It is currently Thu Mar 28, 2024 3:12 am


Tuning - How Accurate?

Existing and forthcoming Hauptwerk instruments, recommendations, ...
  • Author
  • Message
Offline

Jon Hammond

Member

  • Posts: 133
  • Joined: Wed Mar 19, 2003 6:06 am
  • Location: Fairfield Glade, TN, USA

Tuning - How Accurate?

PostWed Nov 02, 2005 2:31 am

To what accuracy should an organ be tuned? For large ensemble tone, is tuning the pitches of each pipe to exact frequency the goal or it is better to allow a range of mistuning to obtain a richer sound? The software tuner I am using is G-tune ver. 2.51. I have measured the frequencies of one commercial sample set, one-half of another and several ranks of a third for comparison. About 2000 samples in all. Changing G-tune's bandpass filters and damping factor per each note yeilds a stable frequency reading accurate to +/- 0.02 Hz with a high limit of 5100 Hz for complex waveforms. In the sample sets I have checked, it is common to see notes of the same key with frequencies that range +/- 2 Hz for frequencies above 1000 Hz. Retuning should clarify ensemble tone of these sample sets. Audsley opines that tuning should be as accurate as possible. But, he didn't have the electronic tools now available. Certain organ synths detune ranks to obtain a richer sound, reeds - sharp, flutes - flat. What would an experienced organ tuner recommend? Any opinions to my question are welcome before I retune/change sample rate/re-save 4000 samples.
Offline
User avatar

giwro

Member

  • Posts: 857
  • Joined: Tue Aug 12, 2003 11:22 am
  • Location: Twin Cities, Minnesota USA

not an experienced tuner, but...

PostWed Nov 02, 2005 11:45 am

IMHO I always like several days/weeks AFTER my tuners have visited and the organ has "settled" a bit. To me, a "dead-on" tuning is just a bit sterile and uninteresting...

I liken this a bit to the whole musical arena, where you will find that it is the little inconsistencies that make music come alive - for example orchestras in which the strings are all very CLOSE to the pitch, but tiny descrepancies add warmth to the tone.

As for Audsley, history has shown he was an armchair theoritician... never built an organ, and many of his ideas are flights of fancy. (that's not to say that he was unintelligent - he really was quite bright, just not a qualified organbuilder!)

I've had some folks remark about my Hauptwerk recordings: "the organ sounds magnificent, but it must be digital - it's too well-tuned"!

If you LIKE that sort of sound, tune away...

If you want a REALISTIC organ sound, leave the discrepancies...

Cheers,
Jonathan Orwig
Coon Rapids, Minnesota USA
http://www.evensongmusic.net
Offline

stevetil

Member

  • Posts: 128
  • Joined: Wed May 25, 2005 11:18 am
  • Location: Oregon, US

tuning

PostWed Nov 02, 2005 5:48 pm

I always tuned pipe organs as close to in tune as possible.
They drift anyway. Temperature and humidity affect
air density which affects tuning of flue pipes.

I "stretch" tuned analog electronic organs a small amount.
It depended alot on the acoustics of the building and the
total number of oscillators.

Our ears hear both volume and frequency on a logarithmic
scale. Doubling the volume of sound only increases the
apparent level by a modest amount. A change of frequency
by a few Hz means whole tones in the 32' octave but a
very tiny perhaps inaudible change in the 1' octave.

The Petersen tuner which I used to use had a dial which
allowed you to detune by fractions of a cent, where 1 cent
is 1/100th of a semitone. To get an idea of what that means,
if you know of a unified organ which derives the 2 2/3' and
1 3/5' pitches from a unison rank, well first of course make
sure that the unison pitches are properly tuned, and then
listen carefully to the mutation pitches which are derived
from the unison voices.

Listen carefully to the 4' and 2 2/3' stops together on single
notes. You can hear a slight waver because the 3rd harmonic
of the 4' is beating against the second harmonic of the
2 2/3' stop. The pitches are close, but not exact. If you
try this on a Hammond tone wheel organ, there is no beat,
because there are no natural harmonics, only near sine waves.
The error that you can hear is just under 2 cents.

Now try adding the 1 3/5' Tierce. You should hear more
of a quiver than a waver because the error is more like
17 cents, at least in equal temperament. A derived Quint or
Nazard is noticible but frequently acceptable, but a derived
Tierce grates on my ears.

Acoustics can be a great help. Reverberation can cover up
a multitude of sins. A dry room is bad for organ music and
singing for a number of reasons. Of course, the very best
music is made on high quality instruments in good rooms.

Cheers, Steve.
Offline

Jon Hammond

Member

  • Posts: 133
  • Joined: Wed Mar 19, 2003 6:06 am
  • Location: Fairfield Glade, TN, USA

PostWed Nov 02, 2005 8:06 pm

Thanks to you both for your information. Hopfully they'll be more information before I summarize so that this enquiry can fade away. When creating celestes and unda maris stops, intentional detuning is done according to prescribed beat rates against in-tune pipes. Pipe organs do change pitch with thermal cycling. Pipes lose tuning at different rates or the tuning sleeve slips. One HW sample set that should be tuned to 440.0 is closer to 441.7. Is that because it was sampled on a hot day? The coefficient of expansion of metal pipes and wood pipes are very different. Are the organs sampled at the same temperature at which they were tuned? Hauptwerk samples do not season at all like a newly tuned pipe organ. Indeed, it is a good thing to purchase an HW organ possessing both accurate room acoustics and as-found tuning. However, my work assessing the measured pitches of some of those tunings reveals that many pipes appear to deviate far from calculated pitch and cleaning up the tuning should clarify the tone of the sforzando ensemble. I do possess a default answer to my question. The 20 stop Ahlborn modules use very short pipe samples and calculated temperament pitches that are modified with factory default deviations individually selectable for each stop. I have charted those numbers. Their words are "pitch adjustments to create various acoustic undulations between stops in order to avoid an unnatural sound activity of the ensemble". These default deviation settings are one manufacturer's opinions. It would be better to source such information from within the hauptwerk forum!
Offline

stevetil

Member

  • Posts: 128
  • Joined: Wed May 25, 2005 11:18 am
  • Location: Oregon, US

tuning

PostThu Nov 03, 2005 8:54 am

Two more little tidbits.

