GrantKing
New member
magazine testing part2
continued from last post:
West liaised with me before running their last tests, even if they did deviate from the plan when half the test anchors would not set when testing on the “edge” as I suggested. We also provided the test Rocna anchor, air freight from NZ.
If you know a test is to be run it is best to provide the anchor (sponsor). This way you can provide a tuned anchor you know is correct in every aspect, especially the correct tip shape and sharpness which is critical for immediate good test results. A number of tuned 10s and 15s, as well as Stowables, should be kept in stock for these occasions. Don’t send a RRR, even if there is a market for the small sizes, the slots on the SARCA and Supremes are universally criticized and receive negative publicity.
Some factors to be considered in any test design:
1. A realistic environment for which general purpose anchors are designed. Not frozen beaches (Starzinger / Practical Sailor), solid rock, dry substrate, etc. Be wary of weed/grass also, it’s extremely difficult to get consistent results and although Rocna should still out-perform, the need for multiple trials would be all the more. Also a broad range of bottom types is essential to weed out the specialist anchors such as Fortress. E.g., sand, mud (hard and soft), clay, sand/mud and stone-shell-molluscs-weed combination, sand/mud and weed, kelp, dead coral debris and a stone/rock combination.
2. Multiple trials in a cyclical set-up (not anchor A trial 1-2-3 anchor B trial 1-2-3, but anchor A trial 1, anchor B trial 1, A-2, B-2, A-3, B-3, etc) to reduce environment “drift” – it’s a hassle to swap anchor and shackle continuously so there is resistance to doing this properly. Ideally the test vehicle should be held stationary over the same bottom for each cycle, maybe between a marked out lane.
3. Willingness to reject obvious aberrations such as a result that a anchor performs better at shorter scope or other impossibilities; data needs more analysis and averaging.
4. Control over environment. Difficult to find consistent bottom in large area at constant depth. See 1 and 2 above.
5. Testing of setting performance, e.g. measure of distance to set and consistency. This needs to be observed in shallow water or with a good dive team. More emphasis should be given to the “quick set” feature, a major advantage of a modern anchor.
6. Test of ultimate holding power, not some abstract low level limit which almost all the test samples will attain.
7. Control over pulls, need to use twin screw heavy powerful boat for even and measured force application, not a small yacht “because it's realistic”.
8. Constant rode at realistic make-up and scope(s). I don’t like rope and chain combinations for testing, there is too much spring created by the anchor as the bottom is very rarely even in consistency. Wire would be better but harder to handle. All chain would be okay as at load it will be straight, so scope ceases to be a consideration, or use lighter chain. It is a constant over all the anchors anyway and the results will be much easier to analyze.
9. Control over set time, set quality improves over time, this must be constant. Set then leave 5 mins under identical conditions then pull, or whatever. I like to see a series of light pulls between no throttle before consistent high load applied to give more realistic results.
10. Fair treatment of “old friends”, e.g. West Marine/SAIL gave way more attention to CQR than was fair because they were so surprised it failed to work. A new unknown would have been written off straight away.
11. No commercial bias, especially in small markets. Look for brands/manufacturers local to the magazine? Are there big spenders w.r.t advertising? Are they touting for ads placed in context with the write-up? Old boys’ network? Typical in NZ and Aus magazines and probably other smaller.
12. Analysis of results:
· Fair weight-for-weight comparison, e.g. Rocna 15 is 33 lb, closest Supreme is 36 lb = 10% difference which they typically won’t account for.
· Results should be “normalised” on a weight-for-weight basis.
· Watch for them asking for the “manufacturer’s recommended size”, otherwise you get rubbish like a 20 kg Rocna compared to a 10 kg Delta.
· No direct comparison of mixed metal types (steel / alloy is typical problem), apples/oranges. Both the alloy Fortress and the alloy Spade are notorious for this one and is the fault of the testers’ lack of understanding.
· Record sensible numbers (holding power is the max force the anchor sustains up to and before it moves – this seems a hard concept for magazines to get their heads around).
· How is a failure to set analyzed? – “infinite” set distance = difficult to graph.
To repeat: If they ask for an anchor and you decide to provide, then make 100% sure it is perfect. Check symmetry, shank straightness, fluke sharpness (sharper is better for test, they don't test longevity/durability!), etc.
As to the German “test”, the results are bizarre and their analysis not consistent with their own graph, but I can’t provide any answers until I see the test anchor. However, we have told you repeatedly that the Chinese anchors that you are shipping are NOT to spec, and we have yet to see one that is acceptable. I assume the anchor that the Germans tested suffers from the same problems. It should not have been provided; rather, send them a tuned sample ex CNC.
