Excessive Wow & Flutter in Sony WM-2 (0.78% RMS with 8Hz Spike)

Hello Walkman Wizards,

I recently picked up a Sony WM-2 to recreate a childhood memory. It needed service, so I dove in completely.

I followed the standard overhaul procedure for this model (as described in great detail by @Valentin). Here is a breakdown of what has been done so far:

Center Gear & Clutch: This unit has that thick plastic center gear and is not cracked. I still completely disassembled it and polished the clutch to remove all patina, just in case.

Mechanicals: Cleaned and relubed the capstan bearings, greased the appropriate points, and added a few drops of oil to the motor shaft.

Belt Path: Polished all surfaces that interact with the belt. The transport runs whisper-quiet and feels mechanically excellent.

The Belt: Replaced the old belt with a brand new Russell Industries SCY5.2 (IC=5.2", C/S=0.035", Wall=0.035") from their PRB line catalogue, which corresponds to Sony part 3-578-222.

Tape Path: Kept the original pinch roller as it looks and feels okay with no obvious deformation. The tape path looks correct when checked with a mirror tape. While I don't have a torque test tape, the take-up tension feels right in my fingers; it is on par with my other decks.

The Problem
Despite the smooth mechanical overhaul, I am getting an incredibly high Wow & Flutter reading of 0.78% RMS. Yikes! When looking at the speed analysis, the dominant spike shows up right in the 8 Hz domain, which points toward a capstan-speed issue. By the way, I did not do a Ferric Chloride etch on the capstan, so it remains smooth and shiny like a standard cassette deck's shaft.

Questions for the Experts:
The Belt: Would sourcing a specialized belt (like from Marian's FixYourAudio) make a significant difference for this 8 Hz spike?

The Pinch Roller : Could a new pinch roller help this?

Any insights or troubleshooting points for the WM-2 transport would be greatly appreciated!

Thanks!

 

Update: After a day of unsuccessful tinkering with the usual suspects, my W&F was still sitting stubbornly high at 0.67% RMS. I finally decided to pull the belt and manually recalculate its inner circumference (IC).

Bingo—it turns out the Russell Ind. PRB catalog specifies the wrong size for this model. An IC of 5.2" is simply too short and causes excessive tension on the transport. A 5.4" IC is a much closer match for the required 44mm diameter path.
Lesson learned: trust, but verify! At least the wall thickness was a correct 0.035".

Expecting a package from Marian at fixyouraudio with the right stuff in a couple of weeks.

 

@euroflash: Alex - I have purchased from Marian for one of my Sony Walkmans, but haven't installed the belts, yet. I, also, got four left pinch rollers coming from him for my Sony TCM-200 tape transports on my Sony TC-K950ES and TC-K700ES decks.

Nando.

 

One thing I want to add: the belts from Russell Industries are one of the best you can get. But of course the diameter has to be right: you have to measure to double check, wouldn't rely on SONY part numbers alone.
I don't have a measurement specifically for WM-II, but I think it should be close to TPS-L2 which is 45mm x 1mm (inner diameter x thickness).

Wouldn't replace the roller if it looks good, on this mechanism excessive W&F is rarely caused by pinch rollers unless they are visibly damaged (glazed, hard as a rock, indentation mark).
Usually this is more of a problem on walkmans with very small rollers (such as WM-EX series), in such walkmans the roller pressure is also smaller so even a roller that visually looks good can cause traction problems.
Also keep in mind some of the new rollers available have eccentricity, so that's something to be check if a new roller is installed as there can be situations where replacing it adds a problem that wasn't there before.

 

@Elite-ist Nando, I’ve installed pinch rollers from Marian on my Sony TC-K679ES and AKAI GXC-760D, and his products are always of excellent quality.

@Valentin, Thank you for the insight. The service manual for the WM-2 specifies 44 x 0.9 mm, so this new length should be a perfect fit. I’ve ordered many belts from Russell Industries over the years, and they’ve always been great for my decks and boomboxes—this was entirely my mistake for relying blindly on their catalog number for this specific Sony.

Thanks also for the heads-up regarding eccentricity in some new pinch rollers; that is definitely a good variable to keep in mind. I'm crossing my fingers that the drop in tension from the correct belt is all it takes to tame this high W&F. This is such a nice Walkman, I'd hate to see it end up as a "souvenir" on a shelf!

 

I received the proper 44mm belt and a new pinch roller from Marian (as always, top-notch quality).

The new belt definitely improved playback stability, but the overall Wow & Flutter is still sitting high at around 0.38% – 0.42% RMS. While it's a step in the right direction, it's still way too high for a WM-2. Swapping in the new pinch roller made no difference at all, just like @Valentin suggested.

Digging deeper, I pulled the mechanism apart again to check for the clutch binding, re-cleaned and relubed the capstan bearing. I lightly sanded the clutch surface and the flywheel with 2500-grit sandpaper, but everything looks clean and spins OK, with some friction.

