Pianotech

Expand all | Collapse all

Steinway B circa 1988 - 2 1/2 questions

  • 1.  Steinway B circa 1988 - 2 1/2 questions

    Posted 02-18-2018 14:01
      |   view attached
    I'm resisting the urge to cross post to multiple lists, and hoping
    I'll get some knowledgeable responses here.
    I've been working with a 1988 Steinway B (#507...). Piano has been
    rebuilt. Attached photo attempts to illustrate two of the three questions

    1) Specific to this piano, it would appear that tenor agraffe holes
    were drilled deep enough so that the level of the 'shelf' (flat part
    of top of agraffe base) is submerged beneath the surface of the
    plate, making the string level of note #21 about 3/16" lower than
    note #51, the last note of the agraffe section. (Tomorrow I could
    get the actual height numbers.) Has anyone ever observed this on
    Steinway B (or any other piano)? What, if anything, did you do?

    2) Apologizing for non-technical terminology, is the profile of the
    string rise in front of the agraffes in the tenor normal for B's of
    this (or any) period? Currently, I have access to a B from the
    1920's, a D, and a few A's. None has this 'feature', although the
    other B has more of it than the other models. I've never paid
    attention to this before, and might have overlooked it this time,
    except for having seen a stellar example of this same issue in small
    Knabe grand I inspected a few days before. It seems like such an
    unfortunate design, both from the perspectives of tuning and
    structural stress on agraffe and plate. Has anyone tried to mitigate
    the severity of the angle?

    1/2) Rear duplex placement - Is there any source of 'correct'
    measurements (or process) for duplex placement? This is a 1/2
    question as I'm currently trying to review the 1872 patent to try to
    understand what the original intent was, but I suspect there is quite
    a bit of more current information, both theoretical and practical...
    short of someone just telling me.

    Thanks -
    David Skolnik
    Hastings on Hudson
    davidskolnik@optonline.net

    ---
    This email has been checked for viruses by Avast antivirus software.
    https://www.avast.com/antivirus


  • 2.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-18-2018 14:07
    One thing might be to measure the string height from keybed, across the scale.
    Is this an attempt to compensate for some structural irregularity?

    ------------------------------
    Ed Sutton
    ed440@me.com
    (980) 254-7413
    ------------------------------



  • 3.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-18-2018 14:54
    Ed -
    As I mentioned, I'll be at the piano again tomorrow (though I really just need my notes, as I've already measured) and can provide precise numbers.  For your theory to make sense, though, the string height from #21 to 51 would have to have ended up fairly consistent, which is not the case.
    I'd be interested in your thoughts on the 'up-angle' of the plate in front of the agraffes.  You would be doing us all a great service even if you could just provide the correct name for this part of the plate structure.
    Thanks​

    ------------------------------
    David Skolnik
    Hastings-on-Hudson NY
    914-231-7565
    ------------------------------



  • 4.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-18-2018 15:37
    Hi, David,

    I'm on the road; and cannot explain more fully at the moment.

    The short version is that the string "plane" in an S&S is not a plane.   It's a compound arc with its highest point at note 62.   This is not terribly clear in the patent drawings.

    I'll be home later this afternoon/evening; and will try to get something more detailed out to you then.

    Thank you very much.

    Kind regards.

    Horace

      Original Message




  • 5.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-18-2018 15:59
    Hi, David,

    Before I forget, would you please send along whatever patent(s) you've been looking at?   I'll troll through my collection to see if I might have anything useful.

    Thanks very much.

    Kind regards.

    Horace
      Original Message




  • 6.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-18-2018 20:00
    Hi Horace -
    The patent I'm looking at is #126,848 May 14, 1872 Improvement in Duple Agraffe Scales for Piano-Fortes - C.F.T. Steinway
    Patent Images
    Just got back from Shape of Water, so I still need to sit with it, though, rough glance, it seems to get quite specific in it's subdivisions.
    Don't forget to comment on the plate issue as well, assuming you have some thoughts.
    Regards -
    David S

    ------------------------------
    David Skolnik
    Hastings-on-Hudson NY
    914-231-7565
    ------------------------------



  • 7.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-18-2018 16:03
    Hi, Ed,

    No...it's part of the design.

