Upgrade for users of 1/2" mylar belt

This thread will be of interest to Galibier, Redpoint and Teres belt drive TT owners, or anyone using or thinking of using a drive belt made from 1/2" mylar. The belt material of choice on these tables for several years has been the silver holographic mylar streamers from sources linked on the Teres and Galibier websites. This outperforms everything else we've tried but like anything it's not perfect - and we've now found something better.

One of the silver holographic mylar's assets can also become a liability. That metallic layer, when new, gives the belt exceptional "grip" on the platter and motor capstan. Minimal slippage on transients is one reason the material works so well. Unfortunately, that layer can wear over time, leaving a silvery gray residue and allowing performance to deteriorate. The only solution has been a good cleaning and a new belt. No big deal, but if there was a better or more stable belt Paul and I wanted to find it.

I'll spare you a recounting of our many experiments and jump to the good news: thanks to (yet another) brainstorm by my resident genius/partner, we've developed a belt that both performs better and lasts longer. I've held off posting until we were sure, but after 4 months of steady use Paul’s idea is still working perfectly. The only negative is that making this new belt takes an hour or more of work spread over two days, but to us it's worth the effort.


1. Cut a length of the silver holographic mylar tape sufficient for your TT, plus 3-4" extra.

2. Remove the silver metallic coating off the backside of the mylar. Paul used an acid etching cream popular with stained glass hobbyists to dissolve the metal - safe, quick and easy (use skin and eye protection).

3. Rinse thoroughly under running water, inspect to make sure you got all the silver off, wipe and hang to dry overnight.

4. Splice as normal to make a TT belt, making sure you tape on the OUTSIDE (which will be smoother than the now bare inside).

5. As always, the best splicing technique is to overlap the ends and cut on a 30-40 degree angle. Apply 1” video splicing tape (*not* tabs) across the belt at the *same* angle and trim away the excess.


Unlike any plain mylar ribbon you could easily buy, the silver holographic streamer has an ultra fine texture embossed on the back side of the *mylar* during manufacturing (*before* the metallic backing is applied). Once the silver layer is applied you can't see this texture, but that's what diffracts light like a million tiny prisms to produce the shimmery rainbow effect. Stripping the metal backing away exposes this texture, which becomes the contact surface of your new belt.

Paul realized this texture must exist and then hypothesized that using it for the working side of a TT belt might provide more “grip” on the platter rim and motor capstan than either plain mylar (which is extremely smooth) or the metallic backing. He was right. This belt produces more lifelike dynamics, both macro- and micro-, than even a brand new silver holographic belt - which heretofore was the best.


Removing the metallic backing exposes bare mylar, which is much sturdier than that fine layer of metal. Under normal use and assuming no accidents, one of these belts should last many, many months, perhaps years, while delivering consistent performance. Ours is going into its fifth month with few visible and no audible signs of wear.

Different motor capstan materials might interact with this belt differently, but I urge anyone with a compatible table to give this a try. Like all our favorite little tweaks, we'd find it hard to go back.


1. Clean any silver/grey residue from your old belt off the motor capstan and platter before mounting the new belt.

2. As many of us have learned, the more inelastic the belt, the more critical motor leveling becomes. That is truer than ever with this new belt. Getting the motor set just right is touchy. Take care that your new belt is riding level in the center of the capstan before you start to play. You don't want it sliding up or down and mangling itself.

3. Motor distancing is also more critical than ever. Since this belt will not slip *or* stretch, tension must be perfect. The right amount is just shy of the tension that would tilt the motor off its feet.

4. Depending on your climate, the belt can build up static potential during use. Not enough to spark, but more than enough to attract airborne dust. I dust the belt's inside with my CF brush after each side before stopping the platter.

5. With this or any belt, always start your platter spinning with a helping push. Just pressing the motor's ON button creates lots of belt-wearing friction as the rapidly spinning motor tries to drag that heavy platter up to speed.

It all sounds like a pain, and it is! But the sonic and longevity results are worth the effort.

can't help but think that beeswax would help with traction; any experience with that?
You know Doug, I always keep an eye out for the things you and Paul come up with even if I don't post for a long time. And this is one of the things I watch for. Thanks buddy, for a relatively easy way to get better performance out of my TT. i wish I had Paul's ears, and his out of the box thinking. I'll let you know how it works on mine.

