ALL VALVE RECEIVER inductor former sizes

-Cheers Ivan, same boat with valves.. Good fun going full circle though.

Hi Ed,

my practical experience with tubes is zero, so I have just a little idea. Maybe grid current starts to flow at a certain point, that changes the amplification factor. In that case you can either reduce the voltage swing on the grid (you did that), or increase the resistor in the cathode - this increases the negative DC voltage of the grid against the cathode and stops grid current from flowing. Every type of tube requires different negative DC grid voltage.

VBR from Ivan

Hi,

All sorted now. In the end after lots of research the oscillator side had too much gain so I tweaked a few values but the real fix was to make the coil tap at 2 turns instead of 4. That fixed the oscillator but the follower triode was still a bit distorted; after decoupling it a little and reducing the drive with an R+C combination the output is clean as a whistle and at a good voltage (providing 1.5v p-p into the mixed is good?). Next time I will use the ECC81 as suggested!

I've also now completed the BFO using an upcycled VHF-41 radio IFT and the drift is in the very low 10's of Hz over time - very pleased with that. Mixer now.

Cheers for the tips,

Regards,

ZL1TET ED.

Hi,

Login working now, thanks! Returning to my local oscillator using the ECC85 distortion issue, I've taken a few measurements. 

All measurements are with a DC coupled probe.

V5b grid  12v p-p  (+/- 6V either side of 0v)       // Clean - top of Tank circuit
V5b cathode 2.2v p-p ( 0v point at 6.24v DC)   // the kink point is 100 mv above 0V
V5b anode solid 150v DC 
V5a cathode [junction of 1M+1n] 9v p-p            // (slightly distorted) This is the input to the grid of the mixer/front-end, what should this value be?

The signal on the hot-side of the tank feeding the grid of V5b is actually pretty clean, no kink whatsoever - the kink is only noticeable on the cathode (v5b). V5b's grid is directly coupled to V5a's grid yet the output is distorting a little. 
Are these two issues connected? Distortion (kink) on the cathode of V5b and the distortion of V5a's output OR there's no relationship between the kink and the output and the distortion is simply caused by overdriving V5b's grid with 12v p-p?

I'd like to understand the reasons behind these problems for the sake of learning; perhaps the amount of distortion I'm seeing in these two areas is normal or acceptable, I'm not sure.

Thanks.

ZL1TET (Ed)

zl1tet Ed wrote:P.S This is what I get when I try to post with zl1tet "Sorry, but this username has already been taken."

Hi Ed,
you post all your texts as a Guest. You are not logged in when sendig it. Yes, "zl1tet" is blocked by a registered user (YOU!), so no guest is allowed to use the same username.
Please log in as a regular member using the temporary password and change it ASAP. According to the statistics, you have never been logged in up to now.

VBR from ivan

Hi,

I've made some progress and have now got a stable linear(ish) across band(ish) output on to what will be the grid input of V4 (front end/mixer). A few more tank circuit tap changes, a dodgy scope probe didn't help.. 
My last issue I'd like to tidy up is the quality of the output sine; it has slight flattening on the negative peak. I suspect this is due to the 'kink' I have still experience as mentioned in my last post (but now inverted and a little smoothed).
My output voltage is 11v p-p, looking at m0xpd.blogspot scope picture of his build of the receiver, it appears he's getting 1 v p-p at the same point. What do you think could be causing the 'kink' in the V5b half of the valve and is it over-driving the follower's output?

Thanks,

(ZL1TET) Ed.
.
P.S This is what I get when I try to post with zl1tet "Sorry, but this username has already been taken."

Hi Harry,

Thank you for your reply. Regarding " We could have updated the old callsign password or changed to a new known password" I'm happy to do either, whatever is easier - maybe delete the old account?

I'm still working on this receiver and am experiencing a few niggles.. I've transferred the project onto an old valve radio chassis in the hope of building the entire receiver. Before I describe my issues, I'm been an Electronic Engineer since the 80's but my valve experience only goes as far as a few audio amplifiers - so pretty limited. I now regret scoffing at the 'valve rubbish' the older Engineers were working on back then; missed out on a valuable chunk of experience I think!   