The dominant 8 Hz frequency spike remains, and that points to the capstan. Is it time to try a Ferric Chloride etch on the capstan to give it some traction, or is there something else in this transport I might be overlooking? I'm running out of ideas here.

 

Thinking about it, realise have seen the flywheel drive gear available online. EDIT: https://www.ebay.com/itm/166988283963
There is even a thread about this here: https://stereo2go.com/forums/threads/desperately-seeking-a-wm-ii-flywheel-capstan.9984/

So that would be the next obvious thing to check.

 

@Valentin Thanks for the reference to the thread. I appreciate it. I have read it earlier.

The whole thing befuddles me. I can’t see any visible wear on the mechanical parts; all the gears are intact, so it really should be working fine. This is my third time dismantling the mech, and now I have completely torn the unit down—literally to the nuts and bolts—to hunt down whatever is causing drag in the transport. This time, I did a deep clean of the capstan bearing and hit absolutely every single part that spins or moves. The mechanics of this WM-II are very sensitive to something.

I’m reassembling it today, so fingers crossed.

 

There are really only 3 other things that can cause increased W&F:
1. Worn capstan bushing. This is a common cause of increased W&F on belt drive units.
While usually a problem mostly on thin players with small capstans, it's something to be considered.
It can best be felt where there is a no oil inside, so for testing purposes you may clean and install it (without the retainer) completly dry.
2. Motor problem. While motor usually creates flutter, problem with the magnet or a resonant frequency in the rotor can cause a lower frequency peak.
These MNF-1600B unfortunately are not the most reliable motors.
3. Servo problem. Least likely, it's something I would consider only if the other 2 are eliminated.
However cleaning the speed adjustment trimpot with contact cleaner is something that can be done before eliminating the other ones.

 

@Valentin , thank you once again. I appreciate your expertise.

I reassembled it yesterday, and all that work made no difference whatsoever. I will continue to track for 1,2,3 possible causes.

 

@Valentin, I did some current draw testing, and the results are interesting. The Service Manual (SM) specifies:
Play Mode: 115 mA , FF / REW Mode: 120 mA

My Readings (without a cassette): Play Mode: 125–135 mA (fluctuating), FF / REW Mode: 168–179 mA (fluctuating) These current numbers seem high. What do you think of these readings based on your experience?

Note: Current increases even further when a cassette is actually loaded. Without the belt installed, the motor spins at 60 mA in Play and 60–66 mA in FF/REW.

A couple of other details:
FG Coil: It shows a clean AC sine wave on the scope and a resistance of 104 Ohms. That side of things looks solid to me.

The Clutch: Could a sticky clutch be causing this excessive drag? I polished the surfaces (see the before/after photo below), but now I'm second-guessing myself. It rotates smoothly but with some resistance. Since I don't have a known-good frame of reference, I'm not sure exactly how much resistance is normal.

I haven't yet looked at the Servo circuit deeply. I need to find Supplement #2 for this generation walkman, as my player has 1 chip 301, and the main SM schematics cover a 2-chip version. Very different design. Nothing is easy, isn't it?

The search for the culprit continues. Once again, thanks!

 

In play you have a 10-15% increase in current draw, which is indeed on the high side (fact it's fluctuating is more of a problem). On FF/REW you have a 30-50% increase, which is significant.
There's something going on here, would check how easy reels turn given the excessive FF/REW current. However motor could be the culprit here.

Motor is drawing 60mA with no load at all ? Can you measure the motor without the rest of the circuits, desoldered ?
Motor should draw 10-20mA with no load, 60mA indicates motor internal friction. Does the motor turn free by hand (in both directions) ?

A frozen clutch can create excessive drag, but if that was the case you would see:
- very high takeup torque, to the point of tape slipping past pinch roller;
- current consumption would jump only with a cassette inserted and would be normal without (which isn't the case here);

 

Thanks, @Valentin . I went ahead and desoldered the motor to test it completely free of the circuit. Without any load, it draws a steady 14.26–14.36 mA with just a minor 0.09 mA fluctuation, and it spins completely free with no perceptible stickiness. So it looks like the motor itself is healthy, and the high current draw is definitely coming from drag elsewhere in the transport.

However, I did find a strange thing during my last teardown. While cleaning the parts, I looked inside the clutch shaft and discovered a tiny, loose ball bearing stuck to the wider, lower side of the shaft. In this picture, it is near the PSB screw to show you its size. I managed to pick it out with a toothpick. I have no idea how it got there or if it's an accidental "migrant" from another part of the mechanism. I see nothing that small on the schematics...

I'm going to tear the mechanism down one more time to carefully double-check the reassembly and see if I can pinpoint where the friction is happening.

Thank you for ruling out the clutch assembly—that is a spot of relief!

