If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#1
|
|||
|
|||
Epoxy Bonding to Aluminum and Magnesium
I'm working on a Hummel Aviation Ultracruiser Plus. The way the
engine mounts is quite different from the mounts on most VW powered homebuilts. The Conventional Wisdom bolts the thing to the firewall using the clucth end bellhousing, often with an accessory case in between. The Ultracruiser Plus is different. Two aluminum angles are bolted to the sides of the magnesium case, in the sump area, and, suitablely reinforced, are used to bolt Berry mounts to "bed" type engine bearers built up out of aluminum, that extend from the forward fuselage. These angles are bolted and epoxied to the side of the magnesium case. My point: what good is the epoxy? It's probably not carrying any loads. A basic tenet of structural design is that the stiffest load path carries the load, and the bolts through the angle and into the case (secured with nuts and washers inside the sump, before the engine is assembled) seem a bunch stiffer that the epoxy. It's not a sure stop against leaks. Epoxy is a wonderful material, but it doesn't bond particularly well or reliably to metals. Plus, it's mechanical properties, from it's modulus of expansion, to it's ductility, are far different that the aluminum, steel, and magnesium sandwich is it the Mayonnaise of. My concern is the epoxy will eventually crack. I lost the reply from Scott Casler of Hummel Engines, I'll paraphrase: "The epoxy is to keep the angles from working and hogging out the holes. The epoxy I use is a real good sealer, you've got to grind it off." My thoughts are this: LAP the angles to the side of the case (instead of sanding with 80 grit), but use Permatex or Curil T to seal things. Use close tolerance bolts in reamed holes in the side of the case and the appropriate Loctite product to seal the bolts. And I'm inclined to put the bolt heads INSIDE the engine. Comments? To see what this installation looks like: http://flyhummel.com/forums/album_pic.php?pic_id=170 Ultracruiser (with 1/2 VW) is the same deal |
#2
|
|||
|
|||
To be honest this all sounds like a bit of a mess.
There are a number of issues here ... however it is difficult to comment in any detail until I know what sort of epoxy is specified and without further details of the surface preparation and details of the design. 1. Most common epoxies have a Glass Transaition Temperature (Tg) of approx 90 deg C. If the epoxy is taken above that temperature then two things happen ... first it softens and the stiffness and strength reduce dramatically. Second irreveraible damage is done to the epoxy and it will never be the same again - even when cooled. Unless you are using one of a small number of epoxies that are designed for high temperature operation (some of these have Tg of approx 400 F, 200 C I think without reaching for a calculator) then it is certain I think that if used on an engine the 90 deg C limit will be exceeded. 2. Next there is the issue of surface preparation. If the plans say prepare by running with 80 grid paper then it is fairly clear that the guy who wrote the plans knows little about what he is doing ! Epoxy metal bonding is reliable if the surfaces are prepared properly but from what you have said I doubt that is the case. 3. Bolted and bonded joints should be avoided because it is difficult to predict the load transfer etc ... I won't go into detail on this - perhaps later. 4. What are the differences in the thermal expansion coefficients of the aluminium and epoxy ... from MIL-HDBK-5J alumnium is approx 12.5e-6 in/in/F and Magnesium is 14.0e-6 ... not a lot of difference perhaps this is not an issue. 5. Galvanic corrosion .. magnesium is at -1.6 V and alumnium at -0.75 V on the galvanic table. That is a big different. You definitely need something to separate the two or the magnesium is going to get gobbled up !! 6. I think galling of the metal is a possibility but if appropriate tolerances are used for the bolts and holes than this would be less of a problem. Hard to say without seeing the drawings etc. Sounds to me like if the holes are flogging out then the design has some fundamental problems and that one should not be relying on epoxy that probably cannot withstand the temperatures to fix it. My gut feel is that you need something between the alumnium and the magnesium for corrosion protection and possibly the help the galling issue. I would assume that this is sufficiently ductile an rubbery that it will not pick up any load and that the fasteners will transfer all the load. I would use a rudder like sealant compound that can take the temperature ... not epoxy. When