Wow & Flutter Meters

[JC]

So, hoped to have left all this mechanical error measurement stuff
behind me back in the 80's but I guess with the renewed interest in
vinyl its all coming out of the woodwork.
Leader LFM=3610 W&F meter, customer wants it calibrated or at least
verified. don't have schematics or full cal manual. Manual I do have
suggests use of Leader LFM-7000 W&F calibrator, don't have that either.

So........ How can I reproduce some of the cal signals, for example JIS,
1%, 4Hz modulation at 1vrms?

JIS is 3KHz sine, so no problem, but how to add the 4Hz modulation? FM
obviously, (AM would be easy), but how? Can't be a 4Hz sweep on 3Khz,
can't be a 4Hz burst? Superimposed 4Hz on 3Khz? How to quantify 1%?

I have a wide range of hardware and software signal gens available but
no idea how to mimic this accurately.

Any of you audio guys have digital waveforms/ test cd's handy you can
share? Help!
TIA

[Clifford Heath]

AD9851 module and an Arduino? It probably wouldn't even be necessary
to calibrate the primary crystal source, or switch to a TCXO.

[Phil Allison]

JC wrote:
>
>
> So........ How can I reproduce some of the cal signals, for example JIS,
> 1%, 4Hz modulation at 1vrms?
>
> JIS is 3KHz sine, so no problem, but how to add the 4Hz modulation? FM
> obviously, (AM would be easy), but how? Can't be a 4Hz sweep on 3Khz,
> can't be a 4Hz burst? Superimposed 4Hz on 3Khz? How to quantify 1%?
>

** 1% is the frequency deviation, 4Hz is the rate.

So +/- 30Hz deviation four times a second.

( The ear is most sensitive to a 4Hz rate)

>
> I have a wide range of hardware and software signal gens available but
> no idea how to mimic this accurately.
>

** Some function generators have an external modulation input.

Set the generator to 3kHz and use a small DC input to get 3030 Hz.

A 4Hz sine wave with the same peak value is then required.



... Phil

[Black Iccy]

On Sun, 17 Jul 2016 19:27:04 -0400, JC <Chipbee4...@yahoo.com> wrote:

>Any of you audio guys have digital waveforms/ test cd's handy you can
>share? Help!
>TIA

Phil just clarified your requirements.
You can generate a waveform tone via Adobe Audition
(Ver 3 or earlier) assuming you can find a copy (Win).
Save as Wav or whatever.

Base Mod by Mod freq (rate)
3Khz +/- 'x'Hz @ rate 'y'
3000 30 4

[JC]

thanks Phil, I'll try it.

[JC]

Thanks. Will try it.

[JC]

On 7/18/2016 12:10 AM, Black Iccy wrote:

Thanks to you and Phil for the info, proved to be very useful and Adobe
Audition seems to be the way to go.

Couple of problems arose.

I need 3 test signals:

JIS (3KHz) 1% @ 4Hz
CCIR (3KHz) 1% @ 4Hz (Same as JIS? Am I missing something here?)
DIN (3.15KHz) 1% @ 4Hz

These should, according to the limited service info I have, produce a 1%
reading on the meter for each relevant setting (JIS, CCIR,DIN)

JIS was giving me 0.5% so I read up on the Adobe Audition settings and I
guess i needed mod by 60 hz, (+/-30Hz). That gave me 1% display with JIS
but then doubles up on the CCIR/DIN giving 2%

JIS is supposedly an effective mean reading, CCIR/DIN are Peak, but I'm
not making sense of these readings, any ideas?

Amplitude of the input signal has no effect as long as its large enough
to register.

TIA.

[Ralph Mowery]

In article <KL9kz.5883$zu3....@fx21.iad>, Chipbee4...@yahoo.com
says...

>
>
> Couple of problems arose.
>
> I need 3 test signals:
>
> JIS (3KHz) 1% @ 4Hz
> CCIR (3KHz) 1% @ 4Hz (Same as JIS? Am I missing something here?)
> DIN (3.15KHz) 1% @ 4Hz
>
> These should, according to the limited service info I have, produce a 1%
> reading on the meter for each relevant setting (JIS, CCIR,DIN)
>
> JIS was giving me 0.5% so I read up on the Adobe Audition settings and I
> guess i needed mod by 60 hz, (+/-30Hz). That gave me 1% display with JIS
> but then doubles up on the CCIR/DIN giving 2%
>
> JIS is supposedly an effective mean reading, CCIR/DIN are Peak, but I'm
> not making sense of these readings, any ideas?
>
> Amplitude of the input signal has no effect as long as its large enough
> to register.
>
> TIA.

The JIS, CCIR, and DIN are standards set by different countries.

JIA Japan
CCIR Mainly countries of or were of the British Empire
DIN Germany

You probably will not be able to adjust for all 3 countries standars at
once.



---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

[JC]

This is the Leader LFM-3160, there is a switch for each JIS, CCIR, DIN.

[c4urs11]

On Thu, 21 Jul 2016 16:30:56 -0400, Ralph Mowery wrote:

> CCIR Mainly countries of or were of the British Empire

It would be troublesome to image the Empire calling it then
"Comité Consultatif International des Radiocommunications".

Cheers

[Bob]

Are you sure the modulation is a since wave (See Phils post above). This looks like a peak to average of a square wave pulse. I would try an actual analog generator with low distortion (in the 4 hz tone also) modulated and calibrated like Phil suggested. One way to check if I am correct is to listen to the tone and see if its frequency smoothly goes up and down or just snaps between high and low points. If it is hard to hear set the mod rate to 1 hz instead of 4.

...Bob

[JC]

Yes, a possibility but I tried a square wave and I still get JIS giving
half the value of CCIR/DIN. A.Audition seems to generate a decent
waveform, I'm monitoring the signal on a scope so I know what I'm inputting.

