interface problem

[marzal]

Every experts:
Recently I have some trouble with the rejected interface of the coupon (not cover the whole interface, but the total length about 200-400um ), and I have tried to change the blasting grits to the new one, and adjust the cooling air, change the coupon to the thick one, but the results are still not good.
the substrate material is Aluminum, and the powder is metco 450NS, the gun is F4 gun.
So what caused this kind of defects? And how to solve it ? what should be pay attention to when do the surface preparation of the Aluminum?
Any help will be appreciate~

[Laurent]

(08-12-2015 07:37 AM)marzal Wrote:  Every experts:
Recently I have some trouble with the rejected interface of the coupon (not cover the whole interface, but the total length about 200-400um ), and I have tried to change the blasting grits to the new one, and adjust the cooling air, change the coupon to the thick one, but the results are still not good.
the substrate material is Aluminum, and the powder is metco 450NS, the gun is F4 gun.
So what caused this kind of defects? And how to solve it ? what should be pay attention to when do the surface preparation of the Aluminum?
Any help will be appreciate~

Hello,
What mesh size do you use to blast substrate?
Do you machine and clean substrate (degreasing) prior to coating?
Thanks
Cheers

[Gordon]

Hi Marzal

In addition to questions by Laurent:

Can you describe your substrate preheating procedure and thickness per pass of the first application of coating?

[marzal]

(08-25-2015 10:00 AM)Laurent Wrote:  

(08-12-2015 07:37 AM)marzal Wrote:  Every experts:
Recently I have some trouble with the rejected interface of the coupon (not cover the whole interface, but the total length about 200-400um ), and I have tried to change the blasting grits to the new one, and adjust the cooling air, change the coupon to the thick one, but the results are still not good.
the substrate material is Aluminum, and the powder is metco 450NS, the gun is F4 gun.
So what caused this kind of defects? And how to solve it ? what should be pay attention to when do the surface preparation of the Aluminum?
Any help will be appreciate~

Hello,
What mesh size do you use to blast substrate?
Do you machine and clean substrate (degreasing) prior to coating?
Thanks
Cheers

Thank you for your attention, the blast grits size is #80,and prior to coating we use ethanol to clean the specimen.

[marzal]

(08-25-2015 11:34 AM)Gordon Wrote:  Hi Marzal

In addition to questions by Laurent:

Can you describe your substrate preheating procedure and thickness per pass of the first application of coating?

Hi,Gordon:
I admire you so much, In this forum, i get so much very professional suggestion from you, thank you we can with you in this forum.
To return to my question, i didn't do the preheating because the substrate is aluminium alloy,and thickness per pass is 0.0145mm/pass.
another question, how to control the thickness per pass ? is it just affected by Rotation speed(RPM) and powder feed rate? which is the best range of the thickness per pass ?

---

In addition information:
the diameter of the fixture is 95mm, and the Rotation is 160rpm

[Gordon]

Hi Marzal

No preheat is wise as long as there is no condensation problem, in which case a gentle indirect warming (from rear side) to about 60 C will be beneficial.

There should be no need to wash in alcohol prior to spraying and after grit blasting. Grit blasting activates the surface and should be sprayed as quickly as possible. Any delay or other processes will only act to deactivate/contaminate the surface. If there appears to be a reason to wash after grit blasting, then it would be best to start again.

Metco 450 is exothermic when sprayed and unlike many spray materials performs best when applied relatively thick per pass. The Metco 450 technical bulletin even states "The best bond coat is obtained if a bond coat 0.004" to 0.006" thick is applied in one pass. Bond coats less than 0.004" thick should not be used to avoid possibility of reduced bond strength. Bond coats of more than 0.006" thick do not add to bond strength"

Bear in mind that the Metco 450 microstructure naturally has many oxide stringers and many of these will line the interface which is normal.

Have you conducted tensile bond strength tests? Have you tried different metallographic polishing techniques to make sure you are not exaggerating interface defects?

[marzal]

(08-26-2015 03:20 PM)Gordon Wrote:  Hi Marzal

No preheat is wise as long as there is no condensation problem, in which case a gentle indirect warming (from rear side) to about 60 C will be beneficial.