    Kind regards.

    Horace
      Original Message




  • 8.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-18-2018 18:16
    "It's not a fault, it's a feature."

    ------------------------------
    Regards,

    Jon Page
    ------------------------------



  • 9.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-18-2018 21:03
    Greetings,
       It is the norm for note 21 on a Steinway B to be significantly lower than note 52.  2-3 mm is not abnormal in my records.   I have heard the expanse of understring felt that the string rests upon between agraffe and tuning pin called the "apron", "string shelf", and "string rest".  I don't know if it really has a specific name.
    Regards,

    ------------------------------
    Ed Foote RPT
    ------------------------------



  • 10.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-19-2018 03:48
    Hi,

    My apologies for the delay...it's been a long day.

    Anyway, here is some more information:

    String height as noted, is critical; and, there are many possible
    variations.

    From notes that I used to prepare a forefinishing class I used to teach
    and from a class done by Michael Mohr some years ago, string height is
    generally measured at note #62, with the optimal numbers being:

    - Models S through B - 7.5"; and,
    - Models C and D being - 7-5/8th"

    These are, however, only very general guidelines as the casting of the
    individual plate varies based, in part, on which of the several wooden
    patterns were used to create the sand mold into which the iron was cast.
    Other important variables include how the plate was fit into the case,
    which, in turn, is based on the actual contours of the seasoned, bent
    rim. Important to remember that the variable of plate fitting a crucial
    one which is often not properly set in the factory, let alone by many
    rebuilders. Especially for a sole proprietor, while taking time with
    this step can be a real PITA, getting the plate in at the proper height
    saves oceans of grief with every aspect of getting the piano to work
    properly thereafter.

    Also, as Ed Foote notes, it is normative for the strings of the lowest
    note of the low tenor on any given scale to be anywhere from 2 to 4mm
    lower than those at note 62. Similarly, it is not unusual to find the
    strings of note 88 to be around 2mm (or so) lower than those of note 62.

    In spite of efforts to bring production between NY and Hamburg into
    greater continuity, manufacturing processes in Hamburg are still
    generally more advanced than those in NY, with the result that there
    continues to be more instrument-to-instrument variation on this side of
    the pond. There's a long story about all of this...for another time.

    Obviously, the "rebuild" introduces many additional variables. I'm not
    sure that I'd be too worried about the below-plane agraffes as long as
    the bearing wasn't excessive for the board, and/or, the string height
    for those last few agraffes didn't wind up more than 4 or 5mm below the
    7.5" (plus or minus...depending on how closely to the optimum number the
    plate was set to begin with).

    Among the problems in doing patent research is that the potential
    patentee (as it were) is often justifiably worried about whether or not
    their R&D work might be stolen by others. So, it's not unusual to have
    to dig through the text carefully for clues from which to assemble some
    reasonable insight into the actual intent of the design submitted.

    One very good example of this are the two main patents associated with
    the (eventual) patenting of the accelerated action...#1,826,848
    (13Oct1931) and 2,031,748 (25Feb1936). Both patents are the eventual
    product of engineering development begun in the mid-1920's, and in
    production (at least in the Model D) at least as early as 1929. The
    specific example of the above noted kind of writing appears on Page 2 of
    the application for the earlier patent (1,826,848), lines 110 through
    119, inclusive. I'm attaching copies of both patent applications...with
    any luck they'll come through.

    OK...this is not as much detail as I had hoped to have. I'm presently
    kicking off a restore from tape of several archived areas of my notes to
    see if I cannot find more relevant documentation for some of the patents
    which come into play with all of this. I know that I put all of those
    documents in a Very Safe Place...now, of course, all I have to do is to
    remember where that is. My sincere apologies. I thought that this
    stuff was more immediately to hand. Maybe Bill Shull has something more
    readily available.

    In the meantime, I hope that this is of some use.

    Kind regards.

