Nicrobraz Technical Articles Library

Vacuum Brazing Honeycomb Seals
 

A Brazing Q&A by Robert L. Peaslee, the "Father of Nickel Brazing"

Question:

We're having difficulty vacuum brazing honeycomb seals for use in turbine engines. Our brazing filler metal is 0.015" thick AMS 4777 transfer tape and, we're brazing at 19750 F (1080° C) for 40 minutes for maximum diffusion. However, the filler metal seems to "chew-up" the thin Inconel* 625 honeycomb, and many of the 1/32" diameter cells appear to be overfilled while other cells are empty. How can we solve this problem, and increase the quality of our brazement?

Answer:

A number of issues must be considered when brazing thin honeycomb material. Some of the more important ones are: flatness of the honeycomb, how it is tacked to the turbine backing ring, the quantity of filler metal used, and the time and temperature of the brazing run. 

Honeycomb Surface. Over the years we've encountered a wide quality range of honeycomb. Most of today's suppliers, however, are producing excellent material, with good cell seam tightness and smooth bottom surfaces. It is critical that the bottom surface be very flat and smooth, since this surface must seat uniformly on the l.D. or O.D. surface of the backing ring. Every cell must make intimate contact with the surface of the ring for a complete braze.

As shown in Figure 1a, when there is an uneven honeycomb bottom surface, resulting in only intermittent contact with the ring surface, an incomplete braze may occur. To compensate for this, some brazers put extra brazing filler metal (BFM) into the honeycomb in order to bridge these gaps. This can be a dangerous practice!

Figure 1b shows the cross section for an ideal honeycomb brazement in which all cells are flush against the base metal, and can therefore be brazed easily with a minimum amount of BFM.

There are three ways in which BFM is usually applied to honeycomb: as powder, transfer tape, or amorphous foil. The first two can be placed inside the honeycomb cells. Foil is usually placed between the honeycomb and the base-metal ring, forcing a positive clearance between the two - which must be closed during the brazing cycle.

When transfer tape is used, it is usually pressed into the honeycomb cells before the honeycomb is attached to the base metal ring. It can be pressed in by rolling it with a hard rubber, wood, or metal roller, which may be manually or machine-operated. Once the tape has been rolled in, and the plastic or paper backing sheet removed, any tape residue is usually scraped-off using a razor blade or other sharp surface. This provides a clean bottom surface to the honeycomb cells, allowing it to make intimate contact with the base metal ring in preparation for spot welding.

When BFM powder is used, the honeycomb is first attached to the ring, before the powder is put into the cells (powder usage is usually limited to honeycomb segments, or to the l.D. honeycomb of full-diameter rings). 

Tack welding. To maintain intimate contact with the base metal, the honeycomb is tack welded to the base metal ring using a spot welder or small capacitive discharge tack welder. Ideally, every cell would be tack welded to the ring during this process, but this is impractical. Thus, to maximize contact, an electrode should be made which conforms to the shape and radius of the honeycomb and the turbine ring. See Figure 2. This electrode can be pressed repeatedly against the honeycomb to maximize the number of contact welds. Once the honeycomb has been sufficiently tacked to the base-metal ring and the BFM is in place, the assembly is ready to braze.

Quantity of BFM. To minimize or eliminate base metal erosion during the brazing cycle, it is important that the correct amount of BFM be used. Some simple guidelines are: (1) the smaller the cell, the higher the number of joints required, and therefore, the more BFM will be needed; (2) as the cells get taller, total joint length increases, requiring more BFM. But how much BFM is enough? Too many shops use too much BFM! Experimentation is required! Before risking expensive parts, run small segment samples in the furnace to determine optimal BFM quantity.

We generally recommend that for 1/16" cells (about 1/2" deep), begin with 0.008" thick transfer tape. This should provide enough BFM to braze the vertical joints and leave extra to form small fillets around the base of each cell. Important note: always leave some base metal visible at the bottom center of each cell! This was demonstrated in Figure 1a. Reduce the quantity of BFM in successive test runs until high-quality brazes result and you can see base metal at the bottom of each cell.

Similarly, for 1/8" cells (approx. 1/4" deep), 0.004"-0.005" thick tape is adequate, while 0.012"-0.015" thick tape products may be needed to adequately braze 1/32" cells that are 1/2" or more deep. Trial and error is usually required to find optimal tape thickness or quantity of BFM powder for each situation. Do not use the same thickness of tape for different cell densities and heights, since they each require a different tape thickness for optimal brazing results.

Braze Cycle. An ideal brazement will exhibit fully melted BFM (no unmelted residue or leftover powder particles in the honeycomb) and well-diffused joints. To achieve this, it is recommended that the brazing temperature for the AMS 4777 material be reduced to 1925-1950° F (1050-1065° C), and the holding time at temperature reduced to 30 minutes. Higher temperatures and longer holding times can increase erosion of thin sheet metal honeycomb structures when too much boron containing BFM is present.

Many brazing shops find it beneficial to apply extra BFM, in paste form, around the outside edge of the honeycomb to insure a complete seal along the outside edges. Optimally, this should not be necessary. If a shop finds that it is absolutely necessary to use extra paste to get a good braze, then re-evaluate the honeycomb setup procedures, in accordance with this article, to try to eliminate this extra step.

* Trademark of Inco Alloys International, Inc.

Nicrobraz Order Entry
248-585-6400
(For all Canadian orders please call: 519-967-9881)
Email:customerservice@wallcolmonoy.com

• Barbara Faremouth, ext. 242
  Customer Service Manager
• Suzy Breen, ext. 238
  Customer Service Representative
• David Riley, ext. 260
  Customer Service Representative
• Darla Lorelli, ext. 234
  Export Customer Service Representative
  

Nicrobraz Technical Services
248-585-6400
Email: nicrobraztechnicalservices@wallcolmonoy.com

• Lydia Lee, ext. 252
  Brazing Products Manager
• Joel Gutierrez, Cell: 281-954-1258
  Business Manager