UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Other Discussions > Homebrew Equipment

Notices

Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members.

Closed Thread
 
Thread Tools
Old 1st Aug 2021, 6:41 am   #41
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
Default Re: DIY FM tuner with 6CW4 nuvistor and ECC85

Quadrature signal path techniques can, in theory, convert and demodulate anything. There are plenty of articles and app notes on what they can do.

Information on their limitations is less common, but crucially important.

They are related to 'phasing method' SSB receivers but are handling wider bandwidths. So the unwanted sideband is full of unwanted signals and needs to be rejected. Rejection is done by cancellation not filtering. Cancellation is a subtractive process. If the I and Q paths are split in the analogue domain you suddenly find yourself needing superb matching between the two channels, and precision phase shifters.

If your channels are only matched to 1% in gain, you only get 40dB suppression of the unwanted stuff. Similarly if your precision in phase terms is one degree, then that's also about 40dB suppression. This is going to take better than 1% resistors and amazing matching of RF amplifiers, mixers etc.

Software calibration can correct this and give you 10 to 20dB improvement, but your receiver is still at a big disadvantage to an classic analogue filtered version.

The escape route is to digitise the incoming signal with only one ADC, and then to play I & Q games entirely in the digital domain. You can have superb precision here and get superb cancellation. I & Q techniques are popular in DSP because they economise on the amount of processor power needed to do the necessary tasks.

There are receivers coming out where the whole 0-30MHz band is digitised in one fell swoop. They are definitely catching on. Their Achilles heel is that the ADC has to be scaled to handle the total peak voltage of all signals combined. Doing this pushes small signals towards the noise floor. They need decent preselector filters to try to reduce the total voltage hitting the ADC so they can be better scaled to handle individual signals.

They're not bad nowadays and the Icom IC-7300 is proving very popular. Their top end transceivers use analogue conversions and narrow filters ahead of their ADC, as well as preselectors. Look at the IC-7700 and IC-7800 families.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 1st Aug 2021, 11:18 am   #42
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: DIY FM tuner with 6CW4 nuvistor and ECC85

Thanks for the explanation.

The IC-7300 uses 16 bit high-speed direct sampling ADC that can capture the RF signal.

Correction: Ziegu G90 uses lower cost 24-bit mixer type ADC, it is not direct sampling.

The greater the bits, the more the dynamic range(the less quantization noise from the ADC);

16 bit ADC has 95db dynamic range
24 bit ADC has 144db dynamic range

Correction: The Marantz 20B and 10B use two-stage grounded grid RF amp with ECC88 and they are not in cascode.

I find this very intuitive, visual explanation of I and Q:

http://whiteboard.ping.se/SDR/IQ
Attached Thumbnails
Click image for larger version

Name:	IC-7300.jpg
Views:	85
Size:	38.5 KB
ID:	238609  
regenfreak is offline  
Old 1st Aug 2021, 2:11 pm   #43
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
Default Re: DIY FM tuner with 6CW4 nuvistor and ECC85

My usual explanation is that a man cycles past on an ordinary bicycle and all you get is a graph of pedal height versus time. You can tell how fast he is pedalling but not whether he is driving his bicycle onwards or whether he is back-pedalling. This is because both pedals are in the same plane and the ambiguity is unresolved.

Now, someone from the Isle of man cycles past (they all have three legs, look at their flag) on his '3-phase' bicycle with three pedals ar 120 degree intervals. Again you get plots of the heights of each pedal, and so long as you know which pedal is which plot you now CAN tell between forwards and backwards pedalling.

Finally, a clown cycles past, erratically, on a bicycle with only two pedals, but fitted at 90 degrees to each other on their shaft. You get two plots, and again you can tell the direction because the pedals are out of the same plane.

With receivers converting to IFs centred on zero (or close tozero) then stuff above the tuned frequency becomes a positive frequency in the IF, stuff below becomes a negative frequency. You need the perspective view of two phases of channels to discriminate between them. On a single channel, negative and positive frequencies are indistinguishable.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 1st Aug 2021, 5:31 pm   #44
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: DIY FM tuner with 6CW4 nuvistor and ECC85

Quote:
Originally Posted by Radio Wrangler View Post
My usual explanation is that a man cycles past on an ordinary bicycle and all you get is a graph of pedal height versus time. You can tell how fast he is pedalling but not whether he is driving his bicycle onwards or whether he is back-pedalling. This is because both pedals are in the same plane and the ambiguity is unresolved.

Now, someone from the Isle of man cycles past (they all have three legs, look at their flag) on his '3-phase' bicycle with three pedals ar 120 degree intervals. Again you get plots of the heights of each pedal, and so long as you know which pedal is which plot you now CAN tell between forwards and backwards pedalling.