    Horace





    On 2/18/2018 6:02 PM, Edward Foote via Piano Technicians Guild wrote:
    > Please do not forward this message due to Auto Login.
    >
    > Greetings,
    > ?? ??It is the norm for note 21 on a Steinway B to be significantly lower than note 52. ??2-3 mm is not abnormal in my records. ?? I have heard the expanse of understring felt that the string rests upon between agraffe and tuning pin called the "apron", "string shelf", and "string rest". ??I don't know if it really has a specific name.
    > Regards,
    >
    > ------------------------------
    > Ed Foote RPT
    > ------------------------------
    > -------------------------------------------
    > Original Message:
    > Sent: 02-18-2018 18:15
    > From: Jon Page
    > Subject: Steinway B circa 1988 - 2 1/2 questions
    >
    > "It's not a fault, it's a feature."
    >
    > ------------------------------
    > Regards,
    >
    > Jon Page
    > ------------------------------
    >
    > Original Message:
    > Sent: 02-18-2018 16:02
    > From: Horace Greeley
    > Subject: Steinway B circa 1988 - 2 1/2 questions
    >
    > Hi, Ed,
    >
    > No...it's part of the design.
    >
    > Kind regards.
    >
    > Horace
    > ?? Original Message
    >
    >
    > Original Message------
    >
    > One thing might be to measure the string height from keybed, across the scale.
    > Is this an attempt to compensate for some structural irregularity?
    >
    > ------------------------------
    > Ed Sutton
    > ed440@me.com <ed440@me.com>
    > (980) 254-7413
    > ------------------------------
    >
    >
    > Reply to Sender : http://my.ptg.org/eGroups/PostReply/?GroupId=43&SenderKey=3db543d7-e7c0-4282-8814-33280ad72b8a&MID=686228&MDATE=756%253d45746%253d&UserKey=3feecf45-4a69-4cff-bbb2-fd6c7eaf0569&sKey=KeyRemoved
    >
    > Reply to Discussion : http://my.ptg.org/eGroups/PostReply/?GroupId=43&MID=686228&MDATE=756%253d45746%253d&UserKey=3feecf45-4a69-4cff-bbb2-fd6c7eaf0569&sKey=KeyRemoved
    >
    >
    >
    > You are subscribed to "Pianotech" as hgreeley@sonic.net. To change your subscriptions, go to http://my.ptg.org/preferences?section=Subscriptions&MDATE=756%253d45746%253d&UserKey=3feecf45-4a69-4cff-bbb2-fd6c7eaf0569&sKey=KeyRemoved. To unsubscribe from this community discussion, go to http://my.ptg.org/HigherLogic/eGroups/Unsubscribe.aspx?UserKey=3feecf45-4a69-4cff-bbb2-fd6c7eaf0569&sKey=KeyRemoved&GroupKey=2bb4ebe8-4dba-4640-ae67-111903beaddf.
    >


    Attachment(s)

    pdf
    US2031748.pdf   667K 1 version
    pdf
    US1826848.pdf   585K 1 version


  • 11.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-19-2018 09:36
    Edited by Jim Ialeggio 02-19-2018 09:36
    The relative length of backscale, string to string, determines how even or smoothly graduated DB loads can be, string to string. So placement and design of the rear aliquots are often intended to achieve equal, or smoothly trended DB loads. Even so, with hitches sharing strings, achieving this level of evenness remains beyond the scope of reality, in my observations. This can be more accurately addressed with adjustable hitches. However, at what point this evenness is valuable or or not, or at what point load is generally in the ballpark, though uneven, is not clear.

    ------------------------------
    Jim Ialeggio
    grandpianosolutions.com
    Shirley, MA
    978 425-9026
    ------------------------------



  • 12.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-19-2018 10:31
    Hi Jim -
    Happy Presidents Sales Day.
    I'm going to be minimal in my response until I re-read some of your previous contributions to this subject, going back a few years.  I suspect I'm not paying enough attention.
    Short version for now: Given that we were speaking of Steinway, and the 'correct' placement of rear duplexes, I don't think that exploring the potential benefit of adjustable hitch pin is particularly relevant.  Now, if you could come up with an adjustable height aliquot, well then, that might make sense.  From the little I've observed of the process of establishing bridge height in a new board installation, the position of the duplex should already have been established.  The basis of that positioning is what I think I'm asking.  Since, in the case of Steinway, we're not dealing with individually adjusted aliquots, I assume there must be some blending of theoretical and practical.  I think I'm first asking about the theoretical.

