Print Page

I have been reading your past Tech Info messages - especially those of interest to me. You have mentioned how the Permeable Tuned Oscillator has very good stability. In looking at my several Heath equipment that I want use (have been in long time storage - yes), I feel that in the process, they need to be stabilized better (both receivers - the HR-10B - and VFOs). I think that this might be a good approach.

I have read this paper by ZL2PD (April 2014), and am considering trying to build using his tuning approach:
ZL2PD
"A Permeability Tuned Oscillator"
http://zl2pd.com/ptovfo.html

My question is - knowing that the frequency vs winding is not a constant - but needs to be spreading toward one end - I am interested in any (and all) advice on winding this coil.

I have obtained a few ceramic coil forms - 1/2 inch outside diameter - by 1 inch long winding area (total length is 1 3/4 inch - the hollow core is about 3/8 inch) (I am searching to find a good brass plug - to tune the coil - per ZL2PD).

Essentially, I am interested in 10 and 20 meters. To start with (to try this) - I have obtained a couple old CB sets - AM (Citifone SS - with the 6AQ5A Final) Their first IF is just above 4 MHz - but I will still need to get correct crystals - so this can adjust - second IF is 455 KHz).
On 10, I am mostly interested in the AM band - 29.0-29.2 MHz. I can put the PTO near 5 MHz - to convert to the 2nd IF, and use a Crystal Osc to the first IF - or I can put the PTO on the 29 MHz converter - using a frequency near 25 MHz (or double up from 12.5 MHz). - 20 Meters would be built similar.

I would probably use the standard 1/4 inch bolt with 20 turns per inch - given the coil's 1 inch tuning length - for the brass plug - and somehow get a frequency marker.
If this were mounted parallel to the front panel, an indicator could slide along the front, giving several frequency indicies. (somewhere finding a gear to make the right angle drive to the PTO from a front panel knob.)
If this were mounted back from the front panel, some type of multiturn indicator would be needed (National?).

advice requested

Cliff W9LCE

Cliff,

It sounds like you want to build a PTO. The article you link to covers the home brewing aspect pretty well. The same author also covers a DDS VFO equally well.

If the goal was to achieve Collins PTO linearity and stability, then it could take you many tries to even remotely approach those goals.

So how would you put a home made PTO into a HR-10 receiver, or some other Heathkit VFO such as a VF-1 or HG-10?

The author you linked too did not get great stability, he even referenced adding a Huff & Puff stabilizer as a future project.

I think I would pick a VFO topology and just go through the trials and tribulations of enhancing stability, and see where the effort takes you.

I went down that road with an old Knight Kit V44 VFO...here is my story:

http://www.cumbriadesigns.co.uk/x-lock_solutions.htm

My story is about 1/2 way down the page.

Jim
Wd5JKO

I believe one of the reasons for choosing a PTO versus a variable capacitor was the ability to achieve very linear change versus rotation which was quite important before electronic counter type readouts and later digital synthesis became the norm. QST had an article, back in the 1970s I believe, that showed the variable winding pitch along with a specially shaped moveable core to achieve a linear change versus rotation. Of course Collins uses the well known corrector stack to allow fine tuning the linearity.

As Jim stated constructing a good PTO is going to be far more difficult than constructing a good capacitor tuned oscillator. If you use good quality components, (the variable capacitor in particular is critical), good construction practice, excellent voltage regulation, and consider solid state then you can end up with a very nice capacitor tuned VFO that is easily stable enough for AM, CW, and SSB although if digital modes were your primary concern then a DDS solution would be desirable. I just ran some informal drift tests on a recently acquired Kenwood R-599A receiver I acquired and when WWV is tuned to zero beat at initial power up there is maybe 50 hertz drift at worse after 20 minutes and once corrected again it stays put. This is typical of a good quality solid state VFO from the 1970s and a reminder to me of what nice and very compact all mode receivers Kenwood produced in the 599 line.