Thanks again!
One of our reasons for trying this experiment was to eliminate the residues left by the silver holographic belt. Beeswax would just replace one residue with another, and would certainly be a magnet for any dust attracted by the belt. It doesn't seem likely that any treatment like that would provide stable performance over the long term (nor, to be honest, would I want it on my platter).

Good to see you. That's still the niftiest arm mount for a Rega/OL style arm I've ever seen. I remember we once discussed belt dressing (as used in car engines) and you warned me away. Piedpiper's idea might be less messy, but a clean interface is best.

There are two sources on the Galibier website, one of them's on the Teres site too. Enjoy.

Anyone who tries this, please report your results. We're always curious...

Doug thanks for the update on mylar belts. I just made my second belt after using the first for 6 months, though the first one was still pulling fine with no noticeable residue. Will sure give Pauls new idea a try. Have a good one.
Nice post Doug,
Illuminating and helpful as always.
Thanks for your (and Paul's) continual efforts to learn and share.
Doug, I'm going to send my 100' roll of mylar down to you so you and Paul will have something to do in your spare time. I can see clotheslines strung all around the living room and kitchen. ;-)

I find this fascinating since I've thrown out a few belts after the silver has worn down. Now I find out that I should have kept them and cleaned more of the stuff off. So, where can one source the etching cream?

Thanks again for sharing your tweaks!

EDIT: a Google search will quickly show sources for the etching cream.
We did consider filling the bathtub with sulfuric acid, dumping a few dozen rolls in and going into business selling TT belts for $500 apiece...

If that Certus table ever needs a new belt, it's on me! ;-)

Whut?!? No cryo treatment?

BTW, I did find an etching cream that does not use some real bad stuff. A bit more expensive, but the add says it can be reused. Second row of product.

etch all

I have tried doug's belt for a few days now on my Galibier gavia. I can hear the benefits of this vs the mylar belt i was using before. The speed stays on better, I believe as there is a tighter feeling in the bass without losing bass.

I will have to continue listening, but I would reccomend this belt to anyone including the designers of these tables.

Now if we can just get Doug to go into the belt making business...
Thanks for the update, Mothra. My etching cream is on the way so I'll get to try this soon.

I think Doug and Paul could start a fashion trend there in CT. by wearing those HazMat suits around the house.
I only wear my HazMat suit for posting on public forums. Acid etching is benign by comparison!

Thanks for the report. If you get your cartridge set up just right (especially VTF and antiskate), you may also hear an increase in microdynamics, sound space info and very low level harmonic detail.
Do you guys have a recomendation for a VPI Scoutmaster TT, I would like to try something other than the belt that comes with the TT.
I've read some good reports about various kinds of thread, string, fishing line, dental floss, etc. I also remember one VPI owner who modified his table to accommodate a wider belt, used 1/4" wide mylar and reported good results.

The difficulty in implementing this kind of belt on a VPI (correct me if I'm wrong) is that the motor capstan and/or platter rim are grooved for a fairly narrow belt. This limits your material choices to whatever will fit, unless you modify your table.

If your platter rim is flat, getting a local machine shop to make you a taller motor capstan might be viable. For a 1/2" belt you need a capstan working surface height of about 9/16" to 19/32", very slightly convex and with lips at top and bottom to stop the belt riding off. The O.D.at the center would be the same as your existing capstan, obviously. Material choice probably isn't super-critical, but the surface should not be highly polished.

The key characteristics in a good belt material are:

1. Lack of elasticity. The more resistant the material is to stretching and rebound the better. Rubber is terrible (sorry). Mylar is dimensionally very stable. At the speeds and loads experienced by a TT belt it wouldn't be far wrong to call it "perfectly" stable, depending on the thickness.

2. Lack of "slipperiness", ie, sufficient friction to prevent slippage when a big transient tries to slow the platter. This is tricky to get right. The biggest risk of slippage is at the very short belt/capstan interface. The belt/platter interface is long enough to make slippage less likely there. The optimum material is very much table dependent. On a Teres motor capstan materials like string, thread, etc. result in quite audible slippage, though they're still better than anything elastic. Mylar will also slip if it's too thick and/or too smooth to make the curve around the capstan, and that's a function of your capstan diameter, material and surface.

