Sign in   |  Join   |  Help
Click here to change your Beoworld Account Details

Beomaster 8000 radio problem [solved]

rated by 0 users
Answered (Verified) This post has 1 verified answer | 29 Replies | 3 Followers

chartz
Top 25 Contributor
Burgundy, France
3,676 Posts
OFFLINE
Bronze Member
chartz posted on Sun, Mar 2 2014 2:45 PM

I would like to refresh that old thread of mine, still unsolved!

Here's my problem: I can't tune above around 92 MHz, after which frequency control is lost and the whole FM band is scanned very quickly - I can hear radios very briefly - and that's it, static, stuck at 108 MHz, but still displaying 92. I have to manually go all the way back to 87.5 to get control back. I want to precise that the counter works because if I turn the wheel I can display any frequency and it stays there. 

Any ideas please?

The video shows the problem at 97 MHz, so this has worsened with time...

Format: MOV
Duration: --:--

Jacques

Answered (Verified) Verified Answer

RaMaBo
Top 150 Contributor
near Munich
434 Posts
OFFLINE
Gold Member
Verified by chartz

Hi,

 

on the pics of the processor board you posted i only can see _one_ CD4013 on a seperate board connected with 4 wires to the main processor board The CD 4013 contains two D-Flipflops inside which are connected to each other so that every flipflop divides the signal at its D-Input by two and both together divide by 4.

 

Bad that you don't have a frequency counter, but maybe you have an oscilloscope?

Applying a manual tuning voltage to the tuner should give you a visible frequency sweep on the screen of the scope when changing the voltage from low to high and vice versa. This sweep will be shown by the highs and the lows getting narrower when 'tuning' upwards and getting broader when tuning downwards.

At PIN 3 of the CD 4013 you should get the half 'pitch' of the pulses (= frequency at PIN3 will be divided by 2) compared to PIN 1 and at PIN 13 the 'pitch' should be doubled ( = frequency at PIN 3 divided by 4).

 

I doubt that the processor itself has a fault because every other function it controls works fine as you stated before so it's 'only' counting that world go wrong. As long as the basic function is working (and it does from 87,5 - 93 MHz !) there can only be a missing or malformed clock signal at PIN 21 of the processor IC3  This will be interpreted by the program for the tuning that the tuning voltage is to low to get the wanted frequency and therefore it decided to raise the voltage even more and also fast (IC3 PIN18 going Low)  to it's limits (which is 19,5 V by design giving 108 MHZ) resulting in the fast tuning to the maximum within 2 seconds.

When tuning down getting nearer to the 'magic' frontier of 93 MHz the processor suddenly gets an frequency which matches the expected frequency and the loop is closed and stable.

 

So checking the frequency at IC3 PIN21 and it's origins towards the base of TR5 should help to isolate the fault.

 

Replacing the CD 4013 should be a first step. It can also be named HEF4013 or MC4013 all are CMOS Chips containing two seperate  flipflops in one case. Try also to replace the wires connecting the additional board to the main board or check atleast the continuity of them from board to board maybe cracked wire or solder joint.

 

Ralph-Marcus

All Replies

Dillen
Top 10 Contributor
Copenhagen / Denmark
10,388 Posts
OFFLINE
Founder
Moderator

This is the Beomaster in which we tested another pair of tuner boards, correct ?
If so, I am almost sure that this is a processor problem since the tuner basically is
controlled by the "up/down/fast" signals from the CPU and I am sure you will see the voltage on
the collector pin of 1TR1 run astray as well.
Can you provide a photo of the processor module and also one where all lettering on the two CPUs is visible ?

Martin

 

chartz
Top 25 Contributor
Burgundy, France
3,676 Posts
OFFLINE
Bronze Member

Dillen:

This is the Beomaster in which we tested another pair of tuner boards, correct ?

Yes, correct.

Dillen:

I am almost sure that this is a processor problem since the tuner basically is
controlled by the "up/down/fast" signals from the CPU and I am sure you will see the voltage on
the collector pin of 1TR1 run astray as well.

