radio's drifting because of weather ?

[gezza123]

HI: Is anybody else having radio driffting off on 40mtr's ,(not fade-ing)dew to the weather condictions,mine are drifting in the same heated room at 20c,also noticed distortion on fm radio merseyside.have to keep ajusting the tuner/BFO/Aerial trim ,is my aerial detecting jack frost ,or is it the Aerial contracting/expanding Gezza

[Herald1360]

Actual drift would have to be the local oscillator- are you sure they're really at constant temperature with no stray draughts?

[HamishBoxer]

Should be no problems at 20C.

David

[Dave757]

Originally Posted by gezza123

HI: Is anybody else having radio driffting off on 40mtr's ,(not fade-ing)dew to the weather condictions,mine are drifting in the same heated room at 20c,also noticed distortion on fm radio merseyside.have to keep ajusting the tuner/BFO/Aerial trim ,is my aerial detecting jack frost ,or is it the Aerial contracting/expanding Gezza

Hi Gezza,

What radio is it?

Kind regards Dave

[gezza123]

thanks for responce guy's

HIave .It is a Lafayette KT-340, the ht power supply has been modified,and a ht choke fitted + an ob2 regulator fitted,also mods to V1 RF cathode resistor r3 330ohm changed to 68ohm for more gain,
also mods to v2 mixer c5 changed from 22pf to 68pf + series 33ohm resistor for better selectivity .
Set has been working fine but now when I power it up it takes ages to settle but still have to repeatidly ajust banspread/tuning,

regards Gezza

[HamishBoxer]

This sounds like duff components,try blowing cold from dryer or freezer around osc stage when its stable to locate suspect.

David

[Sean Williams]

Er Nuff said about the radio - they were known for drifitng VFOs.....

A good look around the VFO for iffy components would do for a start, but I don't think there is much that can be done with this design

Sean

[Skywave]

What Hamish has said above is obviously the first place to start: change all the mica caps. in the local oscillator just to eliminate those first. However, from my own experience, what Sean has reported does have some validity.

Now I've never tried this, but it is a simple thing to try: wrap cotton wool around the local oscillator coil. (Note: some trimming of the alignment may be necessary after doing this). It is coils in LO's that are usually the main cause of drift (defective components aside). That simple expediant should insulate the coil from circulating air currents (hot air - from components beneath the chassis - and cold air, from air outside the chassis being drawn in.) In line with this thinking, it is my common practice to re-mount power resistors that get hot above the chassis.

Afterthought. If the drift is present on ssb signals, but not on AM signals, then the BFO becomes the main suspect - but the suggestion above still applies: just substitute "BFO coil" for "LO coil".
Mind you, it wouldn't do any harm to do both.

Al. [Skywave]

[Sideband]

Originally Posted by gezza123

thanks for responce guy's

HIave .,also mods to V1 RF cathode resistor r3 330ohm changed to 68ohm for more gain,
also mods to v2 mixer c5 changed from 22pf to 68pf + series 33ohm resistor for better selectivity .
Set has been working fine but now when I power it up it takes ages to settle but still have to repeatidly ajust banspread/tuning,
regards Gezza

I presume you used hi stability components for these mods?


SB

[G8HQP Dave]

V1 will now be taking more current, so will get hotter. Is it anywhere near the local oscillator? If this valve has a little gas inside it then it may head towards thermal runaway due to grid current, which could drag the HT voltage down.

You said you changed C5 in the mixer circuit - what does this do? If it means that the oscillator is more closely coupled to the mixer than this could increase drift The same if it means that the valves are more closely coupled to the tuned circuits.

It may be that you have taken a weak design and 'improved' it in a negative direction.

[Dave757]

Hi Gezza,

These kit built sets were never noted for high stability.
I would think the best way to tame it would be to
switch it on a few hours before use and let it 'soak'
well in advance.

A lot of cheaper 1960's sets were not very stable,
(and some of the dearer ones)!

Kind regards

Dave

[gezza123]

HI: Guy's.sorry for delay.
First I must thank you all for the input you have put in for me.Thanks you one and all.

Now I must point out that I never did any mods to the radio,I have the notes off the engineer who built the set in 1964/5.and done the modifications.
May have possibly worked ok back then.

So back to the problem,I have checked all around the OSC and found no components faulty.But I did find a very bad earth on c15,o.o1 on cathode of v5 giving intermittent S Meter results,I need a big iron for that so fixed temp.

I have put part of the front end circuit on for you to see the mods made to
v2grid and v3 osc grid, sorry for the hand drawing .
I have now put it back as was pre the circuit 1964
Set is working ok at minute,not drifting and comes on from cold,so this may have been the problem upsetting the OSC.

Dave G8HQP c5 200pf on mixer v2 was changed to 68pf+68ohm series resistor,also the cathode resistor on v1 I have changed back to 330ohm,the mods was done for better selectivity/gain acording to the notes.
Thank you all again for the help.regards Gezza

[G8HQP Dave]

You said that C5 was changed from 22pF, not 250pF. That threw me off the scent!

