Pianotech

I would look into the action ratio of both pianos to compare them. I would suspect capstan position and/or knuckle spread. There are many articles by Nick Gravania in the journals. Im sure I screwed up the spelling of his name but I highly respect him and the information that he has published through the journal. If you contact him he may share his excel spread sheet for action ratios with you.

Thanks for the pointer to Gravagne.  I studied those articles carefully some years ago.  I'll go back and find them and see what he has to say.

I will indeed be making appropriate measurements before I proceed.  I'm not going to simply copy what I see in the traditionally weighted keyset.

I would look into the action ratio of both pianos to compare them. I would suspect capstan position and/or knuckle spread. There are many articles by Nick Gravania in the journals. Im sure I screwed up the spelling of his name but I highly respect him and the information that he has published through the journal. If you contact him he may share his excel spread sheet for action ratios with you.

The actual pattern is less important than the relationship between SW and AR which will result in the amount of lead as an indicator of that relationship (see my thread on "Spreadsheet for Renner KMD for one discussion of that).  That relationship is something you may have to deal with in removing the assist springs.  The fact that you report 4-5 leads in the lower part of the piano each weighing 17 grams is of concern (depending on the resulting BW) because it suggests that the AR is quite high.  

So, once you establish the AR:SW relationship and the SW curve that in combination with the AR produces inertia that is in the sweet spot and results in a reasonable amount of lead in the keys then you can decide on pattern.  

Lead patterns should have some uniformity and consistency in how they progress through the keyset.  The classic 3-2-1-0 of the early Steinway pianos is a pretty good goal but not always achievable with modern hammers that can be a bit heavier.  4-3-2-1-0 pattern is likely to fall in the realm of acceptable as I outlined in that thread which is 80 - 85% of Stanwood's FW maximums for a medium BW target, say 36 - 38 grams.  That will verify an optimum relationship between AR and SW.  I'm using 1/2"  leads that weight about 14 grams each. 

As far as the actual pattern, in terms of overall inertia it doesn't make that much difference.  More leads located nearer the balance rail will result in slightly lower inertia than fewer leads toward the front of the key, all things being equal.  But the difference is marginal.  The accelerated action design that Steinway used (more lead closer to the balance rail) does help somewhat with the speed of return of the key to the rest position which, along with the balance rail bearing, which was the goal of that system.  It may be noteworthy that Steinway abandoned that method presumably because the marginal benefits weren't enough to counter the production issues, or the removal of the shoe of the key required to use the balance rail bearing, resulted in more key flex and lower levels or "action saturation".  Action saturation, for those who don't now, refers to the amount of flex in the keys and the point at which greater force at the front of the key doesn't result in greater acceleration at the back of the key (or in this case from the capstan position) as a consequence of a key that is too flexible.  

When leading from scratch, I have the first lead always in the same position until no front location is required.  That is typically about 25 mm (1 inch) or so behind the head of the naturals.  The second and third leads should start out equally spaced with the rearmost lead moving back toward the balance rail as the need for its contribution diminishes along with the SW.  This will create a uniformly graduated pattern.  I think trying to figure out some optimum placement for purposes of inertia is not worth the trouble or time.  The benefits are marginal.  The leads contribute a relatively small amount to the overall inertia and there isn't a reasonable substitute at this point.  You will need to place lead in the keys no matter what, so your choices are somewhat limited.  

If interested in pursuing this further you might find some articles by Gravagne but I would probably talk to Dean Reyburn since they are making new keysets and offering preleading for those sets to find out what his reasoning is. I have talked with Gravagne about this in the past and I won't try and restate that conversation but it led me (no pun intended) to abandon the "accelerated action" positioning and focus more on the FW:AR:SW:BW relationships simply targeting a reasonable and smoothly progressing pattern for the leads.  

The actual pattern is less important than the relationship between SW and AR which will result in the amount of lead as an indicator of that relationship (see my thread on "Spreadsheet for Renner KMD for one discussion of that).  That relationship is something you may have to deal with in removing the assist springs.  The fact that you report 4-5 leads in the lower part of the piano each weighing 17 grams is of concern (depending on the resulting BW) because it suggests that the AR is quite high.  

So, once you establish the AR:SW relationship and the SW curve that in combination with the AR produces inertia that is in the sweet spot and results in a reasonable amount of lead in the keys then you can decide on pattern.  

Lead patterns should have some uniformity and consistency in how they progress through the keyset.  The classic 3-2-1-0 of the early Steinway pianos is a pretty good goal but not always achievable with modern hammers that can be a bit heavier.  4-3-2-1-0 pattern is likely to fall in the realm of acceptable as I outlined in that thread which is 80 - 85% of Stanwood's FW maximums for a medium BW target, say 36 - 38 grams.  That will verify an optimum relationship between AR and SW.  I'm using 1/2"  leads that weight about 14 grams each. 

