American vs European FM (de-emphasis)

Most FM receivers not designed specifically for the American market sound a bit heavy on the bass end. This is most noticeable as "thumping" on certain types of music. Almost as if there was an un-wanted resonance with the speaker. Tonight's exercise in the shop was a Grundig Majestic model 4192.

Looking at the schematic, The de-emphasis network R80, 47 K and C31, 1 nF have a time constant of 47 uS. (Note, the standard for most of the world except the United States is 50 uS.) Changing C31 to a 1.5 nF capacitor  modified the time constant to 70.5 uS. I suppose I could have used a 1.6 nF capacitor, to make it 75 uS, but I felt that just modifying the original value by 1.5 was adequate.

On a similar note: Using American radios in Europe would tend to sound a bit "crisp" and lacking in bass response. The obvious solution here would be to reduce the time constant of the de-emphasis network from 75 uS to 50 uS.

Pre-emphasis is used at the transmitter end of the FM transmitter to compensate for the inherent increase in high frequency noise in the de-modulated FM at the receiver.  By increasing the high frequencies prior to transmitting, when they are decreased at the receiver,  the additional (unwanted) noise is decreased as well.

Jeff

To thank the Author because you find the post helpful or well done.

Thank you Jeff for this detailed explanation!  This has been an enigma that I even brought to the attention of our respected master engineer, emeritus, Hans Knoll, a year ago and I was not able to get a solution to this problem!

Now that you have it spelled out, a radio tech can't use the excuse, "the speaker has to be reconed"!

This thread you started is invaluable to all of us who enjoy the great mid century Grundigs of the 20'th century but have to suffer that awful FM thump when listening to rock and jazz music.

I have also seen this problem on Loewe Opta radios, perhaps others can add to this thread

To thank the Author because you find the post helpful or well done.

Hello Omer, Hello Gentlemen.

 

i come back to this theme next year here on this place.

All the best and a happy 2007

hans m. knoll

To thank the Author because you find the post helpful or well done.

Federal Communications Commission Pt. 73, Page 133
§ 73.317 FM transmission system requirements.
"(a) FM broadcast stations
......
(e) Preemphasis shall not be greater than the impedance-frequency characteristics of a series inductance resistance network having a time constant of 75 microseconds. (See upper curve of Figure 2 of § 73.333.) [51 FR 17028, May 8, 1986].
....."




All good wishes for a prosperous New Year !
Rolf

To thank the Author because you find the post helpful or well done.

Trust me, those Grundigs thump on FM (here in the States) if you don't do something!

 

To thank the Author because you find the post helpful or well done.

Rolf included a very nice chart from the FCC regarding pre-emphasis.

However, it is a bit misleading. It shows an idealized curve showing a increasing trend for the high frequencies and an (apparently) absolutely flat low frequency response. A better chart would show end to end (overall) frequency response of the combined transmitter and receiver combination.

The de-emphasis network, typically,  is nothing more than a simple RC low-pass filter. Most radios are designed with the overall sound as an objective. This includes the broadcast standards at the time along with other more esoteric effects such as the speaker and cabinet in addition with the overall response of the audio system.

It was my initial assumption that radios designed for the non-US market pre-emphasis time constant of 50 uS would tend to shift the entire response curve down in frequency. This was borne out with a simple single component value change on the Grundig  Majestic 4192 making the radio's de-emphasis match the current standard used in the US.

Jeff

To thank the Author because you find the post helpful or well done.

Hello Friends,

i think thats not true, that the 50usec deemphasis in the US shifts the whole Audiosignal to the lower end.

I bring here the difference between 50 and 75 usec.
You can see, only above 3Kc is an difference, this can be correct  with the Treble-Control.
3 db is a factor 1:1,414


regards  HMK Attachments:

• Deemphasis 50 to 75 usec (48 KB)
• Deemphasis_in details (29 KB)

To thank the Author because you find the post helpful or well done.

My personal view:

To thank the Author because you find the post helpful or well done.

Good morning, Gentlemen,

as far as I can see we have to find out an answer to the question : Is the described "thumping-effect"
a)    a true technical malfunction or
b)    a result of "improperly input for the considered equipment" ?

This could mean in detail

1. is it caused by the wrong deemphasis (a) or
2. by a technical defect of the low-frequency amplifier/loudspeaker-part of the radio which can be maintained (a) or
3. caused by the special kind of music (b), perhaps in conjunction with the sound intensity/volume you prefer to hear ?

According to Mr Knoll's verification table the possibility No 1 can be separated.
So we still have to consider No 2 and No 3. Mr Birkner obviously tends towards b) and No 3 respectively. I myself think that it could be a mixture of 2 and 3, that means an overmodulation of the amplifier plus distortion from the loudspeaker/s in case you hear your special kind of music very loud. What is about classical music with similar intensity ? In principle there should be the same result. Did you try that ?

And are you sure that the speaker/s are okay ? After switching off the radio you could dismantle the bass-speaker/s and push the membrane with one finger cautiously straight inwards (in direction to the permanent magnet). Any "scratching" sound if you move the membrane ? In case there is any noise the speaker/s must be replaced or repaired.