There is a good reason you don't see a lot of articles for building a homebrew PTO given the mechanical issues. If you are set on using a PTO consider something like a Drake parts unit to salvage the PTO and use a crystal controlled mixer to put the output on the desired frequency. Of course the Collins PTO units are excellent if in good shape but they are larger and more difficult to fit. I guess the Ten Tec PTOs are OK but I am a bit leery of the overall quality given the problems with the grease/slipping/wearing drives. I just picked up a nice Century 21 digital that will at the least need grease and possibly parts to fix the PTO. The only practical way to have multiple frequency range output from a PTO is through using mixers, band switching capacitor tuned VFO units can be practically built although they are at a severe disadvantage to a single range type for stability.

It seems like a LOT more effort then its worth, but that might be what floats your boat.
I have almost no 'ham' gear around because it was all made to a price point, with a lot of bad design thrown in, with Heathkit being about the worst.

If taking something that is really poor and making it work good enough to use is what you enjoy, Heathkit is your top pick.

I am sure you can improve what heathkit did with less fuss then a home made PTO.
I am also sure that it is easy to build a very stable VFO, the LO in my homebrew receiver is very stable, and my VFO/exciter is quite stable.
How stable does it need to be?
Not very for AM and CW, a bit better for ssb, but these days with digital rigs, you need to be able to dial in 7270.000 MHz and be spot on frequency.
You can tell the guys running vintage gear because they are on some odd frequency like 7271.850.
On AM, its not unusual for people to be 2 KHz off.

Heathkit stuff lends itself to some simple easy improvements and its kind of silly to try and turn it into a Collins rig, unless you like that sort of thing.

Did not the Heathkit SB series radios use a pto?
I know the HW series used a cap vfo, and I thought the SB series was an upgrade from the HW series because of the pto.
Not sure a heathkit pto is great, but it might be on the right frequency and easier then building a pto.

If you was to use a digital display, you do not need to have anything track a scale, the frequency counter reads the pto frequency and has an offset to read the correct frequency.
Besides not needing a scale, its self correcting, if something drifts/changes it shows on the display.

While the HW series used a capacitor-tuned VFO that was assembled by the kit builder, the SB series used a factory assembled capacitor-tuned VFO called an LMO, Heath's acronym for "Linear Master Oscillator". I have used the SB-101, 301, 401 using the tube-based LMO, and the SB-102 and SB-303, which had a later version solid state LMO. I found them to be quite stable for SSB and CW operation, and the dial calibration was very linear, less than 500 Hz error at any point between 100 Khz calibration points. The oscillator tuned from 5.0 to 5.5 MHz. I modified one to tune from 3.6 to 4.0 MHz but discarded the dial, and I use that with a counter, as you suggested.

I gotta jump in here. Was working on a PTO tuned colpitts project today but right beside now me is a 80 meter regen PTO tuned that is nothing short of unbelievable. It is a solid state regen rig I designed last year and found great favour in Cuba (ok to say that now) and was featured on their Short Wave Show with Prof Arno Cora. I posted it originally on the Yahoo regen site. It requires almost no retrimming of regen and does SSB like a champ .Above it is a 20meter Solid state Tranmitter AM rig also PTO. The tuning is " definite" and control is much better.
Much harder to build? I don't understand that. You wind a coil, solder a appropriate cap in place of the variable (or use a ceramic hi quality trimmer so you can precalibrate it with a small screw driver through an access hole.
Funny you mention the old DX 60 VFOs. I made mine using a 12ax7 and one of my slug tuned ceramic coil forms. Just a day or so ago I was chastising myself for tuning with a variable cap when everything I needed for PTO is already built into the VFO , just not accessible from the front panel. Having played with it now for awhile I really wonder why I build anything else. Not easy to get a across in words how it works and feels but I have been very impressed.
don

Well, how do you tune it?
Most pto's are multi turn units, and if you want a dial and a scale that reads down to say 1000 Hz, how do you do it?

Caps are 1/2 turn things and you can get 180 degree dial scales and mark things roughly.
I have a pto out of a T368 I was thinking of using for an 80 meter exciter, with a digital display, a multi turn pto works great.
Does not have an old buzzard look though.