Not much of an answer, sorry. But I encourage you to experiment. It can't hurt and you'll certainly learn something from everything you try, as we have.
what sort of concentration etching gel do you use? google searches return various types of acid precentages, 10%-38% acid content. i will be trying this on a tw raven one. i will need 1/4".
seems like all the sources are for 1" or 1/2". does anyone know where one can find 1/4"
Hey Doug.
This sounds like a worthwhile tweak. I'll try to source some etching cream and give it a go.

One of the reasons I'm a believer is because of what I've observed with my Teres 145 when replacing a worn belt with a new one is just what you report. Improved dynamics. A more immediate sense of musical flow. I wasn't aware of the layered construction of the holographic mylar but this makes sense.

Another maintenance tip I'd add centers around the bearing thrust pad. A new thrust pad, say on at least an annual cycle, will restore some sonic performance for these turntables.

Btw, you didn't go with a Verus upgrade?

I'll have to ask Paul if he remembers the concentration. He used up the last he had, so we'll be sourcing some too.

Good tip re: the bearing. Ours is about due for a check.

We did try the Verus last summer. We were still using the silver holographic belt then, but even so the Verus was not an upgrade on our rig. It could not compete with our modified Ref II motor, modified controller and the faster/stronger battery we use vs. the Teres standard. It would be even less competitive against this new belt, which gives more raw 'oomph', the one area where the Verus did have a small advantage.

Our listening sensitivities and priorities may differ from yours, so keep our experience in context.

The Verus's speed controller is adjusted in fixed steps, and no setting yielded exactly 33 1/3. The deviation is too small for most people to notice or care about (and it varies depending on platter diameter), but we both have very good pitch, so we could tell it was running fast or slow. Chris has since halved the size of those steps, so it would be better now.

Also the rubber O-ring was a source of resonance rebound or elastic response to stylus drag events. Again, this was at a level many people don't care about. If we were rock or even regular classical listeners we might not notice. But much of our favorite music is pre-Romantic era on authentic instruments. These recordings are unforgiving (ie, revealing) of even the tiniest system problems. The instruments go from unique and rich timbres to fingernails-on-blackboard in a heartbeat if ANYTHING is wrong. They're a real acid test which we listen to almost daily, but which most people never listen to at all.

The Verus is well liked by most who try it, but its not for everyone. YMMV, as usual.
Hey Doug, I'm sure you and Paul did quite a bit of experimenting but how many belts did you get from a 3 or 4 oz bottle of etching cream?
The etching cream was mostly gone long before Paul thought of using it on a TT belt. He used to do stained glass as a hobby.

That was years ago. You'd have to see our basement to appreciate what a miracle it is that he even found it. ;-)
Interesting re: Verus upgrade and your experience with it. I was in line to try one but then Chris kept putting me off on delivery dates so I just let it drop. I haven't really heard one, except briefly at VSAC '08 on a 265 model. I rarely get a usable impression on audio gear at shows. Too many distractions,... plus room acoustics are rarely good or comparable to my own.

Also interesting re: your comments about your "modified Ref II motor, modified controller and the faster/stronger battery "

This comment suggests that a stronger battery will help motor response to stylus drag "events"? Frankly, with my 145, this is the one area where it falls down in comparison to other turntables. It's really quite laid back in character.

This became obvious when I began work on a TD124 project a while back. The one area where the Thorens was really quite a lot better was in its ability to project percussive sounds into the room. This seems to apply to any music with a strong wide dynamic content, including rock, jazz and classical. Particularly drums. The Teres 145 seems week on projecting a realistic drum sound.

I'm also listening to R2R tape for another source perspective. The tape, particularly with classical titles, can project a very wide uncompressed dynamic scale.

But now I'm rambling.

Update: the cream Paul used is "Armour Etch". Paul's old bottle was half dried out, but a little water brought it back to life. Despite dilution it was still plenty strong enough to strip the metal layer from the mylar.

The link Dan_Ed posted above has Armour Etch on the top line of products. It's nasty stuff, but half the price of the friendlier product Dan spotted on line 2. Be cheap or be green, your choice.