A further possibility could be casing oscillations and/ or resonances of hollow space. Did you check this ?

Please keep in mind : I mainly wanted to exclude possible technical reasons/failures !

Rolf

To thank the Author because you find the post helpful or well done.

To thank the Author because you find the post helpful or well done.

Despite what else has been said about de-emphasis, having the wrong curve for the broadcast material is still an error that needs correction. FM radios that are being used to receive American broadcast standard pre-emphasis of 75 uS need to have the matching de-emphasis time constant.

To thank the Author because you find the post helpful or well done.

The 75 micro-second FM receiver de-emphasis curve chart on page 638 of Radiotron Designer's Handbook volume 4 corresponds exactly opposite (as it should,)- with the transmitter pre-emphasis chart posted above by Herr Nickel.
 I'd post just the chart if I knew how to transfer a scan of it to this forum, but I do not--yet. :-

Is it possible that  an FM radio receiver designed for a 50 ms (Euro) de-emphasis that is reproducing a broadcast using 75 ms (N. Amer..)  pre-emphasis, might produce a bit too much high frequency brilliance and also with a greater chance of passing more h.f. noise?
  A definite answer would be greatly appreciated here.
My few printed references regarding this topic do not address any compatibility issues between the two norms.

 Regards,
 D. Daly

To thank the Author because you find the post helpful or well done.

Minor detail Den, but mS is milli-seconds. uS is micro seconds. But I knew what you intended.
.
From: http://hyperphysics.phy-astr.gsu.edu/hbase/electric/filcap2.html

An excellent explanation of RC low pass filters. Note that the frequency response below the breakpoint F=1/(2*pi*R*C) is NOT a flat line, but continues to curve up. The same thing happens at the transmitter end with the high pass filter.

If you numerically add the two response curves you should get a flat line, IF the two curves have identical breakpoints (time constants). If they do not, you WILL get some slope overall to the response curve.

If you go back up and look at the pre-emphasis chart posted by Rolf, you will see the 3 dB break points at roughly 2.12 KHz for the European 50 uS pre-emphasis, and 3.18 KHz for the American 75 uS pre-emphasis. Note closely however, that as the web site I referenced shows (more clearly) that the audio is NOT flat below the break point, but continues to change as it approaches 0 Hz.

Hans is correct in stating that the differences are slight below the break points.

Bear in mind however, that most FM transmitters are not flat all the way down to DC.  A modern example is the Harris HT 20 CD FM transmitter. http://www.broadcast.harris.com/product_portfolio/prod_media/HAR177339.pdf
With an over all frequency response of 10 Hz to 15 KHz with reference to the specified pre-emphasis curve.

Broadcast equipment from several decades ago, had a somewhat different response at the low end due to technical considerations at the time. And that the FM receivers designed at that time took those limitations into consideration as to their overall sound (frequency response).

One trend in modern FM broadcasting is the use of audio processing on a level unknown at the time. Such as the Orbatron 8500. With a frequency response (relative to the pre-emphasis) from 2 Hz t 15 KHz. This will put a much higher spectral content in the low frequency ranges than what the receivers at the time were designed (and compensated) for.

The simple truth of the matter is that moving the de-emphasis on an FM receiver to match the pre-emphasis used by the transmitter brings the radio much closer to sounding "correct" and it did indeed minimize the "thumping" effect noticed in the radio at various program types and listening levels.

Jeff

To thank the Author because you find the post helpful or well done.

Hi Jeff-
 Sorry for the incorrect abbreviation I used within my last post.
I was hoping it would suffice to convey my meaning since I spelled out the term in the first paragraph.
 The keyboard stroke for the Greek letter used as "micro"  eluded me at that time.....µs.....I've committed it to my poor memory now..:-

 I was basing my thoughts (questions)  regarding the possible adverse reactions after reviewing books on the subject that are admittedly dated for quite some time ago.

  I do realize that varying levels of compression, such as is commonly used in  North American broadcasts may also affect the received signal, along with sometimes other modifying of the generated audio waveforms in order to accentuate (or exaggerate) the bass notes, especially for broadcasts of certain modern music genres.
 It seems that  broadcasts in the U.S.A. can have a noticible variation of audio signals  as long as they remain in compliance with the regulations of the  U.S. Federal Communications Commission.
 (I shall not "speak" for the Canadian or other's regulations)
 
 Another source that served to form my thought is  from a H.W. Sams publication titled "FM Multiplexing for Stereo" authored by Leonard Feldman, E.E.
 It states on page 31 that:
 "At the receiver end, a de-emphasis circuit with the reverse properties attenuates the higher frequencies relative to the "lows" and "mids." The frequencies above 1,500 cycles (1.5khz) are reduced to their original amplitudes."
(The preceding is a brief description or explanation of emphasis in general and not specific to the intent of the book, which is to teach about  FM stereo broadcasting and reception.)

 It is a statement such as this which causes me to think that the emphasis scheme has most to do with the upper frequencies of an FM broadcast rather than the lower ones.

