Non-compensating valve instruments

2nd tenor

Well-Known Member
Non-compensating valve instruments are typically lighter and cheaper than those with compensating valves, and in the right hands are also capable of producing very acceptable music - I've never seen a trumpet with compensating valves and some very expensive orchestral tubas don't have then either.

What problems have players noticed with non-compensating valve instruments and how do they (or would they) work around those problems to still use the instruments to good effect?
 
Last edited:

pbirch

Active Member
Non-compensating valve instruments are typically lighter and cheaper than those with compensating valves, and in the right hands are also capable of producing very acceptable music - I've never seen a trumpet with compensating valves and some very expensive orchestral tubas don't have then either.

What problems have players noticed with non-compensating valve instruments and how do they (or would they) work around those problems to still use the instruments to good effect?
the physics is somewhat complex (and way beyond me) but some valve combination are a little sharp, and so with the addition of extra tubing can be brought into tune,
there are examples around of compensated cornets made in the early 20th C but they didn't really catch on, but most trumpets overcome this problem with adjustable 1st and 3rd valve tuning slides and so "compensate" for the sharpness that way
 

MoominDave

Well-Known Member
Fear not, it's not that complex; all there is to it is this:

1) You have an instrument that has valve slides tuned to lower by set amounts from the basic instrument pitch.
2) When you put a valve down, you're no longer in that basic instrument pitch.
3) So when you add another valve to it tuned to that basic instrument pitch, you're not adding enough tubing. So combinations of valves go sharp.

An example with made-up numbers:
Say you play Eb bass. 1st valve loop is (say) 50 cm long, lowering the bottom C by a tone. 3rd valve loop is (say) 70 cm long, lowering the bottom C by a tone and a semitone.
You put down 1st valve and play a low Bb.
If you then add 3rd valve to try to reach a bottom G, the 70 cm that was long enough to lower low C to low A is not quite enough to lower low Bb to low G - you need (say) 80 cm. That extra 10 cm or so is what the compensating system adds.

Did that help at all?

Having a compensating system is like having an automatic gearbox - you don't have to think about it at all (sort of... Compensation doesn't sort out every combination.). Uncompensated instruments are like manual shifts - you have to be thinking and planning all the time with adjustments. You get some advantages of response with both a manual gearbox and an uncompensated instrument, but at the cost of having to do more things as you go and needing a greater level of skill to do so to advantage.
Smaller instruments don't tend to compensate because the valve loops grow too small and fiddly. Instead, cornets are supplied with valve triggers to do the same job. At this point, I invite you to recall quite how many times you've heard your 2nd and 3rd cornets play low Ds and C#s out of tune... Compensation would undoubtedly result in fewer tuning errors on these notes.
On the larger valved instruments in the lower register, the distance required for tuning corrections becomes enormous - feet of extra tubing required. An uncompensated 4V BBb bass wanting to play a low C# on 1+2+3+4 is short by most of a metre of tubing. So, different solutions tend to be favoured. Some orchestral players favour uncompensated tubas for their less stuffy response in the low register, and these instruments tend to have more valves than the usual 4 in addition to slides that you can pull in and out as you play. A common scheme is to add a 5th valve that when added to the 4th valve makes low F in tune (talking in band treble) - so 5th valve on its own is a long tone. This allows a chromatic register down through the pedals, but requires a certain amount of slide pulling. Low C# is not in bad tune on 1+2+3+4+5 under this scheme.
Having to pull slides as you go is unwieldy, and fast passages in affected registers become a nightmare. The compensating system is a pretty good compromise between tuning, response, and instrument weight. Good enough to still be the state of the art after being patented way back in 1874.
 

2nd tenor

Well-Known Member
Many thanks, Dave - you've made the whole business very clear! :)
As he often does :) , clever stuff and thank you DT.

On the other hand my original question, or at least part of it, seems to have been mostly overlooked somehow. The compensating systems come in two forms: three valve and four valve; nothing is compensated until the last valve is depressed. In other words, besides possibly Eb, nothing within or above the stave (95 % plus of the notes typically played?) is compensated by the very expensive system attached to the valves. Of the notes below the stave the bulk of them don't use the compensating system either 'cause the last valve isn't depressed.

I do appreciated that (lower) Eb, D, Dd, Ab, bottom G and bottom Gb are virtually destined to be sharp with a non-compensating instrument but wonder how a skilled (three valve) player would work around that fault in the instrument to produce music that is reasonably in tune. It must surely have been done in the past before the compensating system was so widely used and before bands could afford the considerable extra expense.
 
Last edited:

MoominDave

Well-Known Member
Glad I could be of help!

