Converting an Amberola V to play 2m cylinders

[gregbogantz]

Ralph, you may be chasing this flutter bug for the rest of your life and still end up frustrated. The belt-drive Edison cylinder machines are notorious for having flutter. It's inherent in their design. Some have it more than others, even within the same model. The design is just conducive to flutter. Assuming that you get the gears clean and the shafts straight and the governor assemblies working properly, the biggest problems are that there is non-uniform drive torque presented to the mandrel shaft by virtue of there being a belt drive of any type. Secondly, most belts are leather, many with a splice in them which lumps and bumps as it passes over the pulleys. Third, many of these belts are old and have taken a "set" in that they are permanently stretched at one or more points along their length so they lump around as they turn. This all conspires to making the torque delivered by the belt non-uniform as the belt travels over the pulleys.

Now, these shortcomings can be largely reduced if the driven end of the belt, i.e. the mandrel shaft, has susbstantial rotational inertia, that is if it has a lot of mass such as a flywheel. That's how modern turntables get away with using belt drive - the better ones have LOTS of platter mass to filter out the flutter. Most Edison belt-drive machines have insufficient mandrel mass to effectively filter out the speed variations delivered to them by the belt drive. Edison himself finally figured this out and improved his designs in his later motors. The Amberola V is one example of having an added flywheel mass on the mandrel shaft in addition to eliminating the belt drive. Furthermore, it has the mass of the mandrel shaft decoupled via the spring wire that isolates the torque input from the mandrel itself. The Opera motor has these same refinements as do the later amberola motors used in the 30 and 50 models. So, you may get lucky and discover a combination of parts that gives you acceptably low flutter in your Triumph, but my guess is that you may never be able to get there from here. I have a Triumph in good condition and it has always fluttered, regardless of what I've tried to do to fix it. I just have resolved to listen to my Amberola 1B when I want to play 4 minute records and my Amberola 1A when I want to play 2 minute records. The 1B motor is much quieter than the 1A, so I prefer it. I also have an Amberola V and and Amberola 50 which both have low flutter, but the 1B has the much better horn.

[pughphonos]

Greg, you are spot-on and I know it fully. I was talking yesterday with one of my local antique phonograph dealers and he said it again (he's said it before) that the joint on the belt is one of those areas that can introduce flutter.

Yes, I could be chasing this problem forever. One can't expect the technology to exceed its limitations. It's just that I'm determined to push those limits to the nth degree and not settle for less. I've already contacted another repair firm and will send the Triumph out to them in a week or two. The last area that I think might be capable of improvement is the 2/4 m feed screw/planetary gear assembly, which seems to be sticking a bit and perhaps causing flutter.

If, once it comes back, it still bothers me too much, I will sell it. I'm happy with my Amberola V and am willing to sell off my 2m cylinder collection and move some resources into early discs. Oh, I've figured out a way to get my model O reproducer onto the Amberola V, so I will still be able to play my wax Amberols.

Thank you all for listening to my frettings about this over the past few weeks. Again, I have not wanted to be unreasonable, and I DID back off the idea that started this whole thread (my modifying my Amberola V). I now recognize that, were I an adult back in 1905 and had a chance to compare belt-driven cylinder machines and turntable-equipped disc machines, I would have invested in a disc machine. This is said with full appreciation of the fact that the fresh 2m wax cylinders of that era were basically superior to disc records (in terms of quality of captured sound). But their belt-driven machines did not do them justice--and for those of us who HATE flutter and pitch variations, we'll put up with a bit less richness of tone just to be assured of evenness of pitch. Unfortunately, when one engages in this hobby and remains faithful to the technologies available at the time, one must choose (unless one can afford the Amberola 1A and 1B, which Greg can--and more power to you! I did debate saving up $ and going that route, but they are also space eaters).

Ralph

[De Soto Frank]

"The design is just conducive to flutter. Assuming that you get the gears clean and the shafts straight and the governor assemblies working properly, the biggest problems are that there is non-uniform drive torque presented to the mandrel shaft by virtue of there being a belt drive of any type. Secondly, most belts are leather, many with a splice in them which lumps and bumps as it passes over the pulleys. Third, many of these belts are old and have taken a "set" in that they are permanently stretched at one or more points along their length so they lump around as they turn. This all conspires to making the torque delivered by the belt non-uniform as the belt travels over the pulleys."


This is why I posted a question the other week about the efficacy of the fabric-reinforced rubber belting being sold for Edison cylinder players on e-Bay. To my little grease-monkey mind, an "endless" (read: seamless), reinforced belt, made from "grippy" rubber, would probably slip less than the traditional leather belt.

Perhaps the belt drive concept is not the beaux-ideal at all, but maybe the performance could be improved ?

As far as I know, the last major car maker to utilize flat leather belting for accessory drive was Ford, in the Model T.

When the Model A came-out in October 1927, it featured a fabric-reinforced rubber V-belt.


I like the bevel-gear drive of the later Amberolas: positive and quiet.

[De Soto Frank]

Ralph:

I wonder if your Triumph has an issue with the mandrel shaft and /or feed-screw being "sprung", cause the shaft to "whip" and rub against the inner / outer neighbor ?

If either one are not perfectly centered on the same axis, that could well cause binding...


Good luck with your continued efforts... I have a couple of warbly Homes... hoping to make them better...

[pughphonos]

DeSoto Frank, I've been wondering the same. Yesterday I loosened various screws securing the drive shaft and mandrel on the Triumph, wiggled them about a bit, and now the 2/4 m conversion works better and I seem to have lost a bit more flutter. I've been playings lots of 4m cylinders today, switching my Diamond B reproducer between my Amberola V and my Triumph D. The Triumph is now "in the ballpark," performing close enough to the Amberola V to restore its retention as a real option.