you use sealant of this type in a joint with fasteners extra largers of safety should be allowed because of the extra flexibility that this produces in the joint (typically an extra factor of 1.5). "Ryan Young" wrote in message om... I'm working on a Hummel Aviation Ultracruiser Plus. The way the engine mounts is quite different from the mounts on most VW powered homebuilts. The Conventional Wisdom bolts the thing to the firewall using the clucth end bellhousing, often with an accessory case in between. The Ultracruiser Plus is different. Two aluminum angles are bolted to the sides of the magnesium case, in the sump area, and, suitablely reinforced, are used to bolt Berry mounts to "bed" type engine bearers built up out of aluminum, that extend from the forward fuselage. These angles are bolted and epoxied to the side of the magnesium case. My point: what good is the epoxy? It's probably not carrying any loads. A basic tenet of structural design is that the stiffest load path carries the load, and the bolts through the angle and into the case (secured with nuts and washers inside the sump, before the engine is assembled) seem a bunch stiffer that the epoxy. It's not a sure stop against leaks. Epoxy is a wonderful material, but it doesn't bond particularly well or reliably to metals. Plus, it's mechanical properties, from it's modulus of expansion, to it's ductility, are far different that the aluminum, steel, and magnesium sandwich is it the Mayonnaise of. My concern is the epoxy will eventually crack. I lost the reply from Scott Casler of Hummel Engines, I'll paraphrase: "The epoxy is to keep the angles from working and hogging out the holes. The epoxy I use is a real good sealer, you've got to grind it off." My thoughts are this: LAP the angles to the side of the case (instead of sanding with 80 grit), but use Permatex or Curil T to seal things. Use close tolerance bolts in reamed holes in the side of the case and the appropriate Loctite product to seal the bolts. And I'm inclined to put the bolt heads INSIDE the engine. Comments? To see what this installation looks like: http://flyhummel.com/forums/album_pic.php?pic_id=170 Ultracruiser (with 1/2 VW) is the same deal |
#4
|
|||
|
|||
"smjmitchell" wrote in message . au...
5. Galvanic corrosion .. magnesium is at -1.6 V and alumnium at -0.75 V on the galvanic table. That is a big different. You definitely need something to separate the two or the magnesium is going to get gobbled up !! Interesting. Where might I information like this for making decisions on what metals can be safely bolted together w/r/t galvanic corrosion? At issue right now is whether it'll be OK to thread my brass fuel-system finger strainers and fuel drains into aluminum blocks, but other such questions will certainly come up in future. Thanks, Greg |
#5
|
|||
|
|||
Thanx for yourlearned and informed reply.
"smjmitchell" wrote in message . au... To be honest this all sounds like a bit of a mess. It does to me too, but on the other hand, there are a fair number of flying installations like this, in spite of my "desk engineer" concerns. 1. Most common epoxies have a Glass Transaition Temperature (Tg) of approx 90 deg C. If the epoxy is taken above that temperature then two things happen ... first it softens and the stiffness and strength reduce dramatically. Second irreveraible damage is done to the epoxy and it will never be the same again - even when cooled. Unless you are using one of a small number of epoxies that are designed for high temperature operation (some of these have Tg of approx 400 F, 200 C I think without reaching for a calculator) then it is certain I think that if used on an engine the 90 deg C limit will be exceeded. That's about 190 deg F, which is a pretty comfortable oil temerature, and this arrangement attaches to the side of the oil sump. But oil temps up to 250 deg F need to be thought about. And that's higher than most epoxy resins and adhesives, even post-cured. I don't know what epoxy is used, I'll find out. 2. Next there is the issue of surface preparation. If the plans say prepare by running with 80 grid paper then it is fairly clear that the guy who wrote the plans knows little about what he is doing ! Epoxy metal bonding is reliable if the surfaces are prepared properly but from what you have said I doubt that is the case. That's actually not outside the "standard of Care" for metal bonding in low-stress operations. Gougeon brothers, makers of WEST and PRO-SET epoxy, suggest sanding and etching for aluminum. The side of the magnesium case is fairly rough, and not particularly flat. Sanding it flattens it, removes the oxide coating, and still leaves some "tooth". 