There's a couple more notes in the mid-to-upper register that are affected under the usual 3V compensating scheme - middle Ab and high Ab. Plus some occasionally useful alternative fingering options such as 2+3 for upper Eb - or 1+3 for high G, A, and B.

Workarounds? Well, pulling slides on the fly has a long history, and lower register notes are reasonably bendable in safety. The compensating system has been widespread in mid-to-high-level British brass banding for much of its life - a century or more. But equally many smaller bands opted for cheaper models. I suspect (based on what one hears in banding today in some places) that these notes have in many cases simply been tolerated out of tune. After a while of hearing Ds and C#s consistently sharp it becomes hard to recall that sharp is actually wrong...

Just Googling around for relevant curiosities...
Here's some rather nice pictures of an old Boosey compensating 3V soprano cornet (compensation on 3rd valve).
Note:
1) They did try this idea out on even the smallest of band instruments.
2) I got my date wrong above, per the inscription on the bell - compensating patent 1879, not 1874. Here's the actual patent, online (crikey, the internet's an amazing thing when used for good), including David Blaikley's rather beautiful technical drawings. See how little changed the layout is from his original design (look at a 3V Sovereign baritone say) - Blaikley was quite the bright spark.
3) The leadpipe goes into the 3rd valve, not the 1st. So the windway first goes through any compensating loops, then adds the main valve loops, rather than vice versa.
4) There is no 2nd valve compensation loop, due to its tiny size. Note how minuscule the 1st valve compensation loop is (4th picture row on the far right)- the 2nd valve one would be about half the size of that (not exactly, but about); it wouldn't have been practical to have one.

So this would have compensated only the 1+3 combination, a pretty rare fingering to find in sop playing. No wonder it didn't catch on. Though isn't it beautiful? The serial number on this instrument dates it to 1885.

Another system which was tried out in the early years of the 20th century was Besson's "Enharmonic" system. This was based on the same principle as the full double French horn, where, like the compensating system, there was a 'master valve' (i.e. the one that compensates), and then the other valves had two separate slides based on whether the master valve was depressed. The difference from the compensating system is that the extra slides were full length (normal+compensating) instead of only having the compensating bit. Then the windway only has to pass through each valve once. The main defect of this system, which offers as good tuning as the equivalent compensating set-up, but with reduced stuffiness on many-valved notes, is that it makes for heavy instruments! Here's 6 photos of a full enharmonic 4V euphonium (link is to first photo - click right for more). It looks bonkers... Surprised they didn't call this model "The Gorgon"...
The principle of the system can be seen more readily on the photos of a 3V equivalent that you can see if you click left from the same link; the windway goes first into the master valve (3rd), then pursues two alternative paths with alternative slides back around 1st and 2nd valves, chosen by depression of the 3rd valve before joining back together at the 3rd valve to head on to the bell.

There are quite a few little variations on these themes out there dating from these years. Must have been a great time to have worked in brass instrument design. Something I've never heard of having been done though is second-order compensation... The systems discussed only fix combinations of two valves at a time - even with a compensating system 1+2+3+4 is appreciably sharp. If you had a 3rd valve compensating subsystem within the 4th valve compensating system, that would fix most of the remaining error... But it's pretty clear why this hasn't been tried - it would be a mechanical and design nightmare of tubing options, tuning returns are rapidly diminishing at this point, and it would turn out very stuffy on these notes.
 
Last edited:

2nd tenor

Well-Known Member
Workarounds? Well, pulling slides on the fly has a long history, and lower register notes are reasonably bendable in safety. The compensating system has been widespread in mid-to-high-level British brass banding for much of its life - a century or more. But equally many smaller bands opted for cheaper models. I suspect (based on what one hears in banding today in some places) that these notes have in many cases simply been tolerated out of tune. After a while of hearing Ds and C#s consistently sharp it becomes hard to recall that sharp is actually wrong...
Using a four valve instrument as if it were a three valve one (i.e not using the fourth valve or any compensation) I found that setting the third slide to give a good (in tune) lower D helps a lot. IMHO the 123 combination isn't dreadfully sharp and the 23 combination not dreadfully flat, but others my find different (YMMV). Intonation on the (quality British) instrument seemed to vary anyway, some open notes being slightly flat or sharp relative to others and middle D being more in tune using the fourth valve rather then the first, etc. To get those notes (23 and 123 combinations) more spot on I'd be interested to learn more about how to lip a note into tune.

