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 > General Vintage Technology > Components and Circuits

Notices

Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets.

Closed Thread
 
Thread Tools
Old 26th Jun 2021, 10:40 pm   #61
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: Relaxation oscillation in LC resonance/ Toko testers

Quote:
The double tuned transformer innards are unknown to me and I'm not sure how you have arranged the connections but I did try sweeping through a parallel resonator on a simulator with 50 ohm ports. In order to get the same response and selectivity as your plot I ended up with 11.2uH and 12pF.
This only needs an external 8pF to tune it down to 10.7MHz.

The plot below is just a single resonator being swept so probably nothing like what is inside your IF can.]
The test transformer is wired like in the attached drawing. Incidentally your single tuned model almost has the same inductance and capacitance as the primary LC! ( L=11.676, C = 10pF). Too much of a coincidence? The FM transformer is japanese, NOS in box with manual, its 3db bandwidth is 100kHz so loaded QL = 107
Attached Thumbnails
Click image for larger version

Name:	trio fm.jpg
Views:	50
Size:	37.9 KB
ID:	236522  
regenfreak is offline  
Old 26th Jun 2021, 11:12 pm   #62
Cruisin Marine
Heptode
 
Join Date: Jul 2018
Location: Worthing, West Sussex, UK.
Posts: 983
Default Re: Relaxation oscillation in LC resonance/ Toko testers

I wonder if the vna approach may not be the answer, perhaps a noise genny (or sweeper) and Spectrum analyser in linear mode could be better?
I swear my old HP 141T/855X mainframe analyser unit, I could detect a gnat passing wind in central Africa I got so used to using that setup.

As loosely coupled in and loosely coupled out as possible of course- letting the resonant cct. absorb just enough to show an effect, the sweep gen could be turned up and the SA turned down to accommodate (within reason and applied attenuators being used of course). The cct loading may be the big factor, but certainly not the only one under applied supply power conditions. Experimentation needed of course to get the desired result (if poss)

This is a form of experimentation I have not really attempted to do in seriousness, as usually one can guess what the cct. resonance is designed to do when powered up especially. We usually know the common IF frequencies used, and if not, a rough measurement will usually lead to the eventual answer we seek.

There may always be far too much difference between non powered and powered up equipment to give a "spot on" answer.
Peaking powered up ccts. in the real world will be the only definite answer anyway.

But, as the company motto always said "Close enough is good enough" well, normally in my case anyway.

Last edited by Cruisin Marine; 26th Jun 2021 at 11:37 pm.
Cruisin Marine is offline  
Old 26th Jun 2021, 11:45 pm   #63
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: Relaxation oscillation in LC resonance/ Toko testers

Quote:
I wonder if the vna approach may not be the answer, perhaps a noise genny (or sweeper) and Spectrum analyser could be better?
As loosely coupled in and loosely coupled out as possible of course, the sweep gen could be turned up and the SA turned down to accomodate (within reason and applied attenuators being used of course). The cct loading may be the big factor, but certainly not the only one under applied supply power conditions.
In fact I tried to do sweep using two different white noise generators with the TinySA.spectrum analyzer and compare the results of two NanoVNAs. White noise generators have poor, non-flat at low frequencies, you i have to substract the noise datum input. In the attached photo, it is sweep of a 40m butterwoth 2nd order bandpass filter; yellow is after correction of the white noise non-uniform distribution, green is the white noise input. If you compare the response with my previous post of the same filter using the NanoVNA F, you can see the side fall-off are wrong. I repeated the test with another more expensive white noise generator , the problem remains. Emm I cannot afford a proper spectrum analyzer new or old.

Quote:
There may always be far too much difference between non powered and powered up equipment to give a "spot on" answer.
Peaking powered up ccts. in the real world will be the only definite answer anyway.

But, as the company motto always said "Close enough is good enough" well, normally in my case anyway.


