<html> <body> Al - You have gotten a good deal of useful information in response to your question.  The problem is a lot like pizza.  How can something which seems to have so few ingredients come out so different? Well, without actually knowing the answer, I'll tell you what I think.  Pinning - It's true that excessively tight pinning in the repetition lever, hammerflange, and even the jack*, could have a negative impact on the checking, but you seem to have eliminated that factor.  Of course, the assumption with regard to the pinning is that the elevated friction prevents the hammertail from addressing the backcheck with enough velocity to both a) sufficiently compress the rep spring, and b) adequately engage the frictive (not a real word) forces that will counter the restorative impulse of the compressed rep spring. Of course, depending upon factors mentioned in previous responses, such as the effective tail to backcheck distance at let-off, the distance the hammer has to travel (downward) before it even makes initial contact with the backcheck will affect how much spring compression has to be overcome (to maintain checking).  Once checking has been accomplished, that pinning friction would seem to add some resistance to the un-checking impulse of the rep spring.  If the pinning is not the real problem, reducing  the friction will undermine other aspects of the action's response. Roughing tails / backchecks - As you pointed out, you've been able to accomplish checking without such measures when you do your own shaping, so, as with easing the pinning friction, this is more likely a compensation for some other failing in the process.  We could get microscopic, and examine what such roughing is actually doing (or not doing), but, when treating the tail thus, you would likely be accelerating the wear on the buckskin, and roughing the leather would seem likely to, at best, create an uncontrolled nap (remember nap?) which could work equally against the initial 'grab' function. Shape - Obviously, there are MANY different concepts of both tail and backcheck profile which seem to work.  The point is how the two work together, and in conjunction with the other factors that create checking.  One thing that I did not see mentioned, with regard to shape, is the congruence of the faces of the tail and backcheck, specifically at the contact surfaces.  I have corrected many such misalignments, on both new and rebuilt actions. How backcheck functions - In the context of the current technology (thus excluding velcro, magnatism, etc.), there is a friction component,  a spring component, and a mechanical interface component. - Ideally, friction should be the least active, since it is hard to limit the action of the friction to one direction. - The mechanical interface should achieve adequate surface to surface contact and engage with the least amount of shock, but sufficiently defined so that it functions consistently, thus the importance of complementary profiles.  Related to the profiles, the reference above to "effective tail to backcheck distance" refers to the actual point at which the two would engage, depending upon their profiles, rather than the simple measured distances of each extremety (tip of tail and top of backcheck) from a defined base. - That leaves the spring function, which is achieved by the combined deflective behavior of the backcheck wire, the backcheck felt, and buckskin (or synthetic) covering.   All other things equal, that combined spring function controls the checking.  For example, you could, in theory, have a totally rigid 'wire' if the correct quality felt and skin were employed. The resilient qualities of the 'skin' material has as much of an effect as the surface 'nap'.  That said, the newer type backcheck wires, considerably more rigid, have to have a profound difference on backcheck behavior.  With regard to Bill Monroe's statement: <blockquote type=cite class=cite cite=""> I wouldn't concern myself where the bend is to establish the angle as long as the angle is correct. </blockquote>the effect might be subtle, but I can see where the location of that bend would, in fact, have an impact.  For example, a wire emerging from the key stick at the correct angle uses its entire length as a spring.  A vertically installed-then bent wire creates two shorter, stiffer 'arms' which may not function in the same way. Last, but probably most to the point, you said: <blockquote type=cite class=cite cite="">The tails are quit rounded now. Do you think they should be a little flatter?</blockquote> For sure.  With excessive rounding you have two problems. First, you reduce the effective contact surface. Second, you lower (increase) the effective point at which tail and check first contact, since the check must still be set back enough to clear the 'bump' in the tail, but now contacts a point further up on the tail than it would were the tail flatter. That's my brief response, I guess. *Theoretically, if jack is pinned too tight, tension of butterfly spring would have have to be increased, which would have an impact upon hammer rise, etc.  It's a stretch, perhaps David Skolnik RPT Hastings on Hudson, NY I could be wrong At 11:47 AM 3/13/2008, you wrote: <blockquote type=cite class=cite cite="">I am regulating a 10 year old Steinway M with all original action parts. I am having a problem with checking. The piano checks 100% on the bench, but no checking at all in the bass section and 85% OK in all other sections.  I have adjusted the angle on the backchecks with no success. Does anyone have any thoughts on what the problem my be and a possible fix?   Al Guecia No virus found in this incoming message. Checked by AVG. Version: 7.5.519 / Virus Database: 269.21.7/1327 - Release Date: 3/12/2008 1:27 PM</blockquote></body> </html>