Not sure why, but I had better luck with caps and air wound coils over slug tuned coil forms and caps.
It seems hard to wind a coil on a ceramic form and have it stay stable.

Quote from: N2DTS on December 26, 2014, 07:04:47 PM

The LMO does have a capacitor that turns 180 degrees, with a gear drive with a shaft that turns 5 full revolutions, 100 KHz per revolution. The larger gear is actually two gears, spring loaded, to reduce backlash. The coil is wound on a ceramic form, and it must be wound under tension to be stable.

I attached a few pictures of an LMO removed from a parts SB-400, It has a single 6AU6. It includes a piston trimmer, connected via a diode, to allow FSK operation. This is an LMO that I modded using a reed relay to switch in an extra capacitor, so the oscillator covers two bands.

VE3LYX
right beside now me is a 80 meter regen PTO tuned that is nothing short of unbelievable. It is a solid state regen rig I designed last year and found great favour in Cuba (ok to say that now) and was featured on their Short Wave Show with Prof Arno Cora. I posted it originally on the Yahoo regen site. It requires almost no retrimming of regen and does SSB like a champ. Above it is a 20meter Solid state Transmitter AM rig also PTO. The tuning is " definite" and control is much better.
Much harder to build? I don't understand that. You wind a coil, solder a appropriate cap in place of the variable (or use a ceramic hi quality trimmer so you can precalibrate it with a small screw driver through an access hole.
***********
is there anything special about winding that coil? - that was my original question. I heard that it needed a regular variation in spacing.

also - the approach given in the ZL2PD paper was challenged as unstable - I was considering it as a possibility to tune the coil more exactly (using a standard 1/4 inch bolt of 20 turns per inch). I have bought ceramic coil forms.

and - I am interested in a way to record the turns - relative to frequency.

Yes - I want to put it on a DX-60 and maybe on the HR-10b receiver. What really got me started in this - was not doing very well with PSK31, with my SB-102. The SB-300/SB-400 (which I have, but haven't worked on) do allow driving with an external osc - and maybe a PTO would stabilizing them.

where can I find a copy of your 80M regen PTO? - I've looked
Cliff

Shouldn't need a circuit. you simply sub the stock coil for a slug tuned coil and install a good quality stable fixed cap in place of the vari capacitor. https://groups.yahoo.com/neo/groups/regenrx/photos/photostream/lightbox/564014676?orderBy=mtime&sortOrder=desc&photoFilter=ALL (find slugabsorb)I stopped posting schematics a couple of weeks ago because there are always folks who haven't tried it but know it wont work critiquing and I had enough of that for a lifetime. I am purely results oriented. If I try it and it works I use it. If it doesn't I move on.
I didn't use a turns counter although would be nice. I listen for it on a digital rx at first to get an idea of where it is or if a first build use my GDO to adjust the coil vrs fixed cap to get the range. Coils are like kissing. The more you make the better you get at it. A lot of exotic theories which are mostly balderdash. I use solid hook up wire with the plastic insulation for all mine. Spacing is automatic and they always work. Try 16 turns ,(for a DX60 VFO use) see what you get with 100pf across it. If it is low add more if high remove some. By the third try you will be an expert.
Yes you go from 1/2turn for the whole band to 15 to 20 turns. Tuning becomes very defined and much easier and hand capacity effect seems to be non existent.
don
Here was my 20M AM PTO project during constuction

Cliff,

The photo on my posts is a car radio used in a 1964 Rambler automobile. The radio is in really rough shape, with some parts missing, but the permeability tuned stages, and mechanism are intact. I believe there are three stages that were likely for RF, Oscillator, and Mixer tuning all ganged together. Maybe your HR-10 has a three gang tuning capacitor. It might be possible to convert the Rambler radio front end to short wave usage. My Rambler radio is yours for the cost of shipping. PM me if interested.

