Drift in an Old VFO

[ashart]

Here's an old-timer question:

Consider a VFO designed and built in the late '30s or early 40's, either ham-built or commercially-built for the ham market, and being used on an AM transmitter.

What was a generally-acceptable specification on frequency drift for AM operation?

Not what was desirable - not what was state-of-the-art - not what "woulda been nice," - not what "coulda been built by a smarter designer," but instead, what was used without often getting criticized by others in the QSO or awakening the local OO?

Did the Meissner Signal Shifter or any other commercial VFO of the day, have a drift spec?

I'll appreciate your ideas.  Tnx es 73.

al, w8vr   (www.w8vr.org)

[KM1H]

Ideally your VFO drifted in conjunction with the other guys receiver. Thats how life was back then. Round tables were a riot and sometimes spread over 5kc