Rambling is encouraged! Our system also used to sound soft in the bass, much as you described. No longer.

Our battery of choice is this: http://www.batterymart.com/p-Big-Crank-ETX20L-Battery.html.
It's liquid rather than gel, so it reacts faster and delivers higher peak currents. The Absorbed Glass Mat internals make it spill proof, for safety, and there's no reason to restrict yourself to low amp/hours. As long as it's 12V you're safe. After several experiments we settled on the 17.5ah model linked above.

Another idea of Paul's, naturally, and it easily out-powers the regular battery on transients and bass. I can rock the house down to the lower limits of our speakers (32 Hz). Way better than what Dan_Ed heard the last time he visited us.
How can you go wrong with a name like BigCrank! ;-) So far I've stayed with the jump start type battery packs that can be found at various places. I don't have an external charger and these come with one built in.

So, there is another item for the list for next time I get down your way. The Anvil, and my Jump n' Charge. Sounds like a game for kindergardeners compared with BigCrank!
We tried our big car recharger (similar to what Thom supplied you?). It sounded rougher than the Big Crank. I remember Thom theorizing that a Teres motor controller would respond more than a Galibier would to battery differences, due to its higher impedance. Paul says the Big Crank has very low internal impedance, so I guess that makes sense - as much as anything electrical makes sense! <;~)

I don't know any sources for 1/4" mylar. You could try what Chris Brady did when he first wanted to experiment with 2 mil vs. 1 mil mylar. He couldn't find 2 mil in 1/2" rolls, so he bought a large sheet and cut 1/2" strips with a straightedge.
Doug, thanks a lot for sharing your worthwhile experience!
I've tried it on my TT with huge success. Your belt replaced McCormic's black mylar tape and profound improvement accross the board was a revelation for me: my playback became better in each and every aspect. That was very educative experience indeed; now I convinced the interaction between platter, belt and motor is critical.
Next thing I want to try is battery power. I have earlier Teres Signature II motor (short case) with no battery power option. Can you recommend any battery/charger combination I can hook up without messing with electronics (my weakest point)? Thanks again.
Hi Livemusic,

Thanks for sharing your results, and glad you were able to hear the profound (indeed) differences that result from optimizing the motor-platter coupling. It's obvious when you think about it, but hearing is believing.

Of course the best implementation for any particular rig can be anything but obvious. It took us 5 years of trial and error with 10-15 belts to reach this point with this table. Some other table might need something different, though the principles are the same: no elasticity and minimal slippage.

For battery power you'll probably want the battery option from Teres. It should still be available, ask Chris. This provides the proper interface to the motor controller (internal to your Sig II) as well as a trickle charger. You could probably DIY if you had electrical skills, but I don't and apparently neither do you! Maybe someone who understands electricity will chime in with an easy DIY.

One tip: although the Teres charger disconnects itself when the motor's running, we still hear a small degradation in sound quality. A few others (not many) have reported the same. We have our charger plugged into an outlet strip with a power switch, which we turn OFF while we're listening.

For the best battery we've found (out of 5 or 6 tried), see the link I posted above.
Well I finally opened last night my new bottle of Etchall to remove the silver layer of my mylar belt (Galibier Serac), following the procedure outlined above by Doug--except that I was working with an already-made belt. I then did several listening passes, with different LPs, alternating the "cleaned-up" belt and a brand new one with its silver layer still in place.

Here is the list of the recordings I used for comparison:
- Vivaldi, Four Seasons, Fall, 1st mvt, Kuijken and La Petite Bande
- Beethoven, String quartet op.127, 1st mvt, Quartetto Italiano
- Stravinsky, Firebird, Infernal Dance of all Kastchei's subjects, Dorati/London Symphony
- Mahler, Symphony n.7, beginning 1st mvt, Haiting/Concertgebouw

So, a good range of ensembles and recording difficulties. In all cases, most noticeable were:
more extension in high register; much tighter bass, with better pitch definition and spatial localization; better sense of perspective in depth; slight recess of mid-bass range, giving a somewhat better separation between instruments (this one threw me off at first because it sounded like less body; but after some time I realized what was being gained rather than lost).