 

Just a casual observation, I do not see the film washer that goes between the center gear hub and driving gear assembly in your image. Just want to ensure that it's not missing. Cheers. Norm

 

@worldwide137 Norm, thanks for pointing that out—there is a red Teflon washer seated between the center gear hub and the magnetic driving gear.

Quick project update: I picked up a second, non-working WM-2 Gen II to establish a comparative baseline. My goal was to compare the capstan, central gear assembly, bearings, and motor. The serial numbers are fairly close, and this donor unit also has a solid plastic central gear, but it came with some battery acid corrosion. Nothing catastrophic—it ate through part of the paper shielding, but the battery contacts are salvageable with a bit of work.

I also managed to source Supplement #2 for the WM-2 Gen II. It turns out the Gen II features a completely different electronic architecture compared to Gen I, along with different specs, voltages, and parts. For anyone interested, the current specs for Gen II are: Play: 110 mA, FF: 116 mA, and REW: 112 mA.

When I swapped the capstans, I discovered that the red thrust washer that sits between the flywheel and the bearing was missing from my original project's capstan. Unfortunately, this detail wasn't shown in the schematics, so the donor unit already proved its worth! After swapping capstans and installing the washer, applying some additional lubrication to the central gear (3 points), and reassembling the transport, current draw dropped right into spec. However, Wow & Flutter is still high. On top of that, swapping the capstans introduced a strange new quirk: the player now struggles with C-60 tapes, yet plays C-120s (!?). Pinch roller pressure measures right on spec at 170 +/- 5 g.

On a positive note, the donor unit’s motor is in good shape—it draws just 9 mA unloaded and desoldered. However, it has a cracked supply reel gear (as expected), so I’m currently waiting on a package from Marian at fixyouraudio.com to finish reassembling the donor mech and test/compare its full functionality. Next, I'll try swapping the central gear assembly, bearings, and motors to see if any of them help. If not, time to start digging into the servo.

A side note: I now understand why restored WM-2s, WM-5s, and DD series units command such high prices. If luck isn't on your side—which seems to be the case more often than not—you practically need two players to build one good one.

Thanks for all the help!

Alex

 

Yes, I recently restored one and aside from one minor issue, the WM-2's are quite remarkable. I record my own bias adjusted cassettes from vinyl with Dolby B. I know that the WM-2 does not have Dolby capabilities, but the Dolby B boosts the upper frequencies just enough to bring just the right amount of clarity to the upper frequencies that the WM-2 can sometimes lack. After the full spa treatment [replaced caps, belts, clean/lube, speed and azimuth adjustments], the audio was outstanding while listening through the Koss Porta Pro headphones. But, you are so correct, it just about takes two units to make one good one - well, if you're lucky. Cheers.

 

@worldwide137 Norm, I am curious what W&F readings you are getting after your restoration? Did replacing the servo caps help lower it at all?

Yes, the build quality of these players is impressive. Sony seemed to spare no expense in design; they are built to last. The sound is good too, particularly on the low end. I stopped using Dolby in recordings some time ago, so this feature is a non-factor for me.

Thanks!

Alex

 

I get around 0.3% WRMS. The unit was non-functioning when I got it so I don't know what the W&F was beforehand. I know the 0.3% sounds bad at first glance, but the peak frequency is in the 'wow' region vs. 'flutter'. Human hearing doesn't pick up on it as noticeably as in the flutter range. For example, you don't hear any of that warbly sound in sustained piano or guitar sounds. That low end is tight and clean. Also, you may already know this, but using higher impedance headphones will make the low end come out more strongly. The Koss Porta Pro, which I highly endorse, is at 60 Ohms. The WM-2 has plenty of power to drive that. If you haven't tried it yet, I would encourage to try listening to a Dolby B encoded recording on the WM-2. It adds the right amount of clarity to make the sound almost 3D-like. Cheers.

 

Thanks, Norm! That 0.3% WRMS gives me a much more realistic target to aim for—I had a feeling getting anywhere near 0.1% RMS on the WM-2 was wishful thinking.

I’ve been using my Technics RP-DJ1200s (32 Ohm) for so long that I'm pretty set in my ways, but I have plenty of Dolby-recorded tapes and will definitely give one a shot as soon as I get the player running smoothly.

Alex

 

Hi Alex! No original W&F specs were published for the WM-2 by Sony or at least made public. I believe the general consensus from the community for that time period was around 0.20 - 0.25% for this class of Walkman. While my 0.3% is not something to brag about, I consider it not bad and luckily for me does not interfere given the spike is in the low frequency range. Given how meticulous I was about the cleaning and lubricating, it could possibly be the belt, maybe. I generally order my stuff from FixYourAudio, but the belt on this particular one came from another well known and highly regarded US-based retailer from eBay.

Going back to one of your earlier posts above, that mystery ball bearing could possibly have come from the cassette door mechanism. When you open the cassette door, you can feel a specific tactile click when fully opened. There are two spring-loaded ball bearings, one on each side of the cassette door mechanism that is responsible for this tactile feel. Cheers! Norm