Finally, a clown cycles past, erratically, on a bicycle with only two pedals, but fitted at 90 degrees to each other on their shaft. You get two plots, and again you can tell the direction because the pedals are out of the same plane.

With receivers converting to IFs centred on zero (or close tozero) then stuff above the tuned frequency becomes a positive frequency in the IF, stuff below becomes a negative frequency. You need the perspective view of two phases of channels to discriminate between them. On a single channel, negative and positive frequencies are indistinguishable.

David
This is a brilliant and funny analogy! I and Q are clown plots!

Once a upon a time, I knew a bloke and a girl who are both Manx. That was 30 years ago and all i can remember they always complained about the weather and there is nothing to do in the island. .
W2AEW produced a series of excellent videos about Q and I that he demonstrated the ideas using a scope:

https://www.youtube.com/watch?v=h_7d-m1ehoY


Mmmm i have spent all afternoon scratching my head, looking at the datasheets and examples applications using 6BA6 and 6AU6. One of the trickest part is to get the right plate and screen voltages for the IF amps and limiters. Just to have to calculate the optimal grid/cathode bias and dropper resistors for the plate...too low plate voltage for the limiter resulting in clipping for weak AM noises but too low gain...I can see very flat response curve on the frequency sweep.
regenfreak is offline  
Old 1st Aug 2021, 8:45 pm   #45
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: DIY FM tuner with 6CW4 nuvistor and ECC85

This explains quite well the concepts of I and Q, Nyquist Theorem, DDC, DUC, decimation and interpretation in SDR:

https://www.youtube.com/watch?v=BK9QkHxeYQI

It is an old video but I find easy to understand.
regenfreak is offline  
Old 6th Aug 2021, 4:50 pm   #46
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: DIY FM tuner with 6CW4 nuvistor and ECC85

At last I have got the 6CW4 tuner working.

The culprit was one of the two "NOS" ECC85 is bad. I have replaced the bad ECC85 by 6BS8 for the oscillator, it comes alive suddenly. The tuner is now working very well with good sensitivity and selectivity. I wire it to the IF input of an existing stereo tuner. It retains very good stereo fidelity. The other modification is that I have drilled bigger holes near all three silver-plated coils. My theory is that the proximity of the coils near the copper ground plane would dampen the oscillation and lower Q of the coils. Copper is far more conductive than aluminium. At VHF frequency, the induced eddy current on the copper plate opposes the magnetic field from the coil reducing its inductance.

Anyway I am happy with it.
Attached Thumbnails
Click image for larger version

Name:	A3.jpg
Views:	138
Size:	77.5 KB
ID:	238906   Click image for larger version

Name:	A1.jpg
Views:	123
Size:	68.2 KB
ID:	238907   Click image for larger version

Name:	A2.jpg
Views:	131
Size:	78.9 KB
ID:	238908  
regenfreak is offline  
Old 22nd Aug 2021, 7:24 am   #47
djsbriscoe
Pentode
 
Join Date: May 2015
Location: Hemel Hempstead, Hertfordshire, UK.
Posts: 132
Default Re: DIY FM tuner with 6CW4 nuvistor and ECC85

Corrected link to Radio electronics 1953-04

https://worldradiohistory.com/Archiv...cs-1953-04.pdf
djsbriscoe is offline  
Old 9th Oct 2021, 1:48 pm   #48
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: DIY FM tuner with 6CW4 nuvistor and ECC85

I just give an update as I haven't been around here much lately.

After some fine alignments, the 6CW4 triple-gang tuner and two-gang tuner (mentioned in previous posts)work very well, on par with my Leak 3 Stereo valve tuner in terms of sensitivity, selectivity and stability.

The attached photos are my third FM tuner with IF and discriminator module. It is stereo with built-in solid state decoder and magic eye tuning indicator. The tuner head uses E88CC in cascode RF amp, 6CB6 mixer and 6C11 oscillator. The oscillator is working, after addressing some self-oscillation and instability issue. I am running into problems with diagnostic techniques of the RF amp and mixer. Anyway, the alignments and debugging start to get difficult as it is not easy to probe around without loading the tuned circuits. the IF stereo module is equally tricky with stagger tuning, lots of messing around with plate, grid, screen voltages and sweep alignments...this is the problem with homebrew ; there is no manufacturer instructions for repair and alignment. VHF circuits are very unforgiving.
Attached Thumbnails
Click image for larger version

Name:	FM1.jpg
Views:	91
Size:	111.8 KB
ID:	242988   Click image for larger version

Name:	FM2.jpg
Views:	84
Size:	116.8 KB
ID:	242989   Click image for larger version

Name:	FM3.jpg
Views:	80
Size:	74.2 KB
ID:	242990   Click image for larger version

Name:	FM4.jpg
Views:	86
Size:	99.4 KB
ID:	242991  
regenfreak is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 4:49 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.