    Happy shopping.

    ------------------------------
    David Skolnik
    Hastings-on-Hudson NY
    914-231-7565
    ------------------------------



  • 13.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-19-2018 11:09
      |   view attached
    David,

    I have attached the factory spec sheet on this given me by Frank Hanson many years ago. It has been annotated and corrected over time. I have noted when and by whom these corrections were made.

    Hope it helps.

    Pwg

    ------------------------------
    Peter Grey
    Stratham NH
    603-686-2395
    pianodoctor57@gmail.com
    ------------------------------



  • 14.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-19-2018 11:15
      |   view attached
    Perhaps this photo is more readable.


    ------------------------------
    Peter Grey
    Stratham NH
    603-686-2395
    pianodoctor57@gmail.com
    ------------------------------



  • 15.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-19-2018 12:07
    Thanks Peter.  I was about jump into the car to head up there since I couldn't read even the 2nd version, but I managed to make it work and am grateful.  You must have quite a trove.  Looking forward to checking out the piano.
    Thanks again

    ------------------------------
    David Skolnik
    Hastings-on-Hudson NY
    914-231-7565
    ------------------------------



  • 16.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-19-2018 12:32
    Theoretically, placement and design is trying to even out the length of each string's backscale relative to its neighbors and within its own unison. If you look at the staggered placement of hitches plus the shared hitches, this theoretically "even" backscale is geometrically impossible to achieve with precision. So, in my view, the aliquot is trying to get within a reasonable tolerance of "even" or "graduated".

    As far as placement in a restring, one locates the original position with locator pins before disassembly. As far as whether the pre-disassembly position may or may not be "correct", the location and sweep of the aliquot is defined, within a small range of adjust-ability as "place it where the backscale strings of the relevant unison actually sit on the aliquot rest for that unison. Given the sweep of the aliquot, one really has a limited amount of choices where the aliquot is located properly to contact its relevant unison's strings. Occasionally, some string's backscales still miss the aliquot.

    This means, aliquot placement, as far as backscale precision, is by definition very general. Claims to any precision exceeding what the geometry will allow, go counter to the geometric possibilities.

    ------------------------------
    Jim Ialeggio
    grandpianosolutions.com
    Shirley, MA
    978 425-9026
    ------------------------------



  • 17.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-20-2018 02:30
    Jim said:
    This means, aliquot placement, as far as backscale precision, is by definition very general. Claims to any precision exceeding what the geometry will allow, go counter to the geometric possibilities.

    I guess the question would be whether,  in the absence of absolute precision, is there any acoustic benefit to being 'close', or conversely, is there a range, beyond which, one might discern a negative impact?

    Using the data provided by Peter Grey, (Many thanks Peter)  I measured the rear duplex distances on this B and found these differences:
    40) +10.7mm
    51) +10.3mm​

    52) +.6.35mm
    68) - 5.60mm

    69) + 4.35mm
    88) + 6.40mm

    It seemed that, in this case at least, it would, in fact, be possible to manipulate the three duplexes to conform to both the distance reaquirements AND the alignment.

    ------------------------------
    David Skolnik
    Hastings-on-Hudson NY
    914-231-7565
    ------------------------------



  • 18.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-20-2018 02:11

    Sent: 02-19-2018 09:35
    From: Jim Ialeggio
    Subject: Steinway B circa 1988 - 2 1/2 questions

    The relative length of backscale, string to string, determines how even or smoothly graduated DB loads can be, string to string. So placement and design of the rear aliquots are often intended to achieve equal, or smoothly trended DB loads. Even so, with hitches sharing strings, achieving this level of evenness remains beyond the scope of reality, in my observations. This can be more accurately addressed with adjustable hitches. However, at what point this evenness is valuable or or not, or at what point load is generally in the ballpark, though uneven, is not clear.