[Black Iccy]

On Thu, 21 Jul 2016 15:54:17 -0400, JC <Chipbee4...@yahoo.com> wrote:

>> Base Mod by Mod freq (rate)
>> 3Khz +/- 'x'Hz @ rate 'y'
>> 3000 30 4
>>
>
>Thanks to you and Phil for the info, proved to be very useful and Adobe
>Audition seems to be the way to go.
>
>Couple of problems arose.
>
>I need 3 test signals:
>
>JIS (3KHz) 1% @ 4Hz
>CCIR (3KHz) 1% @ 4Hz (Same as JIS? Am I missing something here?)
>DIN (3.15KHz) 1% @ 4Hz
>
>These should, according to the limited service info I have, produce a 1%
>reading on the meter for each relevant setting (JIS, CCIR,DIN)
>
>JIS was giving me 0.5% so I read up on the Adobe Audition settings and I
>guess i needed mod by 60 hz, (+/-30Hz). That gave me 1% display with JIS
>but then doubles up on the CCIR/DIN giving 2%
>
>JIS is supposedly an effective mean reading, CCIR/DIN are Peak, but I'm
>not making sense of these readings, any ideas?
>

It looks to be a problem of deviation.
Carrier +/- 30Hz (i.e. 60Hz overall) or
Carrier +/- 15Hz (i.e. 30Hz overall)

Though this is not the OP's machine, the numbers referenced
in here (below) imply a 15 Hz departure from 3150 (or 3000)
http://exodus.poly.edu/~kurt/manuals/manuals/Other/LEADER%20LFM-39A%20Claibration%20Procedure.pdf
* See page 3

My suggested Audition generation will do sine/square/saw/inverse_sine.
The modulation is also sine. Write output waveform to a calibration CD.

[Phil Allison]

JC wrote:

>
>
> Thanks to you and Phil for the info, proved to be very useful and Adobe
> Audition seems to be the way to go.
>
> Couple of problems arose.
>
> I need 3 test signals:
>
> JIS (3KHz) 1% @ 4Hz
> CCIR (3KHz) 1% @ 4Hz (Same as JIS? Am I missing something here?)
> DIN (3.15KHz) 1% @ 4Hz
>
> These should, according to the limited service info I have, produce a 1%
> reading on the meter for each relevant setting (JIS, CCIR,DIN)
>
> JIS was giving me 0.5% so I read up on the Adobe Audition settings and I
> guess i needed mod by 60 hz, (+/-30Hz). That gave me 1% display with JIS
> but then doubles up on the CCIR/DIN giving 2%
>
> JIS is supposedly an effective mean reading, CCIR/DIN are Peak, but I'm
> not making sense of these readings, any ideas?
>
>

** You have been misled by the wording of the manual - the test signal itself is not being specified, merely the NAME of the test signal.

The actual test signal comes from a special calibration unit where the varying parameters of the several standards have been taken into account - such that using each gives a 1% reading on the meter when set to the matching range.

Inside the meter, once the raw W&F signal has been derived, it is first "weighted" via a filter that emphasises frequencies around 4Hz and is then rectified and measured via a peak reading or average reading meter. W&F tests often use RMS values too.

The use of a dedicated calibrator or a test CD seem to be the only methods possible for results that reliably comply with the various standards.



.... Phil

[JC]

Yes, that's what I am doing, same thing though, JIS reads exactly half
of CCIR/DIN with either +/-30hz or +/- 15Hz. I checked the components
for the JIS rms section, bridge/cap etc, all ok. I don't see why I get
half on JIS. Something is either wrong with the calibration instructions
or the meter itself. (I am using some of the LFM-39A instructions but I
believe the cal is applicable to the LFM-3160).

[JC]

Quite possible, but I have looked at the cal methods on several W&F
meters and some are explicit on how to generate the signals for DIN. I
would expect an RMS value for JIS to be around 0.7% of the DIN reading
and not exactly half, but without more info or an example from a tape
machine manual giving both figures for comparison I'm probably stuck. I
can adjust the JIS reading to 0.7% but I need more confidence that it
would be correct to do so.

I'm not really sure how much use this meter will be anyway without a
calibration tape. Using the output signal and recording on a machine of
unknown quality then replaying the tape is just compounding errors.

Thanks for everyone's help.

[Phil Allison]

JC wrote:
>
>
>
> I'm not really sure how much use this meter will be anyway without a
> calibration tape. Using the output signal and recording on a machine of
> unknown quality then replaying the tape is just compounding errors.
>
>

** W&F test tapes simply have a pure tone recorded at 3 or 3.15 kHz.

In the real world, people normally record and play back on the SAME machine - so it makes sense to test a recorder that way using a sine wave generator as the 3kHz source. There will be a higher reading on the W&F meter than given in the maker's specs when frequency deviations add in the same direction during playback. If the machine allows simultaneous record and play, that makes the process all much easier.

FYI:

I regularly do repairs on Roland tape echoes ( models 201,301 & 555 ) which all have a pretty high quality transport with ball race tape guides and a direct drive capstan motor with large pinch roller. Testing is normally done with a square wave input at 500Hz, a single playback head selected and the output monitored on a scope. If the generator is linked to the scope's external sync input, any serious W&F is pretty obvious on the screen.

Using a square wave quickly reveals if the high frequency response is good and consistent for all playback heads, so checking the head alignment. Servicing these machines involves a lot of cleaning and polishing of the tape path and heads plus resurfacing the pinch roller when it has become hard and glazed. The tape's back tension is a critical adjustment as it affects both amplitude and W&F in the playback.

Sometimes the main ball race in the capstan motor has to be replaced - when there is severe W&F or a steady tape speed cannot be maintained.

Lotsa fun.


..... Phil