Indeed maybe we are going down a rabbit hole. But there are always hidden secrets and knowledge that are not written in books and web sites. When i make a valve Fm radio from scratch, i always try to figure out why the manufacturers designed components in certain ways and not other...there were tricks that are intriguing. My natural tendency is asking lots of questions and driving myself nut. The other approach is to forget about it and move on, then i may take up another hobby instead.
Attached Thumbnails
Click image for larger version

Name:	white noise.jpg
Views:	46
Size:	79.8 KB
ID:	236528  
regenfreak is offline  
Old 26th Jun 2021, 11:51 pm   #64
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Relaxation oscillation in LC resonance/ Toko testers

For sure it's always worth experimenting with other methods. If the double tuned transformer (or the radio) comes with a manual I'd expect it to show or state the source and load impedances required to deliver the 100kHz BW at 10.7MHz.

Looking at the circuit values this will probably mean the source and load terminations need to be many tens of k ohms in parallel with about 10pF.

The coupling between the tuned circuits is also probably going to be very light so I'd expect a basic through measurement with a 50R VNA to show very little response at all. Having seen that this must be a lightly coupled double tuned circuit I can see why you want to match up to a higher impedance to test it.

Does the manual state anywhere that there needs to be additional capacitance provided by the valve and maybe even another padding capacitor to tune the circuit down to 10.7MHz? Without this, the component values don't look right for 10.7MHz. It does look like it will be closer to 13-14MHz.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 26th Jun 2021, 11:51 pm   #65
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: Relaxation oscillation in LC resonance/ Toko testers

Quote:
The plot below is just a single resonator being swept so probably nothing like what is inside your IF can.
Attached Thumbnails
What is the difference between a series-thru and shunt-thru measurement of a single LC resonance? My previous post for the Toko FM if transformer was done in series-thru measurement like your simulator. I thought it was wrong way but I am not sure now.
regenfreak is offline  
Old 26th Jun 2021, 11:58 pm   #66
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: Relaxation oscillation in LC resonance/ Toko testers

Quote:
Does the manual state anywhere that there needs to be additional capacitance provided by the valve and maybe even another padding capacitor to tune the circuit down to 10.7MHz? Without this, the component values don't look right for 10.7MHz. It does look like it will be closer to 13-14MHz.
Usually the primary is connected to the plate of a pentode IF ampifier, so i assume that output capacitance of the wirings and the plate make up about 9pF to resonate at 10.7MHz. The secondary is wired to the grid of the next IF amp valve. So the input capacitance (including Miller capacitance) is about 13pF to peak at 10.7MHz. This is very rough guess and i can be wrong!

Quote:
If the double tuned transformer (or the radio) comes with a manual I'd expect it to show or state the source and load impedances required to deliver the 100kHz BW at 10.7MHz.
I used google to translate the japanese a while ago. I could not find it info on input and output impedance. I have been digging around textbooks, trying to calculate it but i am not sure how. I forgot the equations involved.

Last edited by regenfreak; 27th Jun 2021 at 12:06 am.
regenfreak is offline  
Old 27th Jun 2021, 12:07 am   #67
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Relaxation oscillation in LC resonance/ Toko testers

I wonder if you had your VNA croc lead setup like the image below for your S21 measurement? This does seem to give the same response that you were seeing? Can you check your connections to see if this is what you had?

If so, does the response change to look like the second image if you sweep the other resonator as per the revised connections?
Attached Thumbnails
Click image for larger version

Name:	FM_IF_notch.jpg
Views:	40
Size:	55.4 KB
ID:	236529   Click image for larger version

Name:	FM_IF_notch2.jpg
Views:	38
Size:	55.7 KB
ID:	236530  
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 27th Jun 2021, 12:10 am   #68
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: Relaxation oscillation in LC resonance/ Toko testers

Quote:
I wonder if you had your VNA croc lead setup like the image below for your S21 measurement? This does seem to give the same response that you were seeing? Can you check your connections to see if this is what you had?

If so, does the response change to look like the second image if you sweep the other resonator as per the revised connections?
oh let me try again to double check
regenfreak is offline  
Old 27th Jun 2021, 12:21 am   #69
Cruisin Marine
Heptode
 
Join Date: Jul 2018
Location: Worthing, West Sussex, UK.
Posts: 983
Default Re: Relaxation oscillation in LC resonance/ Toko testers

Indeed maybe we are going down a rabbit hole. But there are always hidden secrets and knowledge that are not written in books and web sites. When i make a valve Fm radio from scratch, i always try to figure out why the manufacturers designed components in certain ways and not other...there were tricks that are intriguing. My natural tendency is asking lots of questions and driving myself nut. The other approach is to forget about it and move on, then i may take up another hobby instead.