Trust this is what you are looking for and is of some help,
Regards Peter Smith.
continued from last post:
West liaised with me before running their last tests, even if they did deviate from the plan when half the test anchors would not set when testing on the “edge” as I suggested. We also provided the test Rocna anchor, air freight from NZ.
If you know a test is to be run it is best to provide the anchor (sponsor). This way you can provide a tuned anchor you know is correct in every aspect, especially the correct tip shape and sharpness which is critical for immediate good test results. A number of tuned 10s and 15s, as well as Stowables, should be kept in stock for these occasions. Don’t send a RRR, even if there is a market for the small sizes, the slots on the SARCA and Supremes are universally criticized and receive negative publicity.
Some factors to be considered in any test design:
1. A realistic environment for which general purpose anchors are designed. Not frozen beaches (Starzinger / Practical Sailor), solid rock, dry substrate, etc. Be wary of weed/grass also, it’s extremely difficult to get consistent results and although Rocna should still out-perform, the need for multiple trials would be all the more. Also a broad range of bottom types is essential to weed out the specialist anchors such as Fortress. E.g., sand, mud (hard and soft), clay, sand/mud and stone-shell-molluscs-weed combination, sand/mud and weed, kelp, dead coral debris and a stone/rock combination.
2. Multiple trials in a cyclical set-up (not anchor A trial 1-2-3 anchor B trial 1-2-3, but anchor A trial 1, anchor B trial 1, A-2, B-2, A-3, B-3, etc) to reduce environment “drift” – it’s a hassle to swap anchor and shackle continuously so there is resistance to doing this properly. Ideally the test vehicle should be held stationary over the same bottom for each cycle, maybe between a marked out lane.
3. Willingness to reject obvious aberrations such as a result that a anchor performs better at shorter scope or other impossibilities; data needs more analysis and averaging.
4. Control over environment. Difficult to find consistent bottom in large area at constant depth. See 1 and 2 above.
5. Testing of setting performance, e.g. measure of distance to set and consistency. This needs to be observed in shallow water or with a good dive team. More emphasis should be given to the “quick set” feature, a major advantage of a modern anchor.
6. Test of ultimate holding power, not some abstract low level limit which almost all the test samples will attain.
7. Control over pulls, need to use twin screw heavy powerful boat for even and measured force application, not a small yacht “because it's realistic”.
8. Constant rode at realistic make-up and scope(s). I don’t like rope and chain combinations for testing, there is too much spring created by the anchor as the bottom is very rarely even in consistency. Wire would be better but harder to handle. All chain would be okay as at load it will be straight, so scope ceases to be a consideration, or use lighter chain. It is a constant over all the anchors anyway and the results will be much easier to analyze.
9. Control over set time, set quality improves over time, this must be constant. Set then leave 5 mins under identical conditions then pull, or whatever. I like to see a series of light pulls between no throttle before consistent high load applied to give more realistic results.
10. Fair treatment of “old friends”, e.g. West Marine/SAIL gave way more attention to CQR than was fair because they were so surprised it failed to work. A new unknown would have been written off straight away.
11. No commercial bias, especially in small markets. Look for brands/manufacturers local to the magazine? Are there big spenders w.r.t advertising? Are they touting for ads placed in context with the write-up? Old boys’ network? Typical in NZ and Aus magazines and probably other smaller.
12. Analysis of results:
· Fair weight-for-weight comparison, e.g. Rocna 15 is 33 lb, closest Supreme is 36 lb = 10% difference which they typically won’t account for.
· Results should be “normalised” on a weight-for-weight basis.
· Watch for them asking for the “manufacturer’s recommended size”, otherwise you get rubbish like a 20 kg Rocna compared to a 10 kg Delta.
· No direct comparison of mixed metal types (steel / alloy is typical problem), apples/oranges. Both the alloy Fortress and the alloy Spade are notorious for this one and is the fault of the testers’ lack of understanding.
· Record sensible numbers (holding power is the max force the anchor sustains up to and before it moves – this seems a hard concept for magazines to get their heads around).
· How is a failure to set analyzed? – “infinite” set distance = difficult to graph.
To repeat: If they ask for an anchor and you decide to provide, then make 100% sure it is perfect. Check symmetry, shank straightness, fluke sharpness (sharper is better for test, they don't test longevity/durability!), etc.
As to the German “test”, the results are bizarre and their analysis not consistent with their own graph, but I can’t provide any answers until I see the test anchor. However, we have told you repeatedly that the Chinese anchors that you are shipping are NOT to spec, and we have yet to see one that is acceptable. I assume the anchor that the Germans tested suffers from the same problems. It should not have been provided; rather, send them a tuned sample ex CNC.
Trust this is what you are looking for and is of some help,
Regards Peter Smith.
