Cracks in 304SS in heat exchanger

[griffengm]

We recently encountered an oil fired heatexchanger that shows small cracks in two areas.
The material is .060" 304SS. Both cracks appear to have started in a weld and spread into nearby material, more or less perpendicular to the weld. One crack runs along the radius of a brake (45 deg.) but the other runs across a flat area with no bend.
The only common feature I've been able to ID is that each is in an area where two welds were lapped.
Could it be the hot/cool/hot of the weld laps causing this or is there some other possibility?

Griffy

 

[metengr]

Several questons that might help with your problem;

1. How long was the heat exchanger in service and were the cracks discovered during an inspection? What was the condition of the fireside and airside heat exchanger surfaces?

2. You mentioned the cracks started in a weld and ran into the base metal perpendicular to the weld. What is the specific weld geometry? Is the lap weld you mention actually a seam weld of a corrugated type heat exchanger found on a household furnace heat exchanger?

 

[griffengm]

metengr,
1. How long was the heat exchanger in service and were the cracks discovered during an inspection? What was the condition of the fireside and airside heat exchanger surfaces?

The HX was in service for less than two years, maybe 1000 cycles. Cracks were discovered during service. They are hairline and maybe one inch long. No evidence of leakage. The inside of the unit has some carbon build up indicating less than optimal air/fuel mix. The outside shows what we regard as normal discoloration (slightly yellow brown) over the hottest areas while the remainder is a satin silver.

What is the specific weld geometry?
Both welds are roughly 1/8" fillets. I say roughly because one area has a fairly open angle while the other is a more normal 90 degree fillet. The weld on the open angle looks more like a combination of v-down and fillet.

The cross section of this assembly is octagonal and both crack occur at a corner where welds along two faces join. There is a definite start/stop overlap in both locations.

Hope this helps.

Griffy

 

[metengr]

Giffy;
From your description, the cracks are probably low cycle fatigue cracks related to orignal fabrication residual stresses and thermal cycles.