We all do the same! To a lesser or greater degree.
It is not a hobby, it is either an annoyance we are inflicted with, or a pure gift we are blessed with!
I prefer to think it is the latter.
It is wonderful to question EVERYTHING- that is what Mind Magic truly is!
Cruisin Marine is offline  
Old 27th Jun 2021, 12:21 am   #70
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Relaxation oscillation in LC resonance/ Toko testers

Quote:
Usually the primary is connected to the plate of a pentode IF ampifier, so i assume that output capacitance of the wirings and the plate make up about 9pF to resonate at 10.7MHz. The secondary is wired to the grid of the next IF amp valve. So the input capacitance (including Miller capacitance) is about 13pF to peak at 10.7MHz. This is very rough guess and i can be wrong!
Thanks, that's useful info! I stuck these values into Genesys and then tweaked the K factor and source and termination impedances and managed to get a 100kHz bandwidth centred on 10.7MHz. Is the circuit below similar to what you say above? The ? next to the 9pF cap ( i.e. ?9pF) just means the circuit value is highlighted for real time trimming. I've added 9pF and 13pF to the transformer as in the circuit below.

It could just be a lucky combo of parameters though. I don't know much about valve IF filters so that is my first go at trying to get one simulated.
Attached Thumbnails
Click image for larger version

Name:	FM Filter_10700.jpg
Views:	49
Size:	56.9 KB
ID:	236532  
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 27th Jun 2021, 12:31 am   #71
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: Relaxation oscillation in LC resonance/ Toko testers

I have wired exactly like yours:

primary dip at 14.4MHz -54.4db

seconday dip at 13.4MHz, -51 db

So the secondary is closer to yours.

I was wiring like what i attached in the drawing; Shunt-primary and shunt-secondary with common ground
regenfreak is offline  
Old 27th Jun 2021, 12:35 am   #72
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: Relaxation oscillation in LC resonance/ Toko testers

Quote:
hanks, that's useful info! I stuck these values into Genesys and then tweaked the K factor and source and termination impedances and managed to get a 100kHz bandwidth centred on 10.7MHz. Is the circuit below similar to what you say above? The ? next to the 9pF cap ( i.e. ?9pF) just means the circuit value is highlighted for real time trimming. I've added 9pF and 13pF to the transformer as in the circuit below.

It could just be a lucky combo of parameters though. I don't know much about valve IF filters so that is my first go at trying to get one simulated.
Yes that is it. Wow wow very nice!!!! Thanks for that. Is Genesys software available free?

Last edited by regenfreak; 27th Jun 2021 at 12:41 am.
regenfreak is offline  
Old 27th Jun 2021, 12:43 am   #73
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Relaxation oscillation in LC resonance/ Toko testers

If you were to add 9pF across the input side then does this move the null fairly close to 10.7MHz in each case? See the plot below.

This indicates the resonator on the left is now correctly tuned to 10.7MHz when the external 9pF cap is added.
Attached Thumbnails
Click image for larger version

Name:	FM Filter_10700_padded.jpg
Views:	38
Size:	51.9 KB
ID:	236537  
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 27th Jun 2021, 12:45 am   #74
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Relaxation oscillation in LC resonance/ Toko testers

Quote:
Originally Posted by regenfreak View Post

Yes that is it. Wow wow very nice!!!! Thanks for that. Is Genesys software available free?
Sadly no, but you should be able to do this for free in LTSpice as I think it has a decent transformer model included. I'm off to bed but I'll try and set up the same circuit in LTSpice tomorrow.

