame velocity. Lower turn downs require a bit of special engineering.

The flame must be off the nozzle tip. I have seen some home made stuff that works but also makes me cringe. Its possible if the gas velocities are not high enough the flame velocity will actually back down the system into the fuel supply with some unexpected results. I have experienced that. It is for that very reason acetylene torches have flame blockers installed near the tanks.

The fuel will be varied for different loads. The nozzle will distributed the fuel but does not supply the combustion air. The reason you want even flow and not pulsating is because the controlling the fuel air ratio would be exceptionally difficult for open flame systems like boilers.
Sometimes its the simple things that are not so simple.

Take Care
HMEL
 
I have had some experience in utilizing atomizing burners with steam and air. The air compressors for these systems are specially designed for an even air flow. They are also expensive. You can take out some of the pulse by using a storage tank the bigger the better. You can also add a regulator to even out the line pressure but it has to be sized correctly. The burners will have a turn down ratio which is difficult to go below. Has to do with nozzle velocities and is specific to its heat output. Below 1/3 its tricky but 1/4 is doable with careful attention to flame velocity. Lower turn downs require a bit of special engineering.

The flame must be off the nozzle tip. I have seen some home made stuff that works but also makes me cringe. Its possible if the gas velocities are not high enough the flame velocity will actually back down the system into the fuel supply with some unexpected results. I have experienced that. It is for that very reason acetylene torches have flame blockers installed near the tanks.

The fuel will be varied for different loads. The nozzle will distributed the fuel but does not supply the combustion air. The reason you want even flow and not pulsating is because the controlling the fuel air ratio would be exceptionally difficult for open flame systems like boilers.
Sometimes its the simple things that are not so simple.

Take Care
HMEL


Thanks for the inputs HMEL.

The main reason I chose this particular swashplate compressor design was because it's 10 cylinders pump air sequentially & produce very smooth output air flow.
It will be interesting to see if I can use the compressor alone or if I will need to add a storage tank and/or a pressure regulator to further smooth out the air flow.

Due to the compact size of my burner (12cm diameter) and the use of concentric burner cans, viewing the nozzle is not an easy task; I would need to drill viewing holes into both concentric burner cans to have a good viewing angle. At present, this is the only view I have,...looking directly into the burner at it's lowest setting. From this angle, I cannot tell how close the flames are from the nozzle.


1654489519687.png
 
This air compressor will be used in my Ambitious ORC Turbine project to supply 6 psi nominal air pressure at 70 L/min to a siphon-type fuel nozzle. There are several variations of wobble plate (aka, Swashplate) designs but the one I chose has 5 double end pistons operating inside 10 cylinders and produces a very smooth, steady output airflow. I know I could have used a far simpler vane pump design, but I really wanted to see if I could build a swashplate air compressor.

The compressor is nearly finished, I'm still waiting for two online orders to arrive: a 3:1 reduction belt & pulley set, and a rotary shaft seal, and I will add 10 little air filters, 5 on each end plate, cut from common dishwashing sponges.

Using my lathe to spin the compressor's drive shaft, I was able to chart air pressure outputs at different RPMs, with the fuel nozzle attached to the compressor.

View attachment 136648

A few specs: 85mm diameter x 120mm length. Drive shaft is 8mm diam.
Displacement: 6.86 cubic inches (112.4 cc) per revolution.
Pistons and "block" are 6061 aluminum.
Reed valve plates are 0.008" stainless.

The small screw in the top is used to partially fill the center section with oil to keep both swashplate and pistons lubricated.
View attachment 136649 View attachment 136650

The 5 large-ish round holes in both end caps will be used to hold small disc-shaped air filters.
View attachment 136651

End cap removed shows piston ends.
View attachment 136652

View attachment 136653
The smallest parts shown are made by brazing 6mm ball bearings onto a small flat washer. The washer slides flat along the swash plate while the ball bearing allows for constantly changing angles as the swash plate rotates. The bearing fits into a semi spherical machined surface in each piston and pushes against the centerline of the piston.
View attachment 136654 This pic was taken before I added the O-rings.

End cap and reed valve plates.
View attachment 136655
very nice work and engineering. Ther is bulk gear belt material available and you cn slice ortaer the material gluing with superglue. There are pre made belts too very similar gm ac compressors I as of old.
Byron
 
very nice work and engineering. Ther is bulk gear belt material available and you cn slice ortaer the material gluing with superglue. There are pre made belts too very similar gm ac compressors I as of old.
Byron

Thanks for the compliments Bentwings, and for the belt material suggestions.
Living in Thailand has both advantages and disadvantages; I love the tropical climate and most everything here is much less expensive then in the US,....BUT, sourcing parts locally, like finding a selection of cogged belts and pulleys, is pretty much impossible.
 
Thanks for the compliments Bentwings, and for the belt material suggestions.
Living in Thailand has both advantages and disadvantages; I love the tropical climate and most everything here is much less expensive then in the US,....BUT, sourcing parts locally, like finding a selection of cogged belts and pulleys, is pretty much impossible.
apple and the grievous spell check edited after I Ostend so my note got garbled. The binding process is called skiving or slicing at an angle so belt ends smoothly over lap. I have a couple belt and pulley kits hat make 3:1,ratio I’ll do a double reduction but I won’t have super high belt sped a input speed is relatively low. The GM AC compressors had a face bearing seal that I think is hat you are using we had a special tool to replace it s it s a common fail pint . I’ll be interested in how yours performs. I hav. California ultra quit air compressor that you can hardly hear. If I set it on some medium rubber pad it would be whisper quiet. If fills the tank very quickly . I haven’t hooked up a vacuum source as I really don’t need that now. Its I really hot today . 96 partly cloudy 100 tomorrow and Tuesday sunny . 46% rh so humid too. The news said this is a100yr cycle of weather . No climate change involved just a stern.
 

Latest posts

Back
Top