3. Bolted and bonded joints should be avoided because it is difficult to predict the load transfer etc ... I won't go into detail on this - perhaps later. My point exactly. 4. What are the differences in the thermal expansion coefficients of the aluminium and epoxy ... from MIL-HDBK-5J alumnium is approx 12.5e-6 in/in/F and Magnesium is 14.0e-6 ... not a lot of difference perhaps this is not an issue. There is also the steel bolts to consider, but their lower coefficient of expansion actually tightens up the joint. Perhaps to the point of fracturing the epoxy! I can't find good numbers on cured resin alone, but cured composites have pretty low coefficients, like 2.0e-6. 5. Galvanic corrosion .. magnesium is at -1.6 V and alumnium at -0.75 V on the galvanic table. That is a big different. You definitely need something to separate the two or the magnesium is going to get gobbled up !! Hmm, perhaps. I get that aluminum alloy is about -1.05. But look at the difference between Cast Iron (-.5) and aluminum (-1.05). Lots of cars, with WATER running through their engines, have cast iron blocks and aluminum heads. In this situation, there is no ready source of electrolyte. I don't see this as a huge concern. I'm more bothered by oil leaks. 6. I think galling of the metal is a possibility but if appropriate tolerances are used for the bolts and holes than this would be less of a problem. Hard to say without seeing the drawings etc. Sounds to me like if the holes are flogging out then the design has some fundamental problems and that one should not be relying on epoxy that probably cannot withstand the temperatures to fix it. Amen! Albeit, there are those pesky flying examples to be explained away.... My gut feel is that you need something between the alumnium and the magnesium for corrosion protection and possibly the help the galling issue. I would assume that this is sufficiently ductile an rubbery that it will not pick up any load and that the fasteners will transfer all the load. I would use a rudder like sealant compound that can take the temperature ... not epoxy. When you use sealant of this type in a joint with fasteners extra largers of safety should be allowed because of the extra flexibility that this produces in the joint (typically an extra factor of 1.5). This aligns with my thinking. High Temperature Room Temperature Vulcanizing Silicone rubber looks good for this. The security of the joint would be in the close fit of the mating parts, not in the epoxy. The bolts would be a tight fit in reamed holes, backed up with Loctite Red. "Ryan Young" wrote in message om... I'm working on a Hummel Aviation Ultracruiser Plus. The way the engine mounts is quite different from the mounts on most VW powered homebuilts. The Conventional Wisdom bolts the thing to the firewall using the clucth end bellhousing, often with an accessory case in between. The Ultracruiser Plus is different. Two aluminum angles are bolted to the sides of the magnesium case, in the sump area, and, suitablely reinforced, are used to bolt Berry mounts to "bed" type engine bearers built up out of aluminum, that extend from the forward fuselage. These angles are bolted and epoxied to the side of the magnesium case. My point: what good is the epoxy? It's probably not carrying any loads. A basic tenet of structural design is that the stiffest load path carries the load, and the bolts through the angle and into the case (secured with nuts and washers inside the sump, before the engine is assembled) seem a bunch stiffer that the epoxy. It's not a sure stop against leaks. Epoxy is a wonderful material, but it doesn't bond particularly well or reliably to metals. Plus, it's mechanical properties, from it's modulus of expansion, to it's ductility, are far different that the aluminum, steel, and magnesium sandwich is it the Mayonnaise of. My concern is the epoxy will eventually crack. I lost the reply from Scott Casler of Hummel Engines, I'll paraphrase: "The epoxy is to keep the angles from working and hogging out the holes. The epoxy I use is a real good sealer, you've got to grind it off." My thoughts are this: LAP the angles to the side of the case (instead of sanding with 80 grit), but use Permatex or Curil T to seal things. Use close tolerance bolts in reamed holes in the side of the case and the appropriate Loctite product to seal the bolts. And I'm inclined to put the bolt heads INSIDE the engine. Comments? To see what this installation looks like: http://flyhummel.com/forums/album_pic.php?pic_id=170 Ultracruiser (with 1/2 VW) is the same deal |
#6
|
|||
|
|||
I was thinking ProSeal or similar for the interface...