Back to my local oscillator. I've tried a number of tank circuits using the advice suggested below and have got the overall frequency range about right (4-11 MHz) albeit my tuning cap I think has too much range (off an old Bush valve radio). My main issue is connecting the grid of the actual oscillator (V5b) to the follower's grid (V5A); it has a major effect on the oscillators linearity and adds distortion. I'm using an ECC85 instead of a ECC81, and I have compared the two data sheets and there's not really an enormous difference between the two valves; both have similar S, Mu, voltage ratings etc. I've tried with and without the internal screen (pin 9), no change. When I connect the two grids together it has the feeling like there's an increase of load or capacitance or something. I had to switch back to a bigger tank circuit as a small diameter tank died when connecting the other grid.

My build of the local is as simple as it can be - just the local; I used zeners as suggested and also mentioned in m0xpd.blogspot, 250v on V5a and 150v on V5b (each decoupled with a 100nf on the anode pin). I have also played with the 2K2+10n values coming off the tank's tap into the cathode.  

One other observation using a 20pf capacitor attached to my scope probe onto either side of the 2K2+10n (tank center-tap components), is a kink on the right hand side of the wave form. It's a quarter down from the top peak. It doesn't appear to alter with frequency. Another observation is there's a very small section of tuning toward the lower end of the band where the circuit 'takes-off' after which it quickly comes right. Perhaps these two idiosyncrasies may give you some points respecting my problem?

Appreciate your thoughts,

Cheers, Ed.

Ivan wrote:Hi Edward,

your registration is "zl1tet". You can log in and use full features of this messageboard. Your recent post was done as a guest, though. Maybe the callsign is too short for a guest's username? Guests have restricted rights.

There is no other registered username starting with "zl1". I checked.

A ferrite slug makes a coil easily tunable. Air coils have better Q, but their tuning range is zero or quite little, depending on their design.

Very best regards from Ivan OK1SIP

Hi Ivan, Sorry, I didn't read all of this. I just spent a couple of minutes repeating that which you already did!!
"Too much haste, not enough speed" as they say. I was a bit quick to react.

Ok, catch you later. Have a nice day - Harry

Hello Ed,
No problems with the callsign. As soon as you get the new callsign then let me or Ivan know and we can change the registered account name to the new callsign.
We could have updated the old callsign password or changed to a new known password. Please let us know when you wish to proceed, then we can delete duplicated (dead) accounts.

As regards the coil formers, I think I stated somewhere in the blurb that I used 6mm diameter formers for all the formers. All formers had an adjustable ferrite core fitted. As for wire size, at that time I just used "thin" wire that was robbed from transformers. I found that the only effect of the wire size was that you could get closer turn-spacing and a small increase in the inductance for the same winding area. The wire diameter was in the region of 0.15mm to 0.25mm - "pretty thin"!!

I also used several different valves (tubes) when I built that project and found that there was no noticeable difference between the various tube types, as long as they are ECC<SomethingOrOther>. I still have a load of CV455 (ECC81) and they are really good and stable tubes. I also have a few 6N1P Russian tubes, which I find are not so noisy in audio work (but they have ONLY 6.3V parallel heaters on pins 4 & 5).

Another point I found was that the wires leading to the coils for the higher-frequencies shold be ridgid and as short as possible.

Unfortunately this is one of those projects that I lost during my "life changes", mainly because it was too large. I was able to keep the 2-tube TX and I still have that here now, fully operational, after about 25 years

I don't know if you have any problems sourcing components, but in those days used www.tubesanmore.com (Antique Electronic Supply). They also had a load of ready-made IF transformers, oscillator and antenna coils for tubes, and it was some of those I "reverse engineered" for various projects, then re-calculated for other bands. If you cannot get a speaker transformer then you can wind one yourself (details on my homepages) or use a 230V-6.3V transformer - strip it down and re-assemble the E and I laminations in the same direction, but with a bit of copy-paper between the 2 halves to prevent DC saturation.

