Your opinion about wow&flutter on walkmans

I am curious about what your opinion is regarding wow&flutter on a walkman.

I started to ask myself some questions about what figure should a walkman get, especially considering that most belt-driven ones do no have a specification in the service manual (which is why I don't have high expectations).

@CDV stated in multiple topics, that according to his ear, an acceptable value should be 0.1% - 0.15% or below. However, I did not manage to get that value on anything but disc drive (DD) walkmans. Of course, I am talking about fully serviced units with everything lubricated, cleaned and new rubber.

The best I have got is on the WM-D6C that gets 0.04% WRMS. But even the WM-D6C rises considerably (above 0.1%) if I remove the PLL by desoldering the pad on the PCB. This makes it obvious that the quartz lock contributes significantly to a low wow&flutter (which is what I would expect, that is what it was designed for).
The WM-D3 on the other hand, which doesn't have a quartz lock, but uses the same governor IC (CX20084), has a wow&flutter between 0.08% and 0.1%.

@Simon Green was stating in a topic about belts that he found some belts available on the market to be so and so in terms of uniformity. This made me question the quality of the belts available.

I can confirm that on some walkmans (AIWA HS-P107, AIWA HS-PC202 MK III Dolby C version, Pansonic RQ-S25) that still have the original belt in good condition do obtain a lower figure (about 0.15%) than all the other ones which had their belts replaced (which go closer to 0.2%).

@CDV also stated low figures on a walkman that was never been serviced and still has the original belt.

All of the above made me question: what a typical figure should be ? What should the difference be between a belt driven walkman and a disc drive one ? By how much should quartz lock improve this ?
Of course, on some models it may be impossible to measure what the figure was with the original belt (or other rubber parts), because that's long gone in many cases. But maybe some of you who remember from back in the day can tell.

To my ears, a 0.2% sound ok, but others may have different opinions. I can hear the difference between a 0.2%, a 0.1% and a 0.05%, but consider that the latter is more of a high-end spec and not a necessity.

 

@Valentin, before the votes start to pour in, maybe we should decide on the units. I've researched the topic a bit, and there are/were many standards governing W&F, in particular NAB (U.S.), DIN (Germany), IEC (international; France used it), JIS (Japanese), as well as Soviet ГОСТ 24863-87 that as far as I understand is "aligned" with IEC. They differ in what frequency signal is used for testing, how weighting is done, is it root mean square or peak-to-peak, etc.

In particular, JIS uses RMS measuring; weighted RMS is abbreviated W.RMS or WRMS.

DIN uses Peak-to-Peak measuring, numerically this value is usually about twice higher than WRMS value.

So, we need to decide, whether we measure in WRMS, which is what JIS uses, or in Peak-to-Peak, which is what DIN uses. For example, my Technics RS-TR333 says, 0.1% WRMS or ±0.2% DIN.

I am fine with WRMS.

 

I don’t know if I can say what level of W & F is acceptable, because I have never measured it. One thing I can say is, I know when it’s off but at what figure that is I can’t say.

 

@CDV That's a very important point regarding the standards. For example WFGUI (that we probably all use) can measure WRMS according to DIN standard, or at least that's what is written on the interface.
But according to what you are saying, when measuring weighted RMS, you can only do that according to the JIS standard (a standard that I have never seen in service manual specs).



For example, the spec on the WM-D6C is given either as a peak-to-peak value according to the DIN standards, or as a WRMS value according to NAB standard.
It seems that both NAB and JIS uses weighted RMS and certainly it's a different weighting procedure on each.

So what does WFGUI reads when set on "DIN" ? It clearly displays 2 values, one peak-to-peak and one WRMS. Are both displayed according to DIN standard ? (I assumed that, given what the interface says, but I may be wrong in my assumption)
I am also fine with WRMS because it's a more consistent value being RMS.

@TooCooL4 It would be interesting to measure it: you can record you own test tape if you have a deck with 0.05% WRMS spec. The measuring can be done with WFGUI and a computer sound card.