"smjmitchell" wrote in message u... To be honest this all sounds like a bit of a mess. There are a number of issues here ... however it is difficult to comment in any detail until I know what sort of epoxy is specified and without further details of the surface preparation and details of the design. 1. Most common epoxies have a Glass Transaition Temperature (Tg) of approx 90 deg C. If the epoxy is taken above that temperature then two things happen ... first it softens and the stiffness and strength reduce dramatically. Second irreveraible damage is done to the epoxy and it will never be the same again - even when cooled. Unless you are using one of a small number of epoxies that are designed for high temperature operation (some of these have Tg of approx 400 F, 200 C I think without reaching for a calculator) then it is certain I think that if used on an engine the 90 deg C limit will be exceeded. 2. Next there is the issue of surface preparation. If the plans say prepare by running with 80 grid paper then it is fairly clear that the guy who wrote the plans knows little about what he is doing ! Epoxy metal bonding is reliable if the surfaces are prepared properly but from what you have said I doubt that is the case. 3. Bolted and bonded joints should be avoided because it is difficult to predict the load transfer etc ... I won't go into detail on this - perhaps later. 4. What are the differences in the thermal expansion coefficients of the aluminium and epoxy ... from MIL-HDBK-5J alumnium is approx 12.5e-6 in/in/F and Magnesium is 14.0e-6 ... not a lot of difference perhaps this is not an issue. 5. Galvanic corrosion .. magnesium is at -1.6 V and alumnium at -0.75 V on the galvanic table. That is a big different. You definitely need something to separate the two or the magnesium is going to get gobbled up !! 6. I think galling of the metal is a possibility but if appropriate tolerances are used for the bolts and holes than this would be less of a problem. Hard to say without seeing the drawings etc. Sounds to me like if the holes are flogging out then the design has some fundamental problems and that one should not be relying on epoxy that probably cannot withstand the temperatures to fix it. My gut feel is that you need something between the alumnium and the magnesium for corrosion protection and possibly the help the galling issue. I would assume that this is sufficiently ductile an rubbery that it will not pick up any load and that the fasteners will transfer all the load. I would use a rudder like sealant compound that can take the temperature ... not epoxy. When you use sealant of this type in a joint with fasteners extra largers of safety should be allowed because of the extra flexibility that this produces in the joint (typically an extra factor of 1.5). "Ryan Young" wrote in message om... I'm working on a Hummel Aviation Ultracruiser Plus. The way the engine mounts is quite different from the mounts on most VW powered homebuilts. The Conventional Wisdom bolts the thing to the firewall using the clucth end bellhousing, often with an accessory case in between. The Ultracruiser Plus is different. Two aluminum angles are bolted to the sides of the magnesium case, in the sump area, and, suitablely reinforced, are used to bolt Berry mounts to "bed" type engine bearers built up out of aluminum, that extend from the forward fuselage. These angles are bolted and epoxied to the side of the magnesium case. My point: what good is the epoxy? It's probably not carrying any loads. A basic tenet of structural design is that the stiffest load path carries the load, and the bolts through the angle and into the case (secured with nuts and washers inside the sump, before the engine is assembled) seem a bunch stiffer that the epoxy. It's not a sure stop against leaks. Epoxy is a wonderful material, but it doesn't bond particularly well or reliably to metals. Plus, it's mechanical properties, from it's modulus of expansion, to it's ductility, are far different that the aluminum, steel, and magnesium sandwich is it the Mayonnaise of. My concern is the epoxy will eventually crack. I lost the reply from Scott Casler of Hummel Engines, I'll paraphrase: "The epoxy is to keep the angles from working and hogging out the holes. The epoxy I use is a real good sealer, you've got to grind it off." My thoughts are this: LAP the angles to the side of the case (instead of sanding with 80 grit), but use Permatex or Curil T to seal things. Use close tolerance bolts in reamed holes in the side of the case and the appropriate Loctite product to seal the bolts. And I'm inclined to put the bolt heads INSIDE the engine. Comments? To see what this installation looks like: http://flyhummel.com/forums/album_pic.php?pic_id=170 Ultracruiser (with 1/2 VW) is the same deal |
#7
|
|||
|
|||
Where can U get Belzona products in the US? Bill Hale
|
#8
|
|||
|
|||
(Ryan Young) wrote in message . com...