There should be no need to wash in alcohol prior to spraying and after grit blasting. Grit blasting activates the surface and should be sprayed as quickly as possible. Any delay or other processes will only act to deactivate/contaminate the surface. If there appears to be a reason to wash after grit blasting, then it would be best to start again.

Metco 450 is exothermic when sprayed and unlike many spray materials performs best when applied relatively thick per pass. The Metco 450 technical bulletin even states "The best bond coat is obtained if a bond coat 0.004" to 0.006" thick is applied in one pass. Bond coats less than 0.004" thick should not be used to avoid possibility of reduced bond strength. Bond coats of more than 0.006" thick do not add to bond strength"

Bear in mind that the Metco 450 microstructure naturally has many oxide stringers and many of these will line the interface which is normal.

Have you conducted tensile bond strength tests? Have you tried different metallographic polishing techniques to make sure you are not exaggerating interface defects?

Gordon:
recently we did three sets of test,1:increase the rotation rate to 160rpm,
2:increase the primary gas flow to 55, keep the rotation rate in 100rpm
3:decrease the powder feed rate to 20g/min.

parameter 1, current:630A, primary gas flow rate:45nlpm,secondary gas(H2):9.0nlpm, powder feed rate:30, carrier gas flow:2.5, rotation rate:160rpm
Tensile test result: 44.0MPa,40.4Mpa,39.9Mpa
parameter 2: current:630A, primary gas flow rate:55nlpm,secondary gas(H2):9.0nlpm, powder feed rate:30, carrier gas flow:2.5, rotation rate:100rpm
Tensile test result: 23.8MPa,30.3Mpa,30.4Mpa
parameter 3: current:630A, primary gas flow rate:45nlpm,secondary gas(H2):9.0nlpm, powder feed rate:20, carrier gas flow:2, rotation rate:100rpm
Tensile test result: 42.1MPa,40.0Mpa,35.8Mpa

yes we tried different metallographic cutting parameters, decrease the feed rate of specimen in cutting process.but the problem still have.

[Vadim Verlotski]

Hi marzal,

From your tests is quite clear that the problem is located in an excessive thickness of each monolayer:
The second test parameters have the monolayers reached a maximum thickness and has led to the lowest adhesion strength.

Regards

[Gordon]

Hi Marzal and Vadim Verlotski

(08-27-2015 10:04 AM)Vadim Verlotski Wrote:  Hi marzal,

From your tests is quite clear that the problem is located in an excessive thickness of each monolayer:
The second test parameters have the monolayers reached a maximum thickness and has led to the lowest adhesion strength.

Regards

I would not jump to this conclusion on these tests.

Quote:recently we did three sets of test,1:increase the rotation rate to 160rpm,
2:increase the primary gas flow to 55, keep the rotation rate in 100rpm
3:decrease the powder feed rate to 20g/min.

parameter 1, current:630A, primary gas flow rate:45nlpm,secondary gas(H2):9.0nlpm, powder feed rate:30, carrier gas flow:2.5, rotation rate:160rpm
Tensile test result: 44.0MPa,40.4Mpa,39.9Mpa
parameter 2: current:630A, primary gas flow rate:55nlpm,secondary gas(H2):9.0nlpm, powder feed rate:30, carrier gas flow:2.5, rotation rate:100rpm
Tensile test result: 23.8MPa,30.3Mpa,30.4Mpa
parameter 3: current:630A, primary gas flow rate:45nlpm,secondary gas(H2):9.0nlpm, powder feed rate:20, carrier gas flow:2, rotation rate:100rpm
Tensile test result: 42.1MPa,40.0Mpa,35.8Mpa

How would you know the effect is not down to the increase in primary flow, decrease in feed rate or decrease in carrier flow? If all tests kept all parameters constant except for rotation speed and traverse rate then you have a chance at drawing a conclusion.

Marzal were the bond strength tests showing adhesive (substrate bond) or cohesive (coating) failures? Metco 540 should typically fail cohesively or at Metco 450/top coat interface if coating is normal.

[Vadim Verlotski]

Hi Gordon,

Gordon, you're right, only one parameter may be changed per a one test. Nevertheless, I find that the result is clear:
the influence of powder rate and rotation speed for the thickness of monolayers is significantly greater than influence of the gas flow.