 

Valentin not sure I would want to bother checking the W &F, if I can’t hear it no need to bother. I have Nakamichi CR-7, Sony WM-D6C, Sony WM-DC2 & Sony WM-DD9, all good units that have W & F that I can’t hear.

 

I understand, I thought you may have other belt driven units. The ones you mention all have very low wow&flutter, would guess something like 0.05% WRMS or less. So there's nothing to measure there, they're all really good units.

 

I do have an Aiwa HS-PL707 which is belt and the W & F is terrible, it's only usable while stationary.

 

See this doc for some info. Alex Nikitin ("A.N.T.") made several comments on another forum, from which I gathered my initial idea how the tool works, I also read some of the standards as well as several user manuals for W&F meters. I thought I learned enough to write a descriptive user guide or to make a video but the spec above with NAB using WRMS has knocked the soil out from under my feet, because the manuals that I studied cursory mentioned that NAB used average values, that is, arithmetic mean.

I believe that DIN uses 3.15 kHz signal and JIS and CCIR/EIC uses 3 kHz signal.

The needle jumping around in the bottom right of the app is Peak-to-Peak value. If you use 3 kHz tape, then technically you cannot measure DIN value, because DIN requires 3.15 kHz tape. Also, weighting is slightly different. But I suppose that difference in numbers is not as significant as difference between Peak-to-Peak and RMS. Also, notice it is a "±" value, not just a number. RMS is just percentage.

If I understand correctly:

- if you use 3 kHz tape and select "DIN", then Peak-to-Peak is not exactly DIN, and RMS is not exactly WRMS/JIS.
- if you use 3 kHz tape and select "JIS", then Peak-to-Peak is not exactly DIN, but RMS is WRMS/JIS.
- if you use 3.15 kHz tape and select "DIN", then Peak-to-Peak is DIN, while RMS is not exactly WRMS/JIS.
- if you use 3.15 kHz tape and select "JIS", then Peak-to-Peak is not exactly DIN, and RMS is not exactly WRMS/JIS.

This is what I came to understanding after reading through some manuals. So, I would select "JIS" to measure WRMS using 3 kHz tape that I have. But A.N.T. suggests choosing "DIN" instead, and I am not sure why. He sells both 3 kHz and 3.15 kHz alignment tapes. Maybe ask him on another forum, I am effectively banned there.

EDIT: typing this on a Mac, so have no access to WFGUI. It seems that it has no "JIS" selector, and this is what A.N.T. recommends, "JIS(A) = WRMS = RMS value with DIN filter enabled in WFGUI". Does not make a lot of sense to me, but he is supposedly a pro.

TL;DR: if we are comparing WRMS, then just remove (DIN) from the above selections I will make a more concise table later.

 

I can't seem to be able to edit the response choices, only to add additional ones. Thanks for the manual for WFGUI, I was looking for it, but could not find it. I will read the manual, as I am also trying to understand more clearly what the deal is with those standards.
Because at this point I'm not really sure if I'm doing the measurement correctly. I think this is where a dedicated wow&flutter meter would be handy, I would trust that much more than WFGUI. I am tempted to build one myself, as I have found schematics of old meters online.

I also noticed something about the software: if the speed is not close to 3Khz (let's say 3.05-3.1Khz), it will read higher wow&flutter, although speed deviation is an independent parameter.

However, from what you're saying about the interface, I suppose the interface itself is not very well designed and labeled. That's why it's very misleading in my opinion, because you select DIN, but the WRMS value is displayed according to JIS, not DIN.
If I made that software, I would disable the WRMS reading entirely when "DIN" is selected and add another item in the drop-down menu labeled JIS. And when you select JIS, I would disable the peak-to-peak reading.

From your YouTube videos, I have seen you are using a 3Khz test tape, as I do, so in theory our readings are made in the same test conditions. I will record a 3.15Khz test test just for the sake of it and make the measurements that way, I'm really curious what the difference will be...

 

Maybe we can start a new thread for voting, and discard this one? Or you can self-report your message to the admins and ask them to update the choices.