 This is not so much an argument as it is a reasoning for my thoughts on this.
 I've nothing more to add to this...Hopefully, I can sit back to read some more and learn from it.

 Den

To thank the Author because you find the post helpful or well done.

Dear colleagues,

In support of the thought that "thumping" has mainly to do with today's program material having much higher bass levels than in the 1950s and 60s, I find that my Zenith also has severe bass distortion on certain stations here in the U.S.  In fact, the distortion is at least as bad on this radio as on any of the German 1950s radios I have.

The level of distortion varies from model to model, and probably depends on the bass response curve that each radio was designed to deliver.  As has been pointed out above, many radios have "loudness" compensation that boosts bass response significantly (especially at low volume levels) to provide a more pleasing sound than flat response would produce.  Apparently my Zenith has quite a bit of bass boost.

Another factor which may be playing a role (as pointed out above) is the high level of audio compression used today on some stations to make them sound "loud" compared to other stations on the dial.  This may be pushing the FM demodulators in some antique radios beyond their regime of good linearity.  It also sounds to me as if some stations put in some additional bass boost to maximize the "boom" that seems to be a major objective of certain segments of the American population as they drive around with their car audio systems running at high volume.

Best regards,

Tom

To thank the Author because you find the post helpful or well done.

Hello Mr. Albrecht,

 

i see you now whats going on.

I will bring first bevor i am finish two original Grundig Lab curves they show exact this.

first  Volumecontrol with Bass and treble boost versus variation "VOLUME"

and the simple bass and treble control in a Home-Radio not a Hifi set,  Volume-Control is full open. for these Curves

There is not enough variation on Bass downwards.

I am not finish yet, but i see you on my way.

 

Hans M. Knoll

 

To thank the Author because you find the post helpful or well done.

Thank you Hans....

Back to the de-emphasis.... With 50 uS (factory) instead of 75 uS for American broadcasting, I suppose the resulting sound might be a bit "bright" sounding so the listener would turn up the bass to compensate for it. Hence the thumping effect.

Jeff

To thank the Author because you find the post helpful or well done.

Gentlemen,

I respectfully believe that Herr Knoll and Herr Birkner have assured us that we are discussing different reference frames.

I personally agree with their observations; and there is also the subjective element of different personal auditory  factors depending upon the listener.  After more than 4,000 flight hours as a crewmember in the C-130 turboprop aircraft (Hercules), my hearing is not nearly as keen as the typical 67year old.

I believe Hans Knoll has more than adequately explained the FM emphasis/de-emphasis; in addition, the US tabletop radios rarely ever offered the option of a 4-level or 5-level equalization selection and control as we have seen in the Wunschklang Register on the upper level of Grundig receivers from 1956 onwards.

See the recent forum posting by Herr Thomas Günzel, in the Papers section, "Die Wunschklang - Register 1956" referenced above. The article appeared in the very prominent Funkschau literature which dates from before 1935. . . . . . . . . .

It may be of some benefit for our US members to occasionally take the time to review the German language postings in which these topics are discussed. There is much to be gained and a mutual benefit if we realize the different "cultural tastes" which shape the audio products produced for the European listener - especially as it relates to the enjoyment of Classical music.

Respectfully,
Robert

To thank the Author because you find the post helpful or well done.

To thank the Author because you find the post helpful or well done.

To thank the Author because you find the post helpful or well done.

Dear Radio Friends, Fans of European/German Radios.

We are fully aware of your worldwide appreciation for the technology and even cultural achievments made by engineers, designers, and marketing people in the field of those wonderful sounding radios of the 50's, 60's and 70's..

We must emphasize that we do not want to critisize. We want to assist. Problems often require an illumination from several different aspects. I think we had a pertinent and comprehensive discussion. All essential views have been offered.

I am planning to issue a comprehensive report "Famous sound of German Radios" in RMorg in due time. It will take some time and will hopefully include all aspects.

Ernst Erb asked me to tell you that he wants to close this thread for the time being. I will then ask him to open it again - just for giving a link to the article. Thank you for your understanding. 

********************* in German: *****************************

Verehrte Freunde deutscher/europaeischer Radiotechnik.

Wir wissen die Anerkennung zu schaetzen, die der Technik und Kultur entgegengebracht wird, welche unsere Ingenieure, Konstrukteure und Marktstrategen schufen in den beliebten Radiomodellen der 50er, 60er und 70er Jahre mit ihrem wundervollen Klang.

Wir wollen deshalb darauf hinweisen, dass wir nicht bekritteln sondern assistieren moechten.

Probleme erforden oft eine Betrachtung aus unterschiedlichen Blickwinkeln. Dies ist m.E. hier in ausreichender und sachlicher Form erfolgt. Ich werde zu gegebener Zeit zum "Klang deutscher Radios" einen ausfuehrlichen Bericht im RMorg veroeffentlichen - und hier einen Link darauf setzen.

Meine Arbeit wird eine gewisse Zeit dauern. Deshalb moechte Herr Erb diese Diskussion nun abschließen. Dafür bitte ich Euch um Verstaendnis.

To thank the Author because you find the post helpful or well done.