Yes, a smooth playback via the Triumph does depend largely on those two shafts revolving properly around the axis.

I'll give myself a few more weeks before deciding which machine to sell. It's not that I'm desperately in need of the $. I would just rather focus my attention on a few machines and work them up to maximal performance than have an army of machines. Without the Amberola V, though, I never would have been able to do direct comparisons and really scrutinize how the Triumph was doing.

There is such a variety of folks on this forum and it's great. Different tastes, perspectives, needs. Some are audiophiles (like myself); others are collectors of mechanical variety (the two aren't mutually exclusive). Of course it tends to be the collectors/dealers who have the widest range of knowledge in the field, and hats off to them. But the fussy audiophiles (I stand self-accused) offer an interesting perspective (I believe). As Joe_DS said very well on another topic, audiophiles are not going to just accept an "old-timey" sound as normal for old recordings; right back into the late 1880s it is possible to find good recordings and reproduce them well. Thank goodness for the patient collectors on here who provide such valuable advice/insights.

[Phototone]

It seems that vintage photos showing a recording phonograph at the Edison works appears to have a flywheel addition.

[pughphonos]

Phototone, that's an interesting piece of evidence; supports the general knowledge/view that in the first 40 years or so, recording was superior to reproduction.

This is probably going to be my last post to this string, as I have a GOOD piece of news. My working on the feed screw/planetary gears (on my Triumph model D) eliminated practically all the rest of my flutter. So, having bought that machine on February 8, 2012, and having worked on it myself since then (and sending it off to a great shop), I have finally gotten to a point where the reproduction is as best as it can be. She is now running essentially as evenly as my Amberola V; and as you know how fussy I am, that is saying something.

Of course this means that I have finally gotten a series of delicate arrangements in proper alignment and coordination, and if something slips I am going to have flutter again. But at least I now know how to maintain the machine myself, and if/when some problem re-emerges I trust that I can address the issue. The Amberola V's heavy flywheel helps suppress any flutter coming up from elsewhere in the machine; it's like having a robust immune system that can suppress random illnesses. The Triumph, on the other hand, does not have that robust immunity and therefore has to be kept in "good health" in any number of separate areas. If anyone buys a belt-driven cylinder machine with flutter, my heart goes out to you because it might take you tons of time and aggravation to fix it.

END OF STORY. I will probably sell the Amberola V in a month or so and force myself to stay on top of the Triumph's needs.

[Edisone]

I urge you to keep both machines. The Amberola V is surely scarcer than a Triumph, besides being a fine phonograph & nice piece of furniture. I also believe there's an accidental property of its horn's size & shape which causes records made in Edison's NYC studio, circa 1911-1914, to reproduce with clarity and a sense of spacial placement. On many of my Blue Amberol orchestra & band records, I hear instruments distinctly to the left, the right, closer & farther from the horn. I don't hear that when played through a Cygnet.

[PeterF]

Selling the V would be, in my opinion, a mistake. They do turn up, but Triumphs turn up a lot more frequently. Amberola V is a mini superstar, with a great aesthetic design as well as a robust and well-designed mechanism. They are very special, and rather rare relative to Triumphs. Plus you seem to like yours.

Your Triumph might have a simple issue, or it could have been tampered with over the years by the unknowledgeable or inexperienced - so maybe another Triumph someday will be more to your satisfaction. I'm currently lucky enough to have two Triumphs, and one of them is whisper quiet with perfect unwavering pitch, while the other has a nasty warble that makes it unlistenable. They were the top of the line mechanism design of their day, powerful and stable - so please don't write off the Triumph based on your current example.

My very first machine was a Home, and soon after I acquired it the spring broke while it was playing. After spring replacement the machine had a horrible flutter. Dwayne Wyatt of Wyatt's Musical Americana, through his knowledge and experience - and close observation of the governor shaft while spun in a lathe - found the shaft was bent. The sudden release of the spring had bent the governor shaft. He replaced it and the sweet tone returned. That or a poorly-spliced belt are your Triumph's most likely culprits.

In my years of collecting I've more often than not found that a good cleaning and lubrication will fix almost any ill on an original but aged machine. The ones that have deeper issues are almost always the ones that have been "fixed" or "restored" or "adjusted" by well meaning amateurs, or (in some cases, by the looks of things) blacksmiths.

[Uncle Vanya]

As far as I know, the last major car maker to utilize flat leather belting for accessory drive was Ford, in the Model T.

When the Model A came-out in October 1927, it featured a fabric-reinforced rubber V-belt.

I gather that you don't drive a Ford Car.

Ford changed over to a flat fabric belt quite early on, right about the time that the moving assembly line came in to use. By 1915 Ford was using a rubber coated fabric belt on all production. Some accessory forms offered leather belts, and some owners use them, but they were by no means standard. Re3storers in the 1960's and 1970's commonly installed flat leather belts on their Flivvers because they looked more antique, and because a quality vulcanized fabric belt was not available. In addition, the front crank-shaft oil seal is often leaky on a Ford, and so an oil-soaked leather belt will last far longer than will an oil-soaked natural rubber coated belt. Modern production Ford belts use oil-resistant synthetic rubber, and are much more reliable. The primary trouble with them is due to their tendency to slip off of their pulleys if the working faces of those pulleys are not exactly parallel.

As I recall Overland was using flat leather belts on factory production up until at least 1920.