Regards
Vadim

[marzal]

(08-27-2015 02:30 PM)Gordon Wrote:  Hi Marzal and Vadim Verlotski

(08-27-2015 10:04 AM)Vadim Verlotski Wrote:  Hi marzal,

From your tests is quite clear that the problem is located in an excessive thickness of each monolayer:
The second test parameters have the monolayers reached a maximum thickness and has led to the lowest adhesion strength.

Regards

I would not jump to this conclusion on these tests.

Quote:recently we did three sets of test,1:increase the rotation rate to 160rpm,
2:increase the primary gas flow to 55, keep the rotation rate in 100rpm
3:decrease the powder feed rate to 20g/min.

parameter 1, current:630A, primary gas flow rate:45nlpm,secondary gas(H2):9.0nlpm, powder feed rate:30, carrier gas flow:2.5, rotation rate:160rpm
Tensile test result: 44.0MPa,40.4Mpa,39.9Mpa
parameter 2: current:630A, primary gas flow rate:55nlpm,secondary gas(H2):9.0nlpm, powder feed rate:30, carrier gas flow:2.5, rotation rate:100rpm
Tensile test result: 23.8MPa,30.3Mpa,30.4Mpa
parameter 3: current:630A, primary gas flow rate:45nlpm,secondary gas(H2):9.0nlpm, powder feed rate:20, carrier gas flow:2, rotation rate:100rpm
Tensile test result: 42.1MPa,40.0Mpa,35.8Mpa

How would you know the effect is not down to the increase in primary flow, decrease in feed rate or decrease in carrier flow? If all tests kept all parameters constant except for rotation speed and traverse rate then you have a chance at drawing a conclusion.

Marzal were the bond strength tests showing adhesive (substrate bond) or cohesive (coating) failures? Metco 540 should typically fail cohesively or at Metco 450/top coat interface if coating is normal.

Gordon:
The requirement for the bond strength is 30MPa, but the interface problem still have, the metallergraphic test failed. there are still some interface contamination in the substrate-coating interface, so i considered it may be the interface contamination problem, next i plan to change the blasting parameters and keep the
spraying parameters of test 1:
blasting parameter 1: decrease the blasting pressure from 1.6 to 1.3 bar,
change the flow rate from 1.5 to 1.0. keep the blasting angle on 75°。
blasting parameter 2: keep the pressure on 1.6 bar, keep the flow rate on 1.5kg/min, change the blasting angle from 75 to 55.
so how do you think my plan? i hope you can give me some suggestions, thank you in advance!

---

(08-27-2015 10:04 AM)Vadim Verlotski Wrote:  Hi marzal,

From your tests is quite clear that the problem is located in an excessive thickness of each monolayer:
The second test parameters have the monolayers reached a maximum thickness and has led to the lowest adhesion strength.

Regards

Vadim Verlotski:
thank you for your reply, the thickness may be one of the reasons,but the main problem is the interface problem, the metollographic test failed.

[Laurent]

(08-28-2015 01:54 AM)marzal Wrote:  

(08-27-2015 02:30 PM)Gordon Wrote:  Hi Marzal and Vadim Verlotski

(08-27-2015 10:04 AM)Vadim Verlotski Wrote:  Hi marzal,

From your tests is quite clear that the problem is located in an excessive thickness of each monolayer:
The second test parameters have the monolayers reached a maximum thickness and has led to the lowest adhesion strength.

Regards

I would not jump to this conclusion on these tests.

Quote:recently we did three sets of test,1:increase the rotation rate to 160rpm,
2:increase the primary gas flow to 55, keep the rotation rate in 100rpm
3:decrease the powder feed rate to 20g/min.

parameter 1, current:630A, primary gas flow rate:45nlpm,secondary gas(H2):9.0nlpm, powder feed rate:30, carrier gas flow:2.5, rotation rate:160rpm
Tensile test result: 44.0MPa,40.4Mpa,39.9Mpa
parameter 2: current:630A, primary gas flow rate:55nlpm,secondary gas(H2):9.0nlpm, powder feed rate:30, carrier gas flow:2.5, rotation rate:100rpm
Tensile test result: 23.8MPa,30.3Mpa,30.4Mpa
parameter 3: current:630A, primary gas flow rate:45nlpm,secondary gas(H2):9.0nlpm, powder feed rate:20, carrier gas flow:2, rotation rate:100rpm
Tensile test result: 42.1MPa,40.0Mpa,35.8Mpa

How would you know the effect is not down to the increase in primary flow, decrease in feed rate or decrease in carrier flow? If all tests kept all parameters constant except for rotation speed and traverse rate then you have a chance at drawing a conclusion.