DIN does not measure RMS at all, weighted or not. It is spec'd for Peak-to-Peak. Hardware meters are switchable between standards, like JIS, NAB, DIN, CCIR and separately between what is measured, like WOW, FLUTTER, WOW & FLUTTER or WEIGHTED WOW & FLUTTER. WFGUI shows Peak-to-Peak and RMS at once, no need to switch back and forth, because as I understand, the difference between (W)RMS and Peak-to-Peak is much larger than difference between RMS values measured at either 3 kHz or

Yep, I use a 3 kHz tape, which is perfect for (W)RMS. But I think it won't be a huge difference if 3.15 kHz is used.

I am going to make a combined table with differences. Internet is full of info, I am surprised the Wikipedia article has so little on the topic.

P.S. So far, all the manuals for W&F meters say that NAB value is "average value", but one article reviewing a meter in 1976 says that the meter displays NAB "effective value", which usually means RMS, so it is weird. Pioneer article from 1980 quotes 1965 NAB standard and says "average value". So, no clear understanding regarding NAB, but we probably should not care about it, after all it was mostly used for R2R.

 

I reported, we'll see if an admin can help. At this point I say we should keep this topic, as there are not many responses. The only 2 votes are yours and mine, so not very relevant.



The manual of the WFGUI confirms what I have noted about the speed deviations. The speed deviation is not independent of wow&flutter. I have clearly noticed the closer I adjust the speed to exactly the 3000Hz, the lower the wow&flutter displayed.
I'm not sure if that's how it should be or if this is a limitation of this software, that hardware meters do not have.

Regarding the NAB standards, it is possible that those articles do not use the correct term and by "average" they want to say "RMS". After all, the RMS is an average, it's just not the arithmetic average.
All SONY service manuals that I have read give the WRMS spec according to NAB standard and it's unlikely they're all wrong. The question is what is how big of a difference is between WRMS (JIS) and WRMS (NAB).
My WM-D6C is exactly spot on if I take the WRMS value, at 0.04%. If I take the DIN peak, it's lower than the spec, at about +/-0.1%.

 

I want to add another reference: the JVC CX-F7K that I repaired recently. I measured the wow&flutter after running a couple of tapes in it and the result was 0.14%-0.16% WRMS.
The belt used is the one from DeckTech, which are the only ones that have one specifically designed for this walkman. And given how difficult is to replace it, I didn't want to mess around.
Have to say the result is better than I expected on this one.

This is the link with the restoration: https://stereo2go.com/forums/threads/jvc-cx-f7k-restoration.7222/

 

I have another reference measurement, this time on an AIWA HS-P303. The unit was cleaned, capstan bearings and table reels luricated, pinch rollers cleaned and capstans roughened with ferric chloride.
It is to be mentioned though, that the belt and pinch rollers are original.

 

@Valentin, here is the core portion (on this forum) of my opus magnum on Medium (friend link).

TL;DR: WRMS (DIN) makes no sense @Mister X, please remove the "(DIN)" portion from the poll choices above.

@Valentin, the WFGUI indication looks something like "WRMS (DIN)", but it actually is not. "DIN" corresponds to what the needle instrument shows, while RMS corresponds to JIS value. WFGUI provides two measurements at once: DIN and JIS. The UI can do better.

 

Interesting read on Medium, there are all the details one needs to understand wow&flutter and all other notions related to it. That's a good reference !
A good reference about DIN vs JIS is: DIN is always a +/- value, while JIS is always positive value and is lower than the DIN value.
Yes, about the WFGUI, it's the UI that's problematic. For example the DIN RMS reading shown above the needle, because it's not labeled "WRMS", one could draw the conclusion it's not weighted, but the raw RMS value.

In regard to the value itself, it does depends considerably on what music one listens to: it's the most obvious on music with long constant tones (by long I understand more than one second). Long tones can be everything from vocals to piano.
On this kind of music, everything above 0.1% WRMS is going to be noticeable and obviously the higher we go in value, it will approach an unacceptable level very fast.
On the opposite side of the spectrum, music that is very fast, with constant tones lasting shorter than 1 second, wow&flutter may only be noticeable at the 0.3% WRMS mark. On this kind of music, 0.2%-0.25% WRMS sounds ok.