These angles are bolted and epoxied to the side of the magnesium case. My point: what good is the epoxy? I had further conversation with Scott Casler. ================================================== === Hi I use a epoxy that is similar to JB Weld. Scott ( Hummel Engines) ----- Original Message ----- From: ryoung@ To: humeng@ Sent: Monday, September 13, 2004 3:24 PM Subject: What Epoxy do you use? To bond the aluminum angles to the side of the magnesium case? Thanx for your previous reply to my inquiries. Ryan Young Ultracruiser Plus Builder ================================================== == From the JB Weld web site: Properties (lbs/psi) Tensile Strength: 3960 Adhesion: 1800 Flex Strength: 7320 Tensile Lap Shear: 1040 Shrinkage: 0.0% Resistant to 500º F I'm not sure what "Resistant" means in this context. JB Weld is a steel-filled epoxy adhesive. Belzona is much the same stuff, and I'm intimately familar with both of them. Devcon is another brand, and they have both the widest variety and the best technical information: http://www.devcon.com/techinfo/107.pdf Aluminum filled epoxy. Temperature Resistance Wet: 120°F; Dry: 250°F Adhesive Tensile Shear 2,600 psi Compressive Strength 8,420 psi Modulus of Elasticity 8.0 psi x 10(5) in. Flexural Strength 6,760 psi Coefficient of Thermal Expansion 29 [(in.)/(in). x °F)] x 10(-6) Wowee Kazowee, look how much it expands! Twice as much as aluminum or magnesium! http://www.devcon.com/techinfo/101.pdf Steel filled epoxy putty, more like JB Weld Adhesive Tensile Shear 2800 psi Compressive Strength 8260 psi Modulus of Elasticity 8.5 x 10(5) psi Flexural Strength 5600 psi Coefficient of Thermal Expansion 48 [(in)/(in) x °F)] x 10(-6) Even more expansive, and not much stronger. Application instructions which cut pretty close to Hummel's plans: 1. Thoroughly clean the surface with Devcon® Cleaner Blend 300 to remove all oil, grease, and dirt. This is probably a detergent - RRY 2. Grit blast surface area with 8-40 mesh grit, or grind with a coarse wheel or abrasive disc pad, to create increased surface area for better adhesion 3. Clean surface again with Cleaner Blend 300 to remove all traces of oil, grease, dust, or other foreign substances from the grit blasting. 4. Repair surface as soon as possible to eliminate any changes or surface contaminants. |
#9
|
|||
|
|||
2. Next there is the issue of surface preparation. If the plans say prepare by running with 80 grid paper then it is fairly clear that the guy who wrote the plans knows little about what he is doing ! Epoxy metal bonding is reliable if the surfaces are prepared properly but from what you have said I doubt that is the case. I recently did some testing with West Systems and aluminum. I used 3 test strips, side by side. 1" wide fiberglass tape, 3 inches of bonding surface on 6061. The first strip was applied with no surface impression at all. The second strip was applied after sanding with 80 Grit The third was sanded and etched with an etch (Duramix 4925) The first strip came off with maybe a pound of force, pulling on the end of the fiberglass tape. The second strip (sanded) took maybe twice that - it still came right off The third (etched) strip never did come off, I tore the fiberglass tape It stood up to at least 30 pounds in shear! Surface preparation makes a BIG difference!. |
#10
|
|||
|
|||
|
Thread Tools | |
Display Modes | |
|
|