Like Dave I also suspect that smaller bands, and maybe those outside of the mid-to-high level bands too, did opt for cheaper non-compensating instruments. In theory that must have disadvantaged them relative to richer, and maybe better, bands. However in practice the primary purposes of a band are to provide pleasure for players and audience, so if both of those groups found the output acceptable them simple three valve non-compensating instruments must be perfectly adequate for some purposes. Doing a brief search on Salvation Army instruments supports that concept, in the pictures I found the bulk were non-compensating and likely played well by capable Bandsmen who worked around and compensated for the limitations of their instruments. What did such Bandsmen do?
 
Last edited:

MoominDave

Well-Known Member
Good point about practice not always matching theory exactly. For example, I'm always surprised by how low D and C# on the standard tenor horn layout tend not to be nearly as sharp as they are on other band instruments. They're still high, but the error is basically not intolerable to the ear, hence tenor horns not tending to be fitted with slide triggers until recent years, when the concept's become fashionable in banding.
Straightforward theory does tell us about the different tunings of different harmonics, e.g. upper E on open being flat and upper G on open being sharp - though not nearly so sharp as it actually goes in practice on various instruments.

I can't really offer anything more on what old-time quality players stuck with out-of-tune instruments actually did. I have observed from old recordings and present-day experience that tuning is often not the topmost priority in brass banding - culturally we do ensemble precision as an art form, but can be surprisingly lax on intonation, relying on the large ensemble to average out discrepancies. I think what we've touched on so far probably explains as much as there is to explain about how such players approached intonation. I'd also note that the use of these notes is comparatively rare, and that writers have always been aware of their deficiencies and sought to avoid them where possible. The bass section is where solutions really became vital first.

We can observe what bands were actually using. Gavin Holman at IBEW has undertaken the laborious but magnificent task of scanning in many old band photos. It's interesting to flick through these, noting what equipment was in use where, when, and at which level of playing - BBR has enough historical results that it's usually quite easy to deduce a band's approximate level in a given year.

Tuning 3rd valve flat to make either 2+3 or 1+3 in tune (or splitting the difference) is a common practice, useful also on a 4V compensating system if you wish to use these fingerings - and for when you use these combinations with 4th added. After all, we don't need to use 3 for A; 1+2 does fine apart from strange exceptions (like the common Sovereign baritone top A). Just running some calculations to illustrate how these uncompensated schemes affect this register (note that a cent is 1/100 of a semitone):

1) 3rd valve tuned to make low A in tune when used on its own
Ab 2+3: 16 cents sharp
G 1+3: 30 cents sharp
F# 1+2+3: 54 cents sharp
Under this scheme the Ab is (just about) tolerable, but the G isn't, and F# hugely isn't.

2) 3rd valve tuned to make low Ab in tune on 2+3
Ab 2+3: In tune
G 1+3: 16 cents sharp (note exactly the same as Ab in scheme 1)
F# 1+2+3: 39 cents sharp
Under this scheme the G comes into reach, but the F# still doesn't.

3) 3rd valve tuned to make low G in tune on 1+3
Ab 2+3: 16 cents flat
G 1+3: In tune
F#: 1+2+3: 25 cents sharp
Under this scheme the Ab is starting to drift a bit flatter than desirable, while the F# is still out of reach without bending.

To my mind the optimal scheme is one in between 2 and 3, that leaves Ab a little (but not appreciably) flat, G a little (but not appreciably) sharp, and leaves F# as a special case that has to be solved by other means (most probably lip bending or a slide trigger if available), as tuning to it would spoil too many other things.

Apologies for rambling on. I quite enjoy this subject.
 

2nd tenor

Well-Known Member
No apologies needed at all, I think that your lastest response is particularly helpful. Thank you.

Your comments on the higher E and G notes being out of tune are as surprise to me and some additional detail would be welcome. I have used alternate fingering and (Trombone) slide positions for both notes but didn't realise that they (the open notes) didn't slot that well; everyone is taught to play them open so one assumes that any error is down to oneself or the particular instrument you're playing .....

Suggesting adjustment of the third slide to make low Ab slightly flat and low G slightly sharp seems about right to me: a pragmatic choice :) . Low F sharps are exceedingly rare, I wonder whether they matter that much in the bigger picture of things and, if the note is held, might it be possible (if still difficult) to temporarily pull the first slide to get an in-tune note. In lower level bands perfect intonation is, IMHO, not really a proirity compared to the basics of getting folk to play what's written on the music and in time with the conductor; when that's done the result is pleasing enough for a local concert, fete or bands in the park event and that'll (mostly) do for me.

It would be interesting to hear how notes can be lipped into tune and whether that is more difficult between different types and brands of instrument - I suspect that it is more difficult.

Finally I'd encourage others to contribute their experiences to this thread, there's no need to be reserved or cautious as the mods do sort out folk that are too 'unfriendly'.
 