Tuning variations by temperature are much more dure to the
effect of changing air density within the pipe than anything
to do with the slight expansion or contraction of the pipe
material. Reed pipes vary far less with temperature than flues.

Celestes are usually detuned sharp, but with several celestes it
is nice to have at least one tuned flat. 3 rank celestes are not
unknown. I have also found a 2 rank celeste with one rank tuned
slightly sharp and the other slightly flat.

When I was usuing an electronic tuner I used a sliding scale
with about 19 cents sharp at 8' C, 16 cents at 4' C, 13 cents at 2' C,
and going on down to about 4 cents at top C (note 61) and 3 cents
at note 73 if it went that high. People seemed to like this. Of course,
multiple celestes would all be tuned a little differently.

Steve.
Offline

Jon Hammond

Member

  • Posts: 133
  • Joined: Wed Mar 19, 2003 6:06 am
  • Location: Fairfield Glade, TN, USA

PostThu Nov 03, 2005 9:39 am

Steve - Thank you for the rules of thumb from your experience. That's what we need to know. I suspect you have tuned brand-R similar to my Trio321 with two oscillator sets. How would you tune this organ? Would you stretch the tibia notes in view of the 5-1/3 and 2-2/3 stops? How much stretch - .5, 1.0, 1.5 cents per octave or just to reduce beat notes for quint and nassat? Is the main oscillator also stretched? Since my three keyboards are fitted with the electronic guts from Fatar Studio 610 midi controllers, the old electronics have not been used for over a year. I had tuned this organ against a sine wave output from Yamaha synth. I'd like to re-tune it.
Offline

Martell

Member

  • Posts: 13
  • Joined: Sat Sep 24, 2005 4:01 pm

Retuning samples

PostTue Nov 08, 2005 6:26 am

Do/can we hear sounding differencies between original and retuned organ-samples? (I do not ask for pitch-change)
Offline

Jon Hammond

Member

  • Posts: 133
  • Joined: Wed Mar 19, 2003 6:06 am
  • Location: Fairfield Glade, TN, USA

PostTue Nov 08, 2005 7:20 pm

Hauptwerk pipe samples that are retuned slightly sound the same when played singly, unless the hearer has perfect pitch recognition ability. Maximum change of 1%. For those interested in organ pipe pitch change with temperature and humidity, here is a good article... http://mmd.foxtail.com/Tech/soundspeed.html
Offline

Jon Hammond

Member

  • Posts: 133
  • Joined: Wed Mar 19, 2003 6:06 am
  • Location: Fairfield Glade, TN, USA

PostTue Nov 22, 2005 2:27 am

Purchasers of the Marian Silberman sample set may read in the vendor literature that an evangelical music committee recommends a future retuning of the organ to Kirnberger or other temperament. The sample set is sold with the understanding that it is not digitally retuned. One year ago I retuned my Marian to equal temperament. It retains its sweet sound and works well against the reeds in the SS sample set. However, the instrument deserves to be tuned to a well tempered tuning, as recomended. That would not be Larips so that I'm not visited in my sleep by the spectre of G.S. A week ago I had decided to adopt the Wegscheider temperament used by Prof. Maier to excellent effect for the 1731 Silberman. This week, my research reveals a different path.

Recent organ news about the 2007 AGO convention mentions a dual temperament organ built by Pasi. Further research reveals that the Pasi opus 14 organ evolved from other organs by Fisk and Wegscheider. Pasi op 14 is a 55 stop tracker using a Wegscheider well temperament, but also offers 29 ranks tuned at 1/4 meantone. Zip at equal temp! This requires 20 pipes per octave per 29 stops. The builders' web site leads one to Wegscheider's tuning spreadsheet. Further research reveals that there are three versions of 1/4 meantone usable. This dual temper arrangement enables the organist to lead congregational singing using meantone for music composed prior to 1854 and a well temperament for more modern music and then enables changing temperament on the fly. I doubt that HW2 can do that.

Therefore, I plan to tune the Marian SB16, P8, O4 and G8 on pitch - 20 pipes per octave. I will add the Regal 8 from the SS sample set - tuned on pitch - 20 pipes. I plan an unda maris 8 tuned + 2Hz using the P8. The mixtures will be tuned on pitch to the fundamental of the first overtone of the mixture. The balance of stops will be tuned similar to unda maris but limited to a predetermined beat range from 1/5 to 1/2 beats per second. This will limit all pipe frequencies, including those above 1000 Hz to a beat range below the unda maris beat frequency. It is easy to calculate these frequencies within the cell equations of the spreadsheet. Demos available in a month or so.

Return to Hauptwerk instruments

Who is online

Users browsing this forum: No registered users and 25 guests