Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: Prev 1 [2] 3 Next   Go Down
  Print  
Author Topic: Tube VFO  (Read 12051 times)
KB1WSY
Member

Posts: 671




Ignore
« Reply #15 on: August 05, 2012, 04:16:40 PM »

If you can't find a command tx carcase, then you need a double bearing - front and rear - variable capacitor. Something around 100 pF swing...

If I don't go via the ARC-5 route, I have access via various sources to the Hammarlund MC-140-S variable (140 pF) listed in the original ARRL parts list. It is rather hard to tell from the dreadful ARRL archive scan, but there does appear to be a thick metal plate at the *back* of the capacitor which I assume means it is probably double-bearing. I will investigate further.

... and a ceramic coil former....<<

For the coil, I was thinking of using an Air Dux 816A that I picked up in a dusty box of two dozen Air Dux coils a while back. It is not ceramic, but it is very stiff because of the perspex bar that runs through it and it could be mounted on a couple of standoffs for a very rigid assembly. (It, too, is similar to the part used in the ARRL original but it's better because of the perspex mounting bar.)

I'd suggest also some old APC type trimmers.

Yes I already have a bunch of those, and they are indeed specified in the ARRL design.

If you get stuck, it is quite possible that the RZP junkbox may be able to help, although you would have to wait until I go to the US again next year - unless you have a relation from the UK visiting or one from the US coming here.

Thank you very much for your generous offer and I'll take you up on it if I get stuck. The VFO project, parts wise, looks easier than the others I'm doing. I do visit the UK at least once a year and it will be more frequent in the next couple of years, for various reasons (my sister lives in London, my brother and mother live in Paris, a couple of dear friends live in the southern UK, and starting this autumn my daughter will be at college in France). Would love to get together some time. When I'm in the UK it's either in London or, more often, in Hampshire and your QTH is halfway between those points....

73 de Martin, KB1WSY

« Last Edit: August 05, 2012, 04:20:17 PM by KB1WSY » Logged
W8JI
Member

Posts: 9304


WWW

Ignore
« Reply #16 on: August 05, 2012, 05:30:14 PM »

I'm with Peter on most of his comments.

As for VFO frequency, almost everything used 1.75 - 2 MHz as the base frequency, and then doubled for 80.  Most VFO's used 7 MHz for 40 meters on up, except for VHF stuff.

The reasoning behind this is the VFO only needs temperature compensate on two frequency ranges, as opposed to multiple ranges. Also, most ham rigs were designed to use 160 or 80 crystals on 80, and 40 meter crystals on 40 and higher. There are several higher frequency advantages to multiplying that offset drift multiplication. The VFO becomes nearly immune to transmitter RF leaking back into the VFO, frequency pulling from unneutralized early transmitter stages, and early transmitter stages are more stable on higher bands. Without multiplying, early stages usually lack shielding and neutralization for upper HF stability.

Even VFO's for 80 - 10 often start on 160. The Knight T150 runs the 80 and 40 meter VFO position on 160, and the 40 and up on 80 meters! But that is a little goofy and results in spurs.

http://www.w8ji.com/allied_knight_t150.htm

It's pretty easy to temp comp the VFO on two bands.

As for chassis, I build my VFO's by double-walling the chassis with a thicker gauge inverted chassis. All the VFO frequency determining parts are inside that upper inverted chassis, the tube on the outside.

73 Tom
Logged
KV1E
Member

Posts: 10




Ignore
« Reply #17 on: August 05, 2012, 08:13:41 PM »

Your choice of the W1ICP VFO in the QST article would be as good a choice as any.  I used that one to build a 80 - 10 rig 5 years ago and did a lot of research on vfos since this was my first attempt at building one.  I used sheet aluminum to make a box for the components and had the 12UA7 mounted externally on the back.  The whole box was then fastened to a 3/16" piece of aluminum used as a base.  That was fastened to the chassis with bolts and rubber grommets.  It was just fine for my use.  I didn't plan to use it mobile or in a tank.  Sitting on a desk it seems very stable.  The VFO covers 3.5 to 3.6 MHz.

Logged
G3RZP
Member

Posts: 4391




Ignore
« Reply #18 on: August 06, 2012, 07:22:11 AM »

I haven't found the Airdux or B&W coilstock so good on thermal stability. What I have got for when a round to it is the ceramic former from a BC221: that has a bi-metallic strip temperature compensator, which should be very good.