Without knowing what the valves are it is difficult to say whether the cathode resistor should be 330, 68 or something in between. For some reason I haven't quite grasped, Japanese radios sometimes choose a 'hot' RF valve but then bias it back to lower gain (and higher noise). Maybe the 'hot' valve looks good in the adverts, but would overload the weak multigrid mixer if run at full tilt? They might get better results by running a lower gain valve at full gain.

How was the voltage stabiliser added? If done badly this could increase drift.

[gezza123]

HI, Dave sorry about that wrong information in post #5.

The power supply has been moded so I am trying to draw it out the voltage stay's at 108volts ,I have been listening to the Shannon Velmot Report on 5505Khz most of the day and is stable.not much traffic been on today so could not test 40mtrs.looking at the power it is a bit of a mess so need to tidy it up.have put circuit on so you can see the valve line up.+ what I can make out of the mods to power.you may notice that the ht to the sound output is higher,I think its because of the output valve has been changed to an el91.thanks Gezza.

[Skywave]

I have one of these that's been sitting on my 'one rainy day when I've nothing else better to do' shelf for ages.

The EL91 produces less O/P power than a 6AQ5; an EL91 will draw substantially less HT current. Provided that the biassing has been changed accordingly and that the level of audio out to the speaker or phones is adequate, I'd say that that change was worthwhile. In fact, I implemented this very change in my Eddystone 888A receiver many, many years ago; never had any problems. It was not found necessary to replace the O/P transformer.

However, the revised PSU cct. you've shown doesn't look right. I can understand the stabilised +108v HT being fed to the local oscillator and BFO only, but surely the HT+ feeds to the rest of the set (apart from the take-off for the anode of the O/P valve) should come from the junction of the 2k, 1k5 and the 64uF ?
This arrangement will then require a replacement function switch S1 - two "HT+ out" lines (stab. HT+ and unstab. HT+) now need to be switched independently but in synchronism - using the revised PSU arrangement as shown. (There are ways around 'replacing S2' to achieve the desired result - but then the revised PSU filtering cct. changes).

And talking of which . . . .

If you look carefully at the function switch, S1, and analyse its operation carefully - in accordance with how it is drawn on the schematic - you'll correctly conclude that the drawing is in error. The two 'ears' on the rotating connector segment are clearly shown with a spacing between them which is equal to the distances between any adjacent pair of fixed contacts: OFF / REC or REC / SEND or SEND / Q-MULT. It soon becomes evident that the distance between these 'ears' on the rotating segment should be shown as TWICE the distance between these corresponding fixed contacts. Otherwise you get the absurd conditions of (a) REC joined to SEND (this is actually OK) but (b) next position is SEND joined to Q-MULT (with no HT being fed to the receiver front end! ). The required condition of 'SEND'. i.e. HT switched OFF to all receiver sections but heaters remaining energised, then never arises!

An aside - but still relevant to the Thread -
One of the mods. that I have considered for this set over the years includes replacing the noisy 6BE6 mixer with the much quieter ECF82. Configuration: pentode mixer with the triode section as a cathode follower between the local osc. and the pentode mixer.

Al. [Skywave]

[G8HQP Dave]

As the RF stage is a 6BA6 my comments about 'hot' valves do not apply. You probably could reduce the cathode resistor, as 330 is quite high. The screen grid voltage is a bit high, so I would try 120 or 150 in the cathode.

Could you confirm that R7 and C7 are still present in the HT feed to the oscillator? They will isolate the osc from the voltage stabiliser. Without them the osc sees the varying impedance of the stabiliser, which can reduce stability.

[gezza123]

HI: Al .The bias cap has been changed to a 25uf,and pin 6 changed to pin seven and the sound is good.

The switch has been modified ,the 1k-r7 has been disconnected from the function switch and taken via a choke to the ht line at the junction of the
1.5k/ob2 regulator.
You may be right on the ht to the output stage as I think this may be a bit high,so I will connect this to the junction of the 64uf and see how it works.

Hi: dave.c7 is still in circuit ,but as above r7 has been changed to a choke
thanks to you both for the input,I will let you know how I get on.Gezza

[G8HQP Dave]

I think R7 is best left as a resistor. It is not there as an HT smoother, but to isolate the local oscillator from the HT supply. There is no need for an LF choke here. An RF choke could make things worse by introducing reactance exactly where you don't want it.

[Colinaps]

Have you measured your mains voltage? I bet it changes with the weather!

Cheers,

Colin mm1aps.

[gezza123]

HI Guy's.back again.
checked the mains Collin stable all day at 230v-233.3v thanks .

Dave .I have done what you said ,changed the cathode resistor to 180ohm
also I check the setup in the manual an pin 5 and 6 of v1 the rf amp
have the same voltage on them.I also connected back the 1kohm
in place of the choke,this has reduced the ht a little on the Osc.

Al. I tried connecting to the junction of the 64uf ,had no affect except
the 10watt resistor got quite hot,so put it back.

So what I have done is ,put the 1kohm back in the osc.
removed the choke,and reconnected the two grids of v2 and v3 back together.Took out the BFO and checked and resolder seems ok.that was easy to do.The set seems to work a lot better now and have even recieved the ham report today + shannon airport and shannon velmot and a few of them that blast from the isle of erin.I will change a few more caps when I get them.
So thanks a bundle for all your input ,I will keep the iron on just in case I need you all.Regards Gezza