Last edited:

MoominDave

Well-Known Member
Your comments on the higher E and G notes being out of tune are as surprise to me and some additional detail would be welcome. I have used alternate fingering and (Trombone) slide positions for both notes but didn't realise that they (the open notes) didn't slot that well; everyone is taught to play them open so one assumes that any error is down to oneself or the particular instrument you're playing .....
Brass instruments are designed so that they produce the harmonic series of intervals on open fingering. This isn't an absolute (note the displaced pedal on the cornet), but holds pretty well in the main. The harmonic series are notes whose frequencies are in integer ratios to each other:

Pedal C: 1
Low C: 2
Middle G: 3
Middle C: 4
Upper E: 5
Upper G: 6
Way-flat upper Bb: 7
Top C: 8
Top D: 9
etc.

So for example middle G on an instrument has a frequency 3 times the pedal C on it. Top C has a frequency 8/5 times the upper E on it. These are the frequencies at which the pipe can be persuaded most easily to resonate.

We use equal temperament these days - we expect each semitone to be the same size around the chromatic scale. When we compare the interval sizes we get from the above to equal temperament, we see that we have the following theoretical tuning discrepancies:

Pedal C: In tune
Low C: In tune
Middle G: 2 cents sharp (1/50 of a semitone? Not noticeable to most)
Middle C: In tune
Upper E: 14 cents flat (this can be distinguished easily enough)
Upper G: 2 cents sharp
Way-flat upper Bb: 31 cents flat
Top C: In tune
Top D: 4 cents sharp
etc.

I stress that these are theoretical. Some instruments match them quite exactly, but others deviate from them. A common difference is for the upper G harmonic to be sharper by some way than this. Euphoniums and larger trombones seem vulnerable to that particular G problem, for some reason. Small bore trombones can have a tendency to go in exactly the opposite direction on 5th and 6th harmonics, which can confuse the brain if you play both large and small bore trombones... One needs to reverse all one's usual corrections.

But these provide a basic set of tuning expectations to approach an instrument with. I vividly recall my surprise when my trombone teacher at school corrected me on a passage when I'd played middle C# and upper E in the same 4th position. "Move the slide out half a cm as you go to the E from the C#", then "there, doesn't that sound better?". Yes, indeed it did.

It would be interesting to hear how notes can be lipped into tune and whether that is more difficult between different types and brands of instrument - I suspect that it is more difficult.
No magic here... Relax the embouchure to hit the note below the centre of the slot to lip it down, or tighten it to hit it above to lip it up. One usually has more wiggle room lipping down than up, so don't tune anything too flat.

How much lippage is possible comes down to how tightly the instrument 'slots' on that particular note, a function of harmonic, fingering, design, and lip strength. Anecdotally, trombones have become more tightly slotting over the years, which has upsides and downsides. But then, one should never be lipping up and down when the handslide will do the job for you on a bone...
 

2nd tenor

Well-Known Member
......... so with all the above sort of clear in my mind I went off to practice this morning with a valved instrument. You guessed it, in practice the tuning was much less precise and repeatability inconsistent. I used a reputable tuner to check, its accuracy is assumed to be good but I now wonder about how much trust to put in it. Perhaps those variable results are down to poor playing (operator error), or the instrument, or a big mouthpiece or a combination of multiple factors. Anyway the bulk of my tests used the notes and range that would be uncompensated on a four valve compensating instrument, and the match between theory and (my) practice was looser than I would have expected - easy to be 20 cents out. Of course that's one experimenter's experience on one occasion, I'd be glad to hear the results of similar tests by others.

It was interesting to pull slides and look for some optimum compromise. After adjusting the main tuning slide to give a (tuning) balance between bottom and middle C's I then needed to pull the first slide a little and the third noticeably more; when played the result was quite pleasing in use. I did occasionally find myself thinking that that last note would have sounded just a fraction better through the fourth valve; however playing is by nature full of compromises, trade-offs and balances so such a reaction is an understandable part of that mix - which, amongst others, also includes weight, cost, condition and ability.

The data and experiments lead me to believe that the humble three valve uncompensated instruments get less kindly thought of than is really justified - YMMV- and that it's very much worth while getting your tuner out to check how all the notes produced by your instrument sound. i.e. Don't assume that 'cause middle C on your instrument is in tune that the rest of the notes produced by your instrument will be too, or that you are perfectly consistent in how you produce notes of any particular pitch.

Many thanks to MoominDave for all his (excellent) comments, I hope that the thread will be of wide interest and helpful to others in the future too.
 
Last edited:
Top