Jim
Wd5JKO

Good idea . Many 50s car radios were PTO. Philco (Ford to name another one) Some had very intricate tuning devices and would be far superior to what I did just using a slug tuned core which works but isn't really designed for constant changing. I had to smile BTW. First Job I had when I came to Belleville in 1968 was at Waterson RAMBLER. Since he also sold Jags BMC and British Leyland he had a hard time hiring mechanics and so was the best paying shop in town. Also a very good employer. You made my day mentioning the brand. Thanks.
don

If ur going to use a screw, then at least use a 1/4-24 (fine thread).
Also there is slop in most screw threads... so a means of taking up the slop would help.
They sell reasonably precision, but small assemblies that are designed for precision motion applications.

You might want to use what works out to be a micrometer's business end.
They are sold for not all that much these days, and a used micrometer that is a beater
is just fine, since it the end or the anvil that usually gets trashed. Attach a brass or copper rod
to the end to do the tuning...

I still have an old Stewart Warner table radio, brown bakelite, that used a sweet slug tuned system - the slugs went up and down vertically into a set of coils. Radio worked great. When I was a kid, I used to marvel at that assembly.

_-_-

Also there is slop in most screw threads... so a means of taking up the slop would help.

While true in practice even a coarse thread and a slug is so far ahead of a vari cap set up it kinda doesn't matter.
micometer threads are 40 turns to the inch (40 X.025 = 1.000") I believe a 5/16 NF is 20 turns to the inch which is probable better then required. The tuning even with my set up is very fine and silky smooth. Far better then any varicap set up. Well worth trying n some form as it is surprising. Vari' caps' are not easy to find either in values or dimensions that are suitable. All of the sudden with a bit of creativity they are unnecessary and in fact now perhaps undesirable. PTO really does work that well.
don

Many 50s car radios were PTO. Philco (Ford to name another one) Some had very intricate tuning devices and would be far superior to what I did just using a slug tuned core which works but isn't really designed for constant changing
*********
interesting - I'm well acquainted with one Auto Shop (20 years) - which have a junk yard - I think I'll do some raiding!
**********
well - I did NOT do some raiding - their junk yard does not have cars that old!
and - they knew of none other in town - nor over at Dayton OH!

Jim - you DID get my big YES - on your RAmbler RAdio - didn't you?

Cliff

On the PTO

I have gotten a couple gear drives - right angles - one to one - also a flexible shaft (says it will go 90 degrees - in 6 inches - will it do 2 90 degree angles? - an S shape?)

What I am building is the ZL2PD PTO - setting it directly inside the front of the oscillator - then extending a marker out from the moving shaft through a slot in the front panel -to give an approximate frequency reading.

I will drive it with a 360 degree marked knob - one I have is 1 to 100.

The oscillator tube (Franklin Osc) will be outside the metal house wiring box (deep ceiling fixture size). I am building the PTO inside - so heat will be away from the coil (et al).

ZL2PD
"A Permeability Tuned Oscillator"
http://zl2pd.com/ptovfo.html

also picked up a '50s homebrew transmitter - 6146 final/pair of 6L6 plate modulators - some loose wires, etc

Merle C Rummel
W9LCE

Looks like an interesting project.

Again, a Collins PTO is pretty cheap on the internet and at hamfests, and has ALL of the parts already there, plus a linearizing compensation scheme. You could take that coil out and substitute or remove the windings and rewind.

Also, I'd have used a brass screw not a steel one, and also as I mentioned before 1/4-24 not 1/4-20.

Get a bit of brass rod from a metals supplier - most cities have one or more - which a foot or so is very inexpensive, and a tap and die for 1/4-24 and you can make ur own parts. Dunno but the big box stores might sell it too...

I'd use Delrin or nylon to make the "nuts" that slide on the screw threads, so that you can make a small assembly using two nuts that can be slid on the mount to take up the screw slop - spring loaded on one of the two would be neat. The advantage is that there will be no effect on the inductance of the coil.

For tuning you might want to use a multi-turn pot indicator? That way you might be able to reset to a freq... a single turn to sweep the entire range seems too coarse? (if that was the idea?) Alternately, lashing up a turns counter is an idea.

I did not see a graph showing linearity??