This has got to be one of the most spectacular tweaks I've experienced recently!
Thanks Doug and Paul, for your pioneering efforts!
Thanks for the report Jfd01. :-)

Good job hearing through the trick of flabby mid-bass that gives an impression of body but masks the texture and overtones of real bass instruments. The greater extension in HF's is, I think, due to more stable platter speed and lack of slippage: the tiny waveforms above 15K are easily lost if everything isn't "perfect". Pitch definition is of course dependent on platter speed, and the more pitch-sensitive your ears the more critical this aspect of TT perfomance becomes.

When you said you planned to etch an existing belt (rather than unspliced mylar) I worried about the stability of the glue. No problems there, apparently?

I finally got around to trying this over the weekend. This is a wonderful tweak and I give it the two key lime pie award! I believe this is getting us close to the realm of direct drive, at least as good as an idler.

My perspective on the improvements are similar but, since I listen to mostly rock (a little flab in the bass is not a bad thing), I thought I may have noticed things differently. But I've had some time to digest this overnight and I don't think I would disagree with anything anyone has posted about this new belt configuration. What I've noticed the most is the increase in micro-details. I'm hearing new sounds in the background and around the edges. Bells, wood blocks, chimes and such are not being smeared anymore so they come out to the front. The attack and presence of all instruments is greatly enhanced. I hate to use the cliche about veils being lifted, but there it is.

I found the process of etching the belt very easy. HINT, is it always the side that the mylar wants to curl to. The thought did cross my mind about trying this with an already spliced belt, but the worry of what happens to the splice tape prevented me from trying that. So I simply cut the old splice out and proceeded. This led to a belt that is about 2" shorter than what I normally use, and I'm left pondering if shorter isn't also better. This means that the wrap around the motor capstan isn't as tight so the belt runs smoother. There seems to be no chance of belt flutter with the shorter belt. With the increase in friction provided by the de-mylared surface I suspect that an even shorter belt may work better.

My belt may not be as tight as what Doug gets. I say this because I'm pretty sure I'm no where near tilting my Galibier's motor. I found that there is a point where an increase in tension causes a noticeable thump when the splice went around the capstan. So I backed off to the point where the splice was barely audible with no music playing. That should be good as I find the splices to relax a bit after running for several hours. There is still more tension than I ever ran with the stock mylar.

And that leads me to my only complaint. It is too bad that splice tape comes in beige and not clear! Watching that splice go round and round is disturbing, but I can learn to live with it.
No problem with the splice so far. It hasn't moved.
Like Dan, I'm running the belt with a bit more tension than before, but not near the motorpod tipping point.

I also agree with the increase in micro-dynamics, as it dawned on me while listening to Stockhausen's orchestral piece, Inori (SWF Orchestra, DG). In this piece, Stockhausen created a "scale of dynamics" with 60 degrees, from almost silence to very loud (by varying the mixture of instruments and dynamic indications); there's also plenty of complex attacks and subtle resonances throughout to test the resolution of a system; and it's recorded with beautiful perspective in all dimensions. Plus, it's a great piece of music...
I've always had problems with the belt "stretching" at the spliced area so that the adhesive area of the splice becomes exposed after only a few hours of playing. This exposed area becomes larger as time goes by and the larger it gets, the more noticable the sound as it passes by the pulley.
Is there a know splice material that outperforms all available? That is my main gripe with mylar tape, the stretching.
I haven't yet tried the method Doug outlined in this thread but I'd like to, only if I can find a solution to the stretching of the belt at the splice point. Any ideas?
Frank sm,

Are you using video splicing tape? The silver tabs inevitably spread as you described. So do general purpose tapes, and they can leave glue residue on your platter too.

Splicing tape won't spread. I've had two year old splices that were as tight as new.
I'm using the silver tabs. Can you offer a link to the splicing tape you recommend?
I'm also using Silver holographic tape. Is there a better type of tape than this? I'm not that keen on the color silver anyway. Thanks for your help Doug!
These guys have good splicing tape:
Thanks for finding that link, Jfd01. I've got it at home, but I'm not at home!