    I may have sabotaged myself by trying to combine more than a single topic, which inevitably proves to be far from 'single'.  I'll do what I can.
    Jim
    Other than the Baldwin Accujust hitch pin, and a later Steinway patent (#1,972,511), I'm not sure what documentation you might have supporting your conjecture about the height-adjusting function of the rear duplex.  (To be clear, I believe that the back portion (bridge to hitch) is muted and not intended to function as a resonating duplex.)  I was surprised that there appear to be no serious updates to the original 1872 (#126,848).  In the Steinway book by Roy Kehl and David Kirkland, there are a total of three references to 'duplex', however the second 1878 (#204,109) is actually a 'duplex' agraffe, for models A,B & C, and the third, 1934 (referenced above), a design by Paul Bilhaur (unclear whether it was ever implemented, even experimentally) was intended to be adjustable to modify height.


    ------------------------------
    David Skolnik
    Hastings-on-Hudson NY
    914-231-7565
    ------------------------------



  • 19.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-20-2018 09:16
    I recall Joe Bisceglie talking about the general attitude at the factory about duplex scale positioning which was that it really didn't matter too much since somewhere in the overall partial structure of the piano there would be some partial matching and thus reinforcing no matter where the things were placed.

    I am sure there are many that would strongly disagree with that. I try to set them according to the spec sheet if at all possible. The one I'm currently working on was significantly off from the specs.  I have no hard data to support or deny JB's claim.

    Pwg

    ------------------------------
    Peter Grey
    Stratham NH
    603-686-2395
    pianodoctor57@gmail.com
    ------------------------------



  • 20.  RE: Steinway B circa 1988 - 2 1/2 questions

    Posted 02-20-2018 09:30
    Edited by Jim Ialeggio 02-20-2018 09:36
    <I'm not sure what documentation you might have supporting your conjecture about the height-adjusting function of the rear duplex. 

    Its not a conjecture. I'm not referring to documentation in patents, as patents have no responsibility to refer to geometrical fact. The line between marketing and function are often indistinguishable, and thus simply unnecessarily make this harder to think about than it needs to be.

    If DB load is the goal one is trying to address, one manipulates this load by targeting a specific included angle of positive deflection as the string traverses over the bridge to the rear termination. The calc is simple:   Sine (of the included angle over the bridge) x tension (of that particular string in lbs)= DB load (lbs). I originally had a whole page of calcs to do what this simple Sine x tension calc, from vector something or other accomplishes. Some fine soul posted this formula years ago.

    Since that included angle defines DB (lbs), how does one manipulate the included angle of deflection?

    There are four parameters one can mess with. To increase the included angle, one can 1- raise the bridge height, 2-change (lower) the plate elevation,  3- shorten the hitching length, or 4- lower the height of the aliquot rest.  One can use one or all of these parameters individually or together to adjust the included angle, and thus the DB load of that string.

    Aliquots assign a specific rest height and per-unison hitching length to define the included angle over the bridge. Adjustable hitches, since they have a relatively large range of possible heights, can use the height alone to achieve the included angle, even if the hitching lengths are unequal...that's the big advantage of adjustable height hitches...unequal hitching lengths can be compensated for on the fly.   This is also the reasoning behind adjustable plate heights by plate bolts or Steinway's mid long bridge nose bolt adjustments. Eliminating 1-bridge height (which has been committed already), leaves you with 2-plate height 3-hitching length and 4-height of the aliquot rest. Steinway uses all 3 of these in their design. Plate height, at least in mid-long bridge is adjustable, and their aliquots are designed to address 3 and 4. .

    As far as documentation of a different sort, from Mason, who uses aliqouts to achieve the goals I've outlined, Bruce Clark is very clear that the design of their aliquotes is to even the DB load, and is very proud of the level of consistency they have achieved. Mason has the capability to adjust plate bolts at the perimeter but does not, as far as I am aware mess with mid long bridge plate height adjustability.

    Whether absolute consistency or smoothly trended DB is beneficial or not,  or at what point it is "good enough" is another harder question, which I continue to test the boundaries on. 


    ------------------------------
    Jim Ialeggio
    grandpianosolutions.com
    Shirley, MA
    978 425-9026
    ------------------------------