According to a paragraph in the Medium article, 4th class units (personal stereo and dictation machines) are specified at +/-0.4% DIN, which means a little higher than 0.2% WRMS.
I also want to re-iterate the fact that many belt-driven walkmans out there do not have a W&F specification in the service manual. What does that mean ? To me, it means those fit in the 4th class. Otherwise, the manufacturer would have put a spec.
Of course it is up for debate in which category a walkman should fit, but my opinion is the following:

- the SONY DD with quartz lock can be included in the HIGHEST CLASS;
- the SONY DD without quartz lock will be somewehere between FIRST CLASS and SECOND CLASS;
- very good belt driven ones will be somewhere in between SECOND CLASS and THRID CLASS (more towards the THIRD);
- many belt driven walkmans will actually fit to FOURTH CLASS (or somewhere in between 3rd and 4th);

 

Hardware meters have two sets of buttons. One is MODE or FUNCTION. Another is INDICATION.

WFGUI does not have indication selector, because you cannot switch between DIN, JIS, IEC, NAB and whatnot. WFGUI always uses DIN and JIS, simultaneously.

But it does have function selector, where you can choose a weighting curve out of the four. Three of these curves are weighted, and only the "Unweighted wow and flutter" is unweighted. So, if you choose "Unweighted wow and flutter", both RMS and Peak value are unweighted, if you choose any of the other three options including poorly named "DIN", both RMS and Peak are weighted. It would be much easier if the label changed accordingly, this is a software app after all, not a physical box.

I need to add this to the article, thanks!

 

After reading this thread and most of the article
on Medium, I definitely need to get a cassette to measure my Technics deck.
I would also be interested to find out the value of the DX100, as the mechanism is using a belt from the disc driven brass rotor.
As @TooCooL4 stated, I can’t audibly detect any wow and flutter, but given I replaced the belt recently and one of the pinch rollers needs to be replaced, pre and post values I would like to log.

 

My cheap boombox and walkmans:
Sony CFM-130: 0.07% WRMS, ±0.17% DIN
Sony WM-FX455: 0.22% WRMS, ±0.42% DIN (belt from FixYourAudio)
Sony WM-FX435: 0.15% WRMS, ±0.33% DIN
Sony WM-FX290W: 0.17% WRMS, ±0.33% DIN
Sony WM-FX315: 0.19% WRMS, ±0.39% DIN (belt from FixYourAudio)
Sony WM-FX407: 0.23% WRMS, ±0.45% DIN (belt from FixYourAudio)

My "black plastic crap" double-cassette deck:
Technics RS-TR333: 0.07% WRMS, ±0.24% DIN (interesting that here DIN = 3×WRMS, not 2×WRMS as for most other players)

 

Thanks for measuring and sharing @CDV .
While searching for information for the DD line and potentially related devices, I noticed none of the others (and even the DX100) had W&F specs shared.
Compared to the DD line they seems relatively high, but what is your experience when using them? Above which value it becomes noticeable or unusable!

 

@Emiel, as I mentioned above and chose in the poll, I think that 0.15% WRMS is the highest tolerable value if I want to listen to everything, from hip hop to classical. Between 0.15% and 0.2% is a gray zone, where most pop stuff is sort of ok, but classical is not. Above 0.2% WRMS it becomes too noticeable to me even on pop music. Above 0.25% is intolerable. The FX435 is the only one out of these on which I can listen to Grieg and to Sarah McLahlan, on all other walkmans I hear wobble. If you click on the FX407 video you'll hear Grieg playing, how do YOU feel about it? I think it is quite noticeable.

Switching to CDs and then to file-based digital was a blessing. But CDs and especially files are not as sexy as some cassettes. Also, the whole process of loading a CD into a portable player is undignified, which is why I liked MiniDisc players, they load sort of like cassettes: clunk! open, vzzh! slide in a cassette or minidisc, clunk! close. Anyway, 0.15% WRMS is what I would pay money for, I mean if I were paying more than $20 I want to buy a really good player: beautiful, solid, with good sound quality, compact, metal, etc, and then sell all the cheap ones.

But notice the CFM-130: wow, no wow!