That's the case indeed. The processor module is closed at the moment, in the BM, but here are the photos of the actual PCB, unfortunately with no view on the processors themselves.

Jacques

chartz
Top 25 Contributor
Burgundy, France
3,676 Posts
OFFLINE
Bronze Member

I've found that:

http://beolover.blogspot.fr/2013/09/beomaster-frequency-counter-feedback.html

Same experience, same scope readings basically if memory serves. I replaced the prescaler two years ago, to no avail. So my problem is different, and perhaps processor-related.

Jacques

chartz
Top 25 Contributor
Burgundy, France
3,676 Posts
OFFLINE
Bronze Member

Hi,

I'm still looking for a spare microprocessor module donation. 

I now cannot tune on anything past 91 MHz, so the situation is worsening. At the beginning of the problem, it was 97, then 102 for a few months, yesterday, 92 and now 91. 

All other functions of the Beomaster work as they should.

This is very puzzling.

Jacques

sonavor
Top 50 Contributor
Texas, United States
2,972 Posts
OFFLINE
Gold Member

That is pretty odd. It seems like something is interfering with the signal between the tuner board and the microcomputer board though. If the microcomputer  (IC4) was bad it I think it would just stop working altogether, not slowly deteriorate. That is just a guess of course so I could be completely wrong. I know when I refurbished my BM8000 and broke a lead on the L2 component of the microcomputer board that killed any chance of tuning an FM station. Once I fixed that inductor/transformer, everything started working again. That makes me think something involved in that path might be the culprit. I'm afraid I don't have a spare microcomputer board but be patient, I'm sure a donor part will eventually be found.  If you were in the US I know of a guy selling a used BM 8000 in Denver.  It would be way too heavy to ship to France I think.

chartz
Top 25 Contributor
Burgundy, France
3,676 Posts
OFFLINE
Bronze Member

Thanks for your thoughts John. 

In fact when I monitor the processor's pulses at pins 19 and 20 I can see it go astray at the same frequency.

There must be something wrong in the loop, not the tuner - this was checked by swapping the whole tuner section  - but somewhere on the processor module. 

 

 

Jacques

RaMaBo
Top 150 Contributor
near Munich
434 Posts
OFFLINE
Gold Member
RaMaBo replied on Tue, Mar 4 2014 11:19 AM

Hi Jaques,

 

do you have a frequency counter or can get one for a short time?

To check the divider part on the microprocessor board you can check what frequency range the processor gets offered when manual tuning. You should get a readout of about 384 KHz for 87,5 MHz and about 464 KHz for tuning to 108 MHz. This verifies the correct function of the prescaler (part of the loop). Check the voltage at pin 1 of the prescaler chip µ264 you should read about 3,3 V (always  at 87,5 MHZ as well as at 108,0 MHz and in between).

The tuner module as the second part of the loop is also already checked as you wrote. so there are two parts left: The processor itself and the tuning voltage converter

All this checking should be done with 'AUTO TUNE' switched off so that you get as much control as possible over the loop.

 

Ralph-Marcus

chartz
Top 25 Contributor
Burgundy, France
3,676 Posts
OFFLINE
Bronze Member

Thanks for that, Ralph-Marcus.

No I don't have a frequency counter alas, nor can I get hold of one, and I must admit that it's a problem.

I can confirm that I do get a firm and constant 3.3 V at pin 1 of the prescaler.

The voltage converter has been checked too.

The tuning voltages are normal at 87.5 and 108. Only it jumps from 93 to 108 in 2 seconds.

Also, I remember that injecting tuning voltage from external variable PSU makes the tuner work normally, with no drift control of course.

By the way, why on earth is the flip-flop divided into two 4013 chips?

Jacques

RaMaBo
Top 150 Contributor
near Munich
434 Posts
OFFLINE
Gold Member
Verified by chartz

Hi,

 

on the pics of the processor board you posted i only can see _one_ CD4013 on a seperate board connected with 4 wires to the main processor board The CD 4013 contains two D-Flipflops inside which are connected to each other so that every flipflop divides the signal at its D-Input by two and both together divide by 4.