Removing that silver backing you're not keen on is the whole point of this thread. When you do, only clear mylar is left.
Hi Doug, Thanks Jfd01 for the link. I'd like to make one of these belts. I haven't had the time to even locate where to purchase the "etching creme"
I'd like to see if you can clarify something for me. I've read over your initial instruction and want to make sure I understand the process.
The stripping is actually performed on the side of the belt that would normally be the NON-contact side, the shiny silver side. Then after stripping this surface it becomes the contact surface with the platter, is this correct?
Also, is there a known source and name for the etching creme? Thanks again!
Splicing tape is always white so engineers can more easily find the splice afterwards.
I just ordered the etching creme. I reread the thread and found it, many thanks.
Regarding the splicing tape, Is there any splicing tape available that matches the colour of the tape?
Splicing tape is always white so engineers can more easily find the splice afterwards.

They sure won't have any trouble seeing that splice come around on a clear belt. ;-)


just to clarify a little. The "metal" side has always been the side in contact with the platter and capstan. This surface required a certain tension, no more and no less, to perform at its peak. Over the course of a few weeks or months the metal would have worn sufficiently to cause slippage. So many of us would just make a new belt and sit in wonder at the improvement of a fresh belt. Little did we know that we should have gone in the other direction. But, had we found that Doug and Paul wouldn't have had anything to tinker with. With Paul's new trick there is no more slow erosion of the metal surface and it seems that this textured surface is superior to the old metal surface in every way.

Someone else may have a better method but I'll share mine. I use a plastic pan that is maybe 14x12x8 to contain the etching creme and wear heavy pvc gloves (Home Depot). Remember, it is the side the tape wants to curl to. Since I'm right handed, I start with one end of the tape in my left hand. I dab on the creme using a 3/16" artist's brush to get the end. Then I load the brush, hold it against the tape and pull the tape past the brush with my left hand leaving the treated tape in the pan. You can see the metal vanishing while you're working. Then I take the pan, brush and gloves outside to the faucet to rinse. It really is easy, using a little caution, and takes only a few minutes. The curl is still there, but you won't have any trouble telling which is the smooth side and which is the textured side.
For those looking for another color in splicing tape I got some Professional Archive splicing tape for 1" and 2" recording tape from US Recording and it is a nice shade of blue and works very well.
Here's another source for splicing tape:
a little cheaper, for 100' vs. 82' on the other site I gave you above. I don't know if one is better than the other..

I've never seen splicing tape in anything but white... but I haven't seen everything...

Frank, to answer your question about stripping. It's done on the side where the silver is: if you do it on the other side, nothing will happen. That side then becomes the one in contact with the pulley and the platter, the "inside" side, as you assumed.
Hello Doug you sent me a link to McCormicks for mylar streamers a good while back. What i am using does not appear to have a silver metallic coating. I ordered black, i'm getting great use from it. Do you think i'm using the wrong mylar streamer?
Many thanks for the great tweak and have a good one.
Great description of the stripping process by Dan_Ed above and Jfd01 is correct, the etching cream has no effect on the other side. Mylar is impervious to most acids.

The black mylar streamer from McCormick's is just 1.0 mil thick. That's what Teres supplied with our first table five years ago. It's notably inferior to several materials we've used since, and it won't respond to this tweak.

Chris Brady replaced that black 1 mil material with clear 2 mil mylar belts several years ago. That was a notable upgrade and I remember sending him a big thank you.

Then we tried the silver holographic mylar streamer (also sold by McCormick's), partly because Dan_Ed and others compared Galibier & Teres tables found the Galibier more dynamic. There was little logical reason for that, other than motor-platter linkage, and we did indeed prove that the silver holographic mylar vs. Chris's clear mylar was the difference. We used the silver/holo for about 2 years, with frequent replacements due to the silver wearing off as Dan_Ed described.

Then Paul had the epiphany described at the top of this thread and all is good, very good.

Summary: the black mylar from McCormicks is at least three steps below the belt we're talking about on this thread. If you want a HUGE improvement without all the baby steps, buy a roll of the silver holographic streamer from McCormicks (make sure you don't order the adhesive backed by mistake). Strip the silver off as described above and be prepared for a shock. :-)
Stltrains, the McCormick's item number is 6033665, 1/2"x100' holographic mylar. You're going to love this upgrade.
Dan thanks for the number. With the threat of a hurricane
visiting the area we are in serious planning mode. Hopping this storm will miss the main land. The next few days will give us the good or bad news.