_-_-

One thing to consider about fastener threading is that if it was custom made on a lathe, one could tighten up the tolerances. Most garden variety fasteners are made with a reduced fit, they couldn't be threaded together if they were machined on a 1:1 ratio, so the fit is usually reduced to somewhere around 80-85% to allow for clearance due to foreign materials which may find their way into the threads. My experience is from a basic machine shop class many years ago, but a machinist could elaborate on the subject.

Phil

Getting a used old PTO seems the way to go, but building one yourself is interesting.
For me, its no problem about the dial scale, I just use a freq counter with an offset if needed, it displays the output frequency so there is no calibration needed, and you do not even care if its linear.
The problem I see with a simple setup is the slug moves in and out as you turn it unlike a cap.
I would think there would be little difference between a design that used a fixed coil and a vari cap and a fixed cap and a vari coil.
My vfo/lo's using B+W coil stock and normal variable caps seem very stable in themselves.
Even with average voltage regulation and variable loads on them, they move/drift very little.
My vfo/exciter is just as stable as the 32V3 I also use as an exciter.
The LO in my 2nd receiver is more stable then both of those as the load on the LO is steady.

From my tests, and with what I have access to, it would be much harder making a pto as stable as my vfo's are. Winding wire on slug tuned forms and have it stable is much harder for me to do then just using some very high Q B+W coil stock and a vari cap.

To take up the slack in a sliding assembly, even an acme threaded rod intended for this use, the usual method is to use two nuts "loaded" against each other, often one is held solid, the other is spring loaded. Although in some cases the second nut is just judiciously adjusted and set. The nuts on a steel acme rod are usually bronze so that they set in and are a very low slop fit... I suggested using delrin or nylon for this use.

Before that I suggested a spring and cam (as in some early table radios).

And before that I suggested the body of a micrometer, already nil slop and re-settable and very fine thread.

_-_-

what is Delrin? Nylon I'm familiar with - and had not previously considered using it - it might have less proximity effect on the oscillator coil.

based on the referenced approach by ZL2PD - where a block is screwed in and out - controlling a mounted bronze slug that is inserted into the oscillator coil - it extends out of the middle of the block between the threaded shaft or bolt and his "bamboo rod".

I have been considering getting a bronze bar - maybe 1/2"x3/8" some 2-3" long - cutting treads in (2) 1/4" holes (one toward one end for the threaded control bolt the other centered for the bronze slug) and sliding the bar on something on the other end (to prohibit it from twisting - per his "bamboo rod"). This could as easy be nylon or ?delrin (probably would need to get a larger piece). Either would have less loading effect on the oscillator coil.

He does mention using springs on the bamboo rod, to "improve the feel of the tuning arrangement".

I have obtained a Gates Turns Counter Dial (per your suggestion) - and a 90 degree gear mechanism [in order to place the oscillator box against the front panel of the transmitter (or separate oscillator - I have a damaged Heath HG-10 VFO) -to allow a marginal frequency reading by dial pointer - from the movement of the tuning mechanism].

Jim [WD5JKO] sent me his 1964 Rambler automobile radio - which has a very interesting approach on PTO tuning. I am studying it, to see how I might be able to use it - possibly in an extensive rebuild on a scrapped National NC57 receiver, but doubt if I can duplicate it elsewhere. The radio presents a dial reading by using a pointer that pivots on one end - and multiplies the actual movement. I had thot of that, but they are doing it. (Maybe even make the NC57 be "Amateur Band" with 1st and 2nd IFs.)

(Have you heard of those 12 volt "space charge tubes" used in auto radios in the 50's? - I saw a paper of a Ham receiver using those - I am looking at "Emergency/Disaster" equipment and operations - a totally 12v receiver would be different!)

That is somewhat future - right now, I am just an old man, trying to get "back on the air" - moving to this tiny city lot - and its middle of WINTER.

OK - I got a Collins PTO - where can I get a schematic on this? I've hunted -

It has two tubes - both 6AU6 - the one is a 6AU6W, the other is a 5749/6AU6 - probably a replacement. These may tell where the unit came from.