 

Bad that you don't have a frequency counter, but maybe you have an oscilloscope?

Applying a manual tuning voltage to the tuner should give you a visible frequency sweep on the screen of the scope when changing the voltage from low to high and vice versa. This sweep will be shown by the highs and the lows getting narrower when 'tuning' upwards and getting broader when tuning downwards.

At PIN 3 of the CD 4013 you should get the half 'pitch' of the pulses (= frequency at PIN3 will be divided by 2) compared to PIN 1 and at PIN 13 the 'pitch' should be doubled ( = frequency at PIN 3 divided by 4).

 

I doubt that the processor itself has a fault because every other function it controls works fine as you stated before so it's 'only' counting that world go wrong. As long as the basic function is working (and it does from 87,5 - 93 MHz !) there can only be a missing or malformed clock signal at PIN 21 of the processor IC3  This will be interpreted by the program for the tuning that the tuning voltage is to low to get the wanted frequency and therefore it decided to raise the voltage even more and also fast (IC3 PIN18 going Low)  to it's limits (which is 19,5 V by design giving 108 MHZ) resulting in the fast tuning to the maximum within 2 seconds.

When tuning down getting nearer to the 'magic' frontier of 93 MHz the processor suddenly gets an frequency which matches the expected frequency and the loop is closed and stable.

 

So checking the frequency at IC3 PIN21 and it's origins towards the base of TR5 should help to isolate the fault.

 

Replacing the CD 4013 should be a first step. It can also be named HEF4013 or MC4013 all are CMOS Chips containing two seperate  flipflops in one case. Try also to replace the wires connecting the additional board to the main board or check atleast the continuity of them from board to board maybe cracked wire or solder joint.

 

Ralph-Marcus

chartz
Top 25 Contributor
Burgundy, France
3,676 Posts
OFFLINE
Bronze Member
chartz replied on Wed, Mar 5 2014 10:42 AM

I've got the CD4013s so I'll do that first.

I do have a scope!

Thank you so much for the compehensive explanation! 

Jacques

sonavor
Top 50 Contributor
Texas, United States
2,972 Posts
OFFLINE
Gold Member

Jacques,
Can you post some pictures of your test connections and your measurements as you go?  It will be a nice reference for future Beomaster 8000 work. :)

Søren Mexico
Top 10 Contributor
Mexico City
6,097 Posts
OFFLINE
Gold Member

sonavor:

Jacques,
Can you post some pictures of your test connections and your measurements as you go?  It will be a nice reference for future Beomaster 8000 work. :)

Yes please, and Ralph-Marcus, thank you for your clear explanations, even I understand it.Smile

 

Collecting Vintage B&O is not a hobby, its a lifestyle.

chartz
Top 25 Contributor
Burgundy, France
3,676 Posts
OFFLINE
Bronze Member

First of all, I'd like to say that I'm sorry.

Sorry, because I did not have to measure anything more. So, nothing to show.

I know, this is frustrating.

I'm sorry, again.

THE THING WORKS! THE PROCESSOR IS GOOD!

A replacement of the two flip-flop CD4013 has cured the Beomaster completely. I don't know which one was at fault, but there.

How did Bruce Willis put it already? Big Smile

Anyway, thanks to all, Martin, Søren, John, always supportive of our efforts, real friends, and particularly to Marcus-Ralph who convinced me that the μP was fine.

Wow, this is my two thousandth post Stick out tongue

Jacques

sonavor
Top 50 Contributor
Texas, United States
2,972 Posts
OFFLINE
Gold Member

Great news Jacques. I'm glad to hear your Beomaster 8000 is back in working order. I use the one I refurbished last year almost everyday.  I have one, maybe two, more that are waiting for me to repair so I am always interested in the problems and solutions other BM8000 owners run into.

Page 1 of 2 (30 items) 1 2 Next > | RSS
Beoworld Security Certificate

SSL