Electrola/HMV 102 springs - motor 271-C

[anchorman]

I have an electrola 102 that I'm working on, and it is mostly in good shape. The biggest issue it has is some damage to the spring barrel. I think I may be able to repair, though. I will need a new spring, and need to know the size of the original springs. I have some used springs from other machines that could be made to work, but would like to know the original specifications before substituting something that may be too powerful or too weak.

According to some info I found from the clock and watch making world, going from .024" thickness (what I'm guessing the original was based on replacements available) to 0.026" thickness (the thickness of my available spring material) would give an increase in power of something on the order of 27%. If that is too much power increase, I can order a spring from the phonograph shop that will be 0.025" thick, which is only 13% more powerful than the purported original thickness. .024" springs are not readily available in the USA, and would need to be imported from UK or elsewhere.

I'm also wondering if anyone has tried reversing the springs, so that the natural direction of most of the coil is opposite to how they are wound into the barrel, such as in the attached picture.

This is supposed to give more even power from start to end. I'm not sure that this would be possible to do with already hardened springs, but given that the center needs annealing and forming into the correct crimp shape, then coiling slightly so that the crimp engages properly with the arbor, perhaps it is possible?

[anchorman]

I found the formulas for proper spring barrel fill, and after making a spreadsheet and trying out various thicknesses, the length is largely consistent with the lengths of springs listed in the few motors that I have easy access to from Chunny's website:

https://www.gramophonemuseum.com/garrar ... otors.html

My calculation for the garrard super, and the barrel size and arbor I measured on mine is showing 17', whereas chunny's site lists 16' 9", and The original advertisements there list it as 16' 6".


Lms = PI x ( DB^2 - DA^2 ) / (2 x 4 x Tms)

where Lms = Length of mainspring, Tms = Thickness of the mainspring, Db = diameter of the barrel, and Da = Diameter of the arbor.


Very handy formula if you want to know how long a spring ought to be to fill the barrel properly.

[soundgen]

HMV 102 spring https://www.ebay.co.uk/itm/255500802977 how is your spring barrel damaged If you need to buy a spring I could send a spring barrel FOC

[anchorman]

A previous repairman did not install the old spring properly, and the rivet holding the end of the spring has been pushed outward, making a bulge in the barrel. There's also some minor damage to one of the gear teeth, but this doesn't see to cause any issues when the barrel is put in the machine and rotated. The lid is dented slightly, and bent - easy enough to flatten (I hope).

I was curious what the original spring thickness was, so that I can evaluate the merits of various replacement options. I'm guessing the original was .022" based on the numbers you have provided in your auction on eBay.

Regarding the length, my calculator shows that it could be increased to 16' without issue, and with a .025" spring, could be the ordinary length of 14'. I'm currently having some other springs made up by a (slightly more local to me) supplier here in the states, and was wanting to make sure I was getting the correct size for maximum performance.

The spring that was in the machine was only .020" thick, definitely not the right length, as it was kludged together in various ways.

[Inigo]

That's an interesting formula, meaning that the total volume occupied by the spring (Lms•Tms) is only 50% of the total barrel free space (π/4•(Db²-Da²)). Interesting. I would have said that the spring would fill 80% of the space, not as little as 50% ...
One never ceases learning. Thanks a lot!
Despite this formula, it's clear that some models used different percentages of barrel infill, for we've learned here that the same motors were fitted with different springs depending on the machine model. It may vary depending on the required power*duration. It's clear that smaller machines were filled with the same motors but in different power arrangements. Would like to know if the results given by the formula mean the average or recommended value, or the maximum recommended... I tend to think it's a maximum that must not be surpassed for the motor to work properly.

[anchorman]

my understanding is that if you over-fill the spring barrel, you cannot get the full power/run time theoretically available from the spring. same if you undefill. longer spring = longer run time, thicker and/or wider spring = more power available. The power delivery curve is not very even, regardless, in such a simple motor, with more torque being available when the motor is fully wound, and progressively less as the spring unwinds. This is due changing radius of the arbor as the spring is wound onto it (or unwinds from it). Clockmakers did some ingenious things to make their motors have more equal power delivery, from the spiral/chain drive Fusee mechanism, to changing the shape of the mainspring (winding it for some portion with the natural curve of the spring reversed from it's ordinary inward spiral.

Not really necessary with a gramophone motor to go to such extremes. Governor design to prevent wow/flutter, is more important IMHO than absolute speed over a long period of run time. you wouldn't want the speed to change during the playing of a record, but keeping it dead on to the very end of the spring is less important than in a clock, perhaps? Ideally you don't use the gramophone until the spring runs dead, and even a rudimentary governor can keep the speed relatively constant over the duration of playing a record.

[Inigo]

But I've noticed repeatedly that good gramophones run pretty steady until the very end, when they lose power suddenly until the TT stops dead. Then you lift the soundbox and the residual power only moves the TT for some more time, already almost no power left. I mean that the system of spring+governor works very well in good motors until the very end, provided you're playing soft records in good condition. So the design is clearly in the direction of good surplus spring power, dissipated by the friction on the governor pad, which gets softer and softer with time until the moment when the friction at the needle is equal or greater, then the governor starts to fail, and the power falls suddenly. A good design Indeed.

[anchorman]

HMV 102 spring https://www.ebay.co.uk/itm/255500802977 how is your spring barrel damaged If you need to buy a spring I could send a spring barrel FOC

I managed to fix the bulge in the barrel, but the rivet is a little dodgy looking still, chewed up a bit around the edges, where the broken spring eye slipped past it. If you had just a barrel you'd like to sell to me, I would appreciate it. I'm actually curious to have two barrels so I can quickly interchange them to compare spring strength, without having to pull them from the barrel in between trial runs. could I buy a spring and barrel from you and also another barrel too?

[anchorman]

I am currently also working on documenting governor spring and weight sizes in the various motors I have in order to share that information with Brian Parlier from the Phonograph Shop. Today I took apart a columbia 202A portable to have a look at the motor. It fortunately has an identical spring barrel to the HMV 271-c motor, and since I am closer to done with the electrola, I'll be stealing the spring barrel for now. The motor for this machine is unlabeled, but it is mostly identical, other than having a horizontal crank rather than sticking out at an angle as they do on the HMV machines. I took a bunch of pictures, which I will post for posterity later. The columbia motor has a cast iron frame, which to me is a nice touch, despite the extra weight it adds to a portable.

[anchorman]

Some pictures of the Columbia 202 motor, which is for all intents and purposes identical to the HMV 271-c… except for the winding crank angle.