Formers for laboratory standards were often made so that they expanded more lengthways than diametrically: by getting the ratios of the expansions right, you had a very temperature stable coil.

If you want to see a once popular in Europe (but God knows why) VFO/exciter, search for 'Geloso VFO'. 6J5 Clapp oscillator on 3.5 or 7, 6AU6 multiplier and 6L6 or 6V6 (depending on model) buffer/multiplier. Nasty little coils on muckite formers, and a chassis in about 18 SWG aluminium.
Logged
KB1WSY
Member

Posts: 671




Ignore
« Reply #19 on: August 06, 2012, 08:43:11 AM »

Your choice of the W1ICP VFO in the QST article would be as good a choice as any.  I used that one to build a 80 - 10 rig 5 years ago and did a lot of research on vfos since this was my first attempt at building one.  I used sheet aluminum to make a box for the components and had the 12UA7 mounted externally on the back.  The whole box was then fastened to a 3/16" piece of aluminum used as a base.  That was fastened to the chassis with bolts and rubber grommets.  It was just fine for my use. The VFO covers 3.5 to 3.6 MHz.

Did you build the whole QST project including the buffer/multiplier? If so, did you put the 2nd tube also on the same box or did you segregate it elsewhere?

Approximately what combination of L/C did you find optimal to get the 3.5 to 3.6 MHz range? (That's exactly the range I had in mind, unlike the QST circuit which covers 3.5 to 4.0.)

What component provenance did you use for the tuned circuit? Relatively "medium quality" such as in the QST design or did you branch out and use something heavy duty like ex-military stuff?

Thanks for the encouragement, it's great to hear from someone who's actually built that QST design recently.

Formers for laboratory standards were often made so that they expanded more lengthways than diametrically: by getting the ratios of the expansions right, you had a very temperature stable coil.

Yes I read about that somewhere, very clever. Perhaps the optimal approach would be to use high stability components in the oscillator tuned circuit as you suggest, and then relegate my Air Dux or B&W stock (I have both) to the buffer/multiplier output coil where, if I am not mistaken, stability is much less critical. Am I mistaken?

I'm looking forward to being opportunistic at hamfests, and seeing what I can pick up, rather than pre-designing everything on the basis of what could be unobtainium or expensive parts. There is plenty of time, I can get on the air with my "Novice Rig" some time this autumn and live with the frustrations of crystal control for a while, that should provide plenty of motivation to build the VFO! (But I have lots of Xtals -- about two dozen, split between 40m and 80m -- so it shouldn't be *too* frustrating, hopefully).

73 de Martin, KB1WSY
Logged
W8JI
Member

Posts: 9304


WWW

Ignore
« Reply #20 on: August 06, 2012, 06:17:20 PM »

Yes I read about that somewhere, very clever. Perhaps the optimal approach would be to use high stability components in the oscillator tuned circuit as you suggest, and then relegate my Air Dux or B&W stock (I have both) to the buffer/multiplier output coil where, if I am not mistaken, stability is much less critical. Am I mistaken?

The main issue with the B&W coils are the spacer rods expanding. The plastic rods are far worse than the copper.

Coils wound on porcelean or dry wood are far more stable than a B&W air coil for heat, although wood will give you very slow long term drift from moisture changes...so you need to seal it.

I successfully used an old test tube I swiped from the science room at school.

Tom
Logged
G3RZP
Member

Posts: 4391




Ignore
« Reply #21 on: August 07, 2012, 12:49:11 AM »

Some of the pre WW2 German military equipment used silver deposited onto a ceramic former for a stable inductor. I suppose that the silver loaded conductive paint on a glass tube might do the same, but the results are unlikely to be worth the trouble.
Logged
W8JI
Member

Posts: 9304


WWW

Ignore
« Reply #22 on: August 07, 2012, 04:54:34 AM »

Some of the pre WW2 German military equipment used silver deposited onto a ceramic former for a stable inductor. I suppose that the silver loaded conductive paint on a glass tube might do the same, but the results are unlikely to be worth the trouble.

I always just wound tight with warm soft wire, and used electrical varnish. Copper, glass, ceramic, phenolic, and wood are way better than the plastics used in air coils. I just can't figure why people put Airdux in VFO's, other than ease of construction.
Logged
N2EY
Member

Posts: 3856




Ignore
« Reply #23 on: August 07, 2012, 06:41:18 AM »

There's an easier way.

Get an old ARC-5 tx. One that's been hacked up is fine, as long as the oscillator compartment, chassis and PA tuning cap are good.

Cut it down as described by W2EWP in the book "Command Sets" (free download from the Glowbug website)

Build your favorite VFO circuit on it, using the ARC-5 coil and capacitors.

Much easier than metalwork.

If you decide to use the W2YM/W1ICP design, put the 12AU7A on the ARC-5 chassis and the 6AU6 buffer on another chassis, to keep the heat down. The buffer chassis could be a larger chassis that the cut-down ARC-5 chassis sits on.

Or keep the ARC-5 chassis original size and put the buffer stage in the front.

73 de Jim, N2EY
« Last Edit: August 07, 2012, 06:43:30 AM by N2EY » Logged
KB1WSY
Member

Posts: 671




Ignore
« Reply #24 on: August 08, 2012, 12:18:21 PM »

Today I obtained, from a former ham family a one-hour drive from my home, a James Millen variable capacitor "new in box" that will be the centerpiece of my VFO. Pictures here:

http://tinyurl.com/cgqwqfz

http://tinyurl.com/8dg2gec

Cost was about $15 not including gas. Now I just need to settle on the right inductor (or build it from scratch).

This 2-section capacitor is 35pF per section and I think 70pF will probably be just about right to get the range I want (the CW portions of 80 and 40).

Note that the main shaft and the reduction shaft are at 90 degrees to each other. Unless I use some kind of "right angle converter" mechanism, this means that if the dial is on the front panel, the main tuning knob will be on the side of the box. Or, I could put the main tuning knob on the front and use the outside edge of a homebrewed cylinder as the dial. I think N2EY has been through this before? Suggestions, Jim, you are the expert with cereal bowls or whatever? How did you do it on the Southgate 7?

The mechanism is absolutely stable and free from backlash, which argues against adding any kind of "right angle converter" that might pollute the mechanism with a little bit of play.

Ergonomically speaking, having the knob on the right hand side of the box might even be better than on the front panel (at least if you are right-handed).

IMHO this VFO will rock! "Build it like a tank" eh?

73 de Martin, KB1WSY
Logged
N2EY
Member

Posts: 3856




Ignore
« Reply #25 on: August 11, 2012, 04:39:59 AM »

Today I obtained, from a former ham family a one-hour drive from my home, a James Millen variable capacitor "new in box" that will be the centerpiece of my VFO. Pictures here:

http://tinyurl.com/cgqwqfz

http://tinyurl.com/8dg2gec

Cost was about $15 not including gas. Now I just need to settle on the right inductor (or build it from scratch).

This 2-section capacitor is 35pF per section and I think 70pF will probably be just about right to get the range I want (the CW portions of 80 and 40).

It's nice but I wouldn't use it for a VFO. It's meant for transmitter use in a push-pull circuit.

The plates are aluminum IIRC. You want brass or invar for a VFO.


Note that the main shaft and the reduction shaft are at 90 degrees to each other. Unless I use some kind of "right angle converter" mechanism, this means that if the dial is on the front panel, the main tuning knob will be on the side of the box. Or, I could put the main tuning knob on the front and use the outside edge of a homebrewed cylinder as the dial. I think N2EY has been through this before? Suggestions, Jim, you are the expert with cereal bowls or whatever? How did you do it on the Southgate 7?

A drum dial- a cylinder mounted on the end of the capacitor shaft. Similar to what is done in the BC-221 and LM frequency meters.

The mechanism is absolutely stable and free from backlash, which argues against adding any kind of "right angle converter" that might pollute the mechanism with a little bit of play.

Ergonomically speaking, having the knob on the right hand side of the box might even be better than on the front panel (at least if you are right-handed).

IMHO this VFO will rock! "Build it like a tank" eh?

73 de Martin, KB1WSY

There are much easier ways, though.

73 de Jim, N2EY
Logged
W8JI
Member

Posts: 9304


WWW

Ignore
« Reply #26 on: August 11, 2012, 05:33:13 AM »

Today I obtained, from a former ham family a one-hour drive from my home, a James Millen variable capacitor "new in box" that will be the centerpiece of my VFO. Pictures here:

http://tinyurl.com/cgqwqfz

http://tinyurl.com/8dg2gec

Cost was about $15 not including gas. Now I just need to settle on the right inductor (or build it from scratch).

This 2-section capacitor is 35pF per section and I think 70pF will probably be just about right to get the range I want (the CW portions of 80 and 40).

You should be OK on capacitance. A wide range of capacitances will work, if you pick the correct external component values. That looks like a workable capacitor provided it does not have backlash in the drive. The drive looks like it might be spring loaded. The thing I strongly don't like about that capacitor is the floating shaft at the open ends. That might be a problem with shaft centerline flexing movement and external shaft ground paths, especially if you load the shaft with an indicator or use a metallic connection to it.


Quote
Note that the main shaft and the reduction shaft are at 90 degrees to each other. Unless I use some kind of "right angle converter" mechanism, this means that if the dial is on the front panel, the main tuning knob will be on the side of the box. Or, I could put the main tuning knob on the front and use the outside edge of a homebrewed cylinder as the dial.


I would never drive a VFO through any unnecessary angle conversions, so I would align the tuning shaft in the knob direction. The frequency indicator could use an angle conversion scheme without harm, because backlash would not matter nearly as much. I would worry  above main shaft rigidity.

Quote
The mechanism is absolutely stable and free from backlash, which argues against adding any kind of "right angle converter" that might pollute the mechanism with a little bit of play.

Right. Although any ill effect would be reduced because it would be followed by a gear reduction, I would still avoid it.
 
I have very successfully used regular variable capacitors of many types, with no special attention to anything but good solid bearings and shaft grounds. The only thing that concerns me is the main rotor shaft is not grounded at outer ends and has no support. In my opinion, that is a potentially serious issue. First, you probably cannot rigidly "ground" and mount the main-shaft exiting the capacitor. That means you cannot hang any large metallic things on it, so you would need to use insulated couplings and not deflect the shaft with variable pressure. If you flex the main-shaft or erratically form a ground path, you are into unwanted frequency movement issues.

I've never had issues using a wide variety of double bushing capacitors, but what you generally want is a compact, rigid, capacitor with fairly thick plates and good rotor shaft to frame electrical paths at BOTH ends of the rotor.

73 Tom




 
Logged
KB1WSY
Member

Posts: 671




Ignore
« Reply #27 on: August 11, 2012, 06:46:56 AM »

Thankyou Jim and Tom. After obtaining the capacitor I did start thinking a little bit along the lines you've laid out (especially about the lack of a bearing at both ends of the shaft). Your expertise helped seal it for me. But it's a "keeper" and I'm looking forward to finding an application for it in the future, just not in the VFO -- meanwhile it is a nice ornament for my desk!

73 de Martin, KB1WSY
Logged
KB1LKR
Member

Posts: 1899




Ignore
« Reply #28 on: August 11, 2012, 10:27:54 AM »

Foe what it's worth, here's a web link on 6AG7 & miniature version 6CL6 tubes.
http://faculty.frostburg.edu/phys/latta/ee/6cl6xmtr/6cl6why6cl6.html

"In 1950, an important article in QST magazine, "Crystal-Controlled Oscillators, A Review of Modern Crystals, Circuits and Tubes" (QST, March 1950, C. Vernon Chambers, W1JEQ) addressed all of the issues above, as well as the choice of which tube to use. In that article, various electron-coupled circuits were tried along with a variety of tubes: the 6AG7, 6F6, 6V6GT, and 6L6. Among the many conclusions in the article, one came through loud and clear, which I quote here: "Of the four tubes tested the 6AG7 is by far the best from every standpoint." As a result of that article, virtually all crystal oscillator circuits in the ARRL handbook for the next 15 years featured or recommended the use of the 6AG7."

"The 6CL6 is released by GE as the miniature version of the 6AG7 in May, 1953
On the first page of the May, 1953 issue of QST, an ad appeared by GE announcing the 6CL6 as the new, miniature replacement for the 6AG7. The new 6CL6 was the electrical equivalent of the 6AG7 in a cheaper, smaller, package..."

See his other articles:
http://faculty.frostburg.edu/phys/latta/ee/wing2e26xctr/2e26oscillator.html
http://faculty.frostburg.edu/phys/latta/ee/wing813amp/amplifierpics1.html
etc.
Logged
G3RZP
Member

Posts: 4391




Ignore
« Reply #29 on: August 11, 2012, 01:56:43 PM »

And later on, when the Vackar became the range, a different set of tubes became magic........

Being cynical (who, me?) I have never seen that much deiference except in power output.

My choice is a 6SJ7 with fairly low voltage - 150 volts  - in an ECO. Maybe a 6SJ7Y., although how much difference it makes, I know not.

Then a 6AG7 buffer - or a 6CL6 or even a 5763.

Let's face it, we ain't looking for WWV stability here!
Logged
Pages: Prev 1 [2] 3 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!