Marzal were the bond strength tests showing adhesive (substrate bond) or cohesive (coating) failures? Metco 540 should typically fail cohesively or at Metco 450/top coat interface if coating is normal.

Gordon:
The requirement for the bond strength is 30MPa, but the interface problem still have, the metallergraphic test failed. there are still some interface contamination in the substrate-coating interface, so i considered it may be the interface contamination problem, next i plan to change the blasting parameters and keep the
spraying parameters of test 1:
blasting parameter 1: decrease the blasting pressure from 1.6 to 1.3 bar,
change the flow rate from 1.5 to 1.0. keep the blasting angle on 75°。
blasting parameter 2: keep the pressure on 1.6 bar, keep the flow rate on 1.5kg/min, change the blasting angle from 75 to 55.
so how do you think my plan? i hope you can give me some suggestions, thank you in advance!

---

(08-27-2015 10:04 AM)Vadim Verlotski Wrote:  Hi marzal,

From your tests is quite clear that the problem is located in an excessive thickness of each monolayer:
The second test parameters have the monolayers reached a maximum thickness and has led to the lowest adhesion strength.

Regards

Vadim Verlotski:
thank you for your reply, the thickness may be one of the reasons,but the main problem is the interface problem, the metollographic test failed.

Hi Marzal,
We yet don't have any problem with bond coat 450NS on aluminium base.
However we changed two parameters:
#1
Cleaning further to FPI (residuals of revelator)
#2
Sand mesh from 50-60 down to 20-30.

It seems that both mods improved drastically.

Hope it helps
cheers

[marzal]

Hi Laurent:
Thank for you help!
I don't known how you prepare the substrate metal before you do the spraying?
the size of the grit which we use is 80 mesh, and the roughness about 3.8um.

From now, we just care the results of specimen, otherwise we don't do the FPI before plasma spraying.

[Laurent]

(09-02-2015 06:42 AM)marzal Wrote:  Hi Laurent:
Thank for you help!
I don't known how you prepare the substrate metal before you do the spraying?
the size of the grit which we use is 80 mesh, and the roughness about 3.8um.

From now, we just care the results of specimen, otherwise we don't do the FPI before plasma spraying.

Hi Marzal,
In that case try to use sand blast grit 30-40 mesh in place of 80.
Whether you don't FPI before, try to clean ande degrease prior to blast.
Thanks

---

(09-03-2015 02:44 PM)Laurent Wrote:  

(09-02-2015 06:42 AM)marzal Wrote:  Hi Laurent:
Thank for you help!
I don't known how you prepare the substrate metal before you do the spraying?
the size of the grit which we use is 80 mesh, and the roughness about 3.8um.

From now, we just care the results of specimen, otherwise we don't do the FPI before plasma spraying.

Hi Marzal,
In that case try to use sand blast grit 30-40 mesh in place of 80.
Whether you don't FPI before, try to clean ande degrease prior to blast.
Thanks

FYI roughness is of ~2 um.

[Joris Kraak]

Hi Marzal,

I did not read everyone their comments and I am sure they are very useful and apologize if I repeat someone else but I suggest the following:

1. use low pressure for grit blasting 1.5 - 2.0 bar
2. make sure everything is prepared so that after the grit blasting you can start the coating immediately! Keep this time as short as possible.
3. make sure you use a thickness per pass as low as possible because high thickness generates heat/stress/oxidation/expansion differences.
4. cool your parts down with enough air jets and measure frequently what the temperature is.
5. if this not works you can always apply a thin layer of molybdenum, I remember we did this on a large parts for a marine application about 11 years ago. Metcoloy 2 did not work on aluminum :-)
6. do not pre-heat the part Aluminum will oxidize immediately

Good luck,
Joris