As far as the actual pattern, in terms of overall inertia it doesn't make that much difference.  More leads located nearer the balance rail will result in slightly lower inertia than fewer leads toward the front of the key, all things being equal.  But the difference is marginal.  The accelerated action design that Steinway used (more lead closer to the balance rail) does help somewhat with the speed of return of the key to the rest position which, along with the balance rail bearing, which was the goal of that system.  It may be noteworthy that Steinway abandoned that method presumably because the marginal benefits weren't enough to counter the production issues, or the removal of the shoe of the key required to use the balance rail bearing, resulted in more key flex and lower levels or "action saturation".  Action saturation, for those who don't now, refers to the amount of flex in the keys and the point at which greater force at the front of the key doesn't result in greater acceleration at the back of the key (or in this case from the capstan position) as a consequence of a key that is too flexible.  

When leading from scratch, I have the first lead always in the same position until no front location is required.  That is typically about 25 mm (1 inch) or so behind the head of the naturals.  The second and third leads should start out equally spaced with the rearmost lead moving back toward the balance rail as the need for its contribution diminishes along with the SW.  This will create a uniformly graduated pattern.  I think trying to figure out some optimum placement for purposes of inertia is not worth the trouble or time.  The benefits are marginal.  The leads contribute a relatively small amount to the overall inertia and there isn't a reasonable substitute at this point.  You will need to place lead in the keys no matter what, so your choices are somewhat limited.  

If interested in pursuing this further you might find some articles by Gravagne but I would probably talk to Dean Reyburn since they are making new keysets and offering preleading for those sets to find out what his reasoning is. I have talked with Gravagne about this in the past and I won't try and restate that conversation but it led me (no pun intended) to abandon the "accelerated action" positioning and focus more on the FW:AR:SW:BW relationships simply targeting a reasonable and smoothly progressing pattern for the leads.  

Keith, so with the change to CF shanks what was the strike weight reduction in the bass and treble?  Sounds like ~2.5 grams in the bass which would more than offset the assist spring contributing which is typically set around 10 grams. 

Seems like you might have been able to get away with not even changing the shanks with a SW reduction of that amount. 

Generally speaking, in terms of tone, mass is a benefit in the bass and a detriment in the treble. 

The actual pattern is less important than the relationship between SW and AR which will result in the amount of lead as an indicator of that relationship (see my thread on "Spreadsheet for Renner KMD for one discussion of that).  That relationship is something you may have to deal with in removing the assist springs.  The fact that you report 4-5 leads in the lower part of the piano each weighing 17 grams is of concern (depending on the resulting BW) because it suggests that the AR is quite high.  

So, once you establish the AR:SW relationship and the SW curve that in combination with the AR produces inertia that is in the sweet spot and results in a reasonable amount of lead in the keys then you can decide on pattern.  

Lead patterns should have some uniformity and consistency in how they progress through the keyset.  The classic 3-2-1-0 of the early Steinway pianos is a pretty good goal but not always achievable with modern hammers that can be a bit heavier.  4-3-2-1-0 pattern is likely to fall in the realm of acceptable as I outlined in that thread which is 80 - 85% of Stanwood's FW maximums for a medium BW target, say 36 - 38 grams.  That will verify an optimum relationship between AR and SW.  I'm using 1/2"  leads that weight about 14 grams each. 

As far as the actual pattern, in terms of overall inertia it doesn't make that much difference.  More leads located nearer the balance rail will result in slightly lower inertia than fewer leads toward the front of the key, all things being equal.  But the difference is marginal.  The accelerated action design that Steinway used (more lead closer to the balance rail) does help somewhat with the speed of return of the key to the rest position which, along with the balance rail bearing, which was the goal of that system.  It may be noteworthy that Steinway abandoned that method presumably because the marginal benefits weren't enough to counter the production issues, or the removal of the shoe of the key required to use the balance rail bearing, resulted in more key flex and lower levels or "action saturation".  Action saturation, for those who don't now, refers to the amount of flex in the keys and the point at which greater force at the front of the key doesn't result in greater acceleration at the back of the key (or in this case from the capstan position) as a consequence of a key that is too flexible.  

When leading from scratch, I have the first lead always in the same position until no front location is required.  That is typically about 25 mm (1 inch) or so behind the head of the naturals.  The second and third leads should start out equally spaced with the rearmost lead moving back toward the balance rail as the need for its contribution diminishes along with the SW.  This will create a uniformly graduated pattern.  I think trying to figure out some optimum placement for purposes of inertia is not worth the trouble or time.  The benefits are marginal.  The leads contribute a relatively small amount to the overall inertia and there isn't a reasonable substitute at this point.  You will need to place lead in the keys no matter what, so your choices are somewhat limited.  

If interested in pursuing this further you might find some articles by Gravagne but I would probably talk to Dean Reyburn since they are making new keysets and offering preleading for those sets to find out what his reasoning is. I have talked with Gravagne about this in the past and I won't try and restate that conversation but it led me (no pun intended) to abandon the "accelerated action" positioning and focus more on the FW:AR:SW:BW relationships simply targeting a reasonable and smoothly progressing pattern for the leads.  

All of this is immensely helpful.  Thank you.

Keith, so with the change to CF shanks what was the strike weight reduction in the bass and treble?  Sounds like ~2.5 grams in the bass which would more than offset the assist spring contributing which is typically set around 10 grams. 

Seems like you might have been able to get away with not even changing the shanks with a SW reduction of that amount. 

Generally speaking, in terms of tone, mass is a benefit in the bass and a detriment in the treble.