Anyway, good luck with the project. Very best regards from Harry - sm0vpo

-----------------------

PS - I just searched the members list for " zl1t* " in the users and I only found " zl1tet ", registered  4ᵗʰ April 2024 - DS

Hi Edward,

your registration is "zl1tet". You can log in and use full features of this messageboard. Your recent post was done as a guest, though. Maybe the callsign is too short for a guest's username? Guests have restricted rights.

There is no other registered username starting with "zl1". I checked.

A ferrite slug makes a coil easily tunable. Air coils have better Q, but their tuning range is zero or quite little, depending on their design.

Very best regards from Ivan OK1SIP

Hi Ivan,

Appreciate the pointers! My call sign is (or at least was) ZL1TET, but it's been a long time since I've used it; I'm finally getting back into building stuff. I tried to post this message using zl1tet as my username but site wouldn't let me; I have a feeling I used this forum before using a different email address, not sure.. Anyway, I've contacted NZART to renew my licence so I'll see what they'll decided regarding my old call sign.

I've had some good progress with this receiver starting with the local oscillator (4.0 - 11.0 MHz) . I decided to use an ECC85 as I don't have any ECC81s; before your reply I wound an air coil on a 42mm former as a start but the output from the local was distorting half way up the band and the frequency was well under what it should have been. Frequency aside, I was curious to understand how the distortion was being created halfway across the tuning range; the amplitude from a loosely coupled secondary to attached to my scope showed a large amplitude output when distorting - so that was my focal point. Evidently the distortion was down to overdriving the valve due to my tank inductor. Using some very handy inductance and Q plus Tank Fo calculators (M0UKD's website) I discovered that my 42mm coil had a big change in Q particular at the point of distorting, and onward from that point. So together with your advice and the calculators I ended up with a 12mm air coil 18mm long that had a reasonably flat and lower Q. Using this new coil fixed my distortion, as well as giving me a reasonably linear output at the right frequencies. Good fun!

I'm moving onto the front end and IF now; retrieved a few old valve radio carcasses from the loft for some IF tranys.                  

Regarding your comment about using slugs, apart from making the inductor smaller at lower frequencies, what other advantages would they offer (at 4.0 - 11.0 MHz) and in particular in a local oscillator?

Cheers, Edward.

Hi,

maybe all coils should be wound like L1, i.e. "on a 1/4" (6mm) former fitted with a ferrite slug". Just my guess. I would use litz wire for 0,5 - 4 MHz and cca 0,5 mm magnet wire for shortwave ranges. The coils must be tuned using the slug anyway. But let Harry tell the proven parameters.

You wrote "zl1tetd" here, but you registered as "zl1tet". I suppose it is your callsign, am I right? What is your first name?

VBR from Ivan OK1SIP

And the wire size too please!

Hi,

I'm in the process of making the All Valve Receiver and I've noticed the coil former sizes for L1 and L2 are not listed. Could you let me know what the sizes are please? The only coil information is listed below:

L2

• 550 KHz - 1.65 MHz - 200 turns, tapped at 28 turns. (updated*)
  
  
• 1.5 MHz - 4.20 MHz - 78 turns, tapped at 12 turns.
  
  
• 4.0 MHz - 11.0 MHz - 27 turns, tapped at 4 turns
  
  
• 10 MHz - 25 MHz - 12 turns, tapped at 2 turns
  
  
• 20 MHz - 32 MHz - 5 turns, tapped at 1turn
  
  
• Space for extra band to be used as 'bandspread'.
  
  

L3

• 550 KHz - 1.65 MHz - 110 turns, tapped at 10 turns, + 68pf. (updated*)
  
  
• 1.5 MHz - 4.20 MHz - 70 turns, tapped at 11 turns.
  
  
• 4.0 MHz - 11.0 MHz - 24 turns, tapped at 4 turns
  
  
• 10 MHz - 25 MHz - 12 turns, tapped at 3 turns
  
  
• 20 MHz - 32 MHz - 5 turns, tapped at 1.5 turns
  
  
• Space for extra band to be used as 'bandspread'.
  
  

 

Thanks!