However, someone else with more experience of LTSpice might be able to set up the simulation better than I can. I've only used LTSpice a few times.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 27th Jun 2021, 12:47 am   #75
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: Relaxation oscillation in LC resonance/ Toko testers

Yes the trim caps can be regarded as the dynamic inter-electrodes capacitance ( and Miller capacitance) of the grid and the plate. It is a function of the gain and operating conditions of the IF amplifier valve.
regenfreak is offline  
Old 27th Jun 2021, 12:48 am   #76
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: Relaxation oscillation in LC resonance/ Toko testers

Thanks for all this. Goodnight I am off to bed too
regenfreak is offline  
Old 27th Jun 2021, 12:37 pm   #77
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: Relaxation oscillation in LC resonance/ Toko testers

The Japanese FM IF comes in a box set ( two IF transformers and a ratio detector transformer). The user manual provides the sweep response curves and recommended IF valve 6BA6 (see photos) which is variable-mu pentode. Usually it uses a plate voltage of about 75V.

From the 6BA6 datasheet:

Anode impedance ra = 250k
input impedance 4.5K
Inter-electrodes input capacitance = 5.5pF
Inter-electrodes output capacitance = 5pF
anode resistor 10k (for 75V plate voltage)
qm = 4.3mA/V

Cin=Cgk+Cga(1+A) and A = Vout/Vin

A= -gm X [ra/(ra +RL)]
Attached Thumbnails
Click image for larger version

Name:	trio.jpg
Views:	44
Size:	59.5 KB
ID:	236558   Click image for larger version

Name:	trio2.jpg
Views:	38
Size:	61.6 KB
ID:	236559  
regenfreak is offline  
Old 27th Jun 2021, 3:06 pm   #78
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Relaxation oscillation in LC resonance/ Toko testers

Thanks. I think there is enough info now to have a go at making a similar filter on a bare piece of PCB.

The VNA plot below is a first attempt but I made this with inductors that probably only have a Qu of 200 at 10.7MHz.

However, I set the source impedance to 220k ohm and the load impedance was 68k ohm. I used the same inductances as the original and adjusted the coupling to get a 100kHz bandwidth. I padded the capacitances with the extra 9pF and 13pF as in the previous posts.

I've used series resistive (very lossy) matching to set up the port impedances to make the plot below so that is why it shows about -65dB on the response. However, the passband looks about right.

My nanovna struggles to make the same plot. I think I'll have to upgrade the nanovna firmware to allow a lower resolution bandwidth and a higher source power. I'll have a go at this in a bit and post up a nanovna plot of the filter.

It might be best to try and repeat what I'm doing with the real IF filter as it only took a few minutes to set up the test below and that included soldering the parts together on a test board. There aren't many parts involved.

The second plot is the Genesys model of the filter and it seems to agree quite well with the filter I just built using the same components and transformer coupling factor.
Attached Thumbnails
Click image for larger version

Name:	10700bpf.gif
Views:	40
Size:	17.4 KB
ID:	236570   Click image for larger version

Name:	10700bpf_model.jpg
Views:	41
Size:	61.5 KB
ID:	236571  
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 27th Jun 2021 at 3:18 pm.
G0HZU_JMR is offline  
Old 27th Jun 2021, 3:32 pm   #79
regenfreak
Heptode
 
Join Date: Nov 2018
Location: London SW16, UK.
Posts: 655
Default Re: Relaxation oscillation in LC resonance/ Toko testers

Thanks. I have tried to replicate your early reflection experiment on a Toko10k series FM black 10.7MHz IF. I have added reflected test sockets on the board.

The S11 test does not work, the Img S11 is a flat line; no sign change throughout.

Then i have done S21, please attached. The Toko FM 10.7MHz has L = 4.1mircro H. I dont know C, may be it is 50pF. Data sheet does not state it. Q may be 50.
Attached Thumbnails
Click image for larger version

Name:	toko s21.jpg
Views:	29
Size:	36.4 KB
ID:	236572   Click image for larger version

Name:	Snap 2021-06-27 at 15.15.27.jpg
Views:	27
Size:	21.0 KB
ID:	236573  
regenfreak is offline  
Old 27th Jun 2021, 3:34 pm   #80
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Relaxation oscillation in LC resonance/ Toko testers

What are the two RF chokes for that are trying to hide in the green Trio box?
Could these be external matching components? Could you take a nice image of the Trio datasheet on its own and then pdf or zip it up and post it onto the forum? Otherwise the image quality will be too poor as the forum limits the size and quality of jpgs. It would be nice to see the plots on the datasheet.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 10:29 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.