I've also gotten a couple LMOs - from SB401s - so I am getting some decent Oscillator sources (although my SB400 didn't hold stable enough to do PSK31). I did find a schematic for these.

But I still am going to go PTO - just for stability.
- maybe build my variant of the one by ZL2PD.

Maybe I will be stable enough to even do PSK31 - which is where THIS whole thing started.

I don't know

there is no metal drum - this was mentioned in the purchase (I do think there are things I don't know) - - no wonder I got it so "cheap". Well, it still may be better than what I have -

I have downloaded a couple papers about it - one showing several pictures of internals - and its operation - the other listing some statistics on 17 types of the PTO.

The covering seems to be a hard clothlike material - from what I have found out, this is likely an inside insulating layer, with the mechanism inside. I had already determined that I needed to mount it inside a cage or can - just to give it necessary isolation from nearby radiations.

The shaft is a 1/8th inch stub - from pictures, there was a gear head mounted on it, which I do not have.

According to the schematic I am missing the "plug" connections for the wires. I do have the "coax" shielded output wire (to P801), and a woven shield with three wires inside (to P802 - wires A, B, C - shield being D, E - yes the ad mentioned that an "oven wire' had been "cooked" - E, F) The one paper that I downloaded, mentioned that PTO 70E-15 used in the 51J did not have the thermostatic oven, so I guess I am without!

I do intend to mount this in a separated case and plug its output in the "crystal"/VFO plugs of one or other transmitters - as use or repairing requires. I will be building a small outside power supply for it - such as I have on the LM Meter.

I started building an exciter using the pto out of a T368.
It runs from 1.4 to 3.2 MHz or something like that.
I want it to do 80 and 40 meters, and the pto will cover both if I double and triple it.
I would have done a vfo, but would have to switch something to get the range, which would have added another switch.
Since the pto has been sitting around for 25 years or more, it was time to use it.
output to a 6AG7 into a 6146, I will also make provisions to be able to plug xtals in.

Not sure if the pto will run all the time and the 6AG7 and 6146 get keyed or it all gets keyed.
Frequency display is an issue, a direct pto tune is about 20 turns so any pre made dial scale is out.
A frequency counter on the output would work but would not be on all the time.
I use the srd-iq as a frequency reference and can just spot, but some sort of display on the vfo would be nice.
80 and 40 meters only, for now, may play around with modulating it at some point.
Front panel is drilled and power transformer holes on chassis drilled so far.

Sounds like a simple set up is getting over complicated. I have several here. Shouldn't take more then an hour or so to build. I have two solid states. A 20m TX on AM and a 80M receiver known as slugabsorb. It is a regen and it perm tuned. the absorb comes from the use of a pot and an absorption loop to keep the regen just below osc point. Both tune very nicely and smoothly. Take you coil form (mine are both same size BTW). Double freq in megs will give you the number of turns for the same approx freq on 40M (40M only)So 14 turns will get you in the park. Put a fixed cap across it to bring it where you want. IE 100 pf or 200pf or whatever you need then all tuning is by your slug. I always build a hartley and centre tap (and I do mean centre or middle) for solid state I like a FET For tubes a triode (Kiss principle) The rx I built on 80 does a great job and even SSB can be handled with ease. I have a slugged coil in my Hb VFO for my DX60B and have asked myself a thousand times why I am using a variable cap to tune it when I could be using the slug if i had mounted it in a spot where I could access it. Someday I hope I will correct that oversite on my part. I also have in the works a slug tuned 40m rig cathode modulated using a 12ax7 osc buffer and a 829B final in parallel. It is a rainy day project so it will get done when it gets done. You can discuss this for a year or build it in a hour. I vote for the build even if just the perm tuned osc by itself. Famous American race engine builder Smokey Yunick once said "One good experiment beats 1000 expert opinions." I have found it to be sound advice and I am confident so will you. Go fer it!
don

The AM Forum

THE AM BULLETIN BOARD => Technical Forum => Topic started by: W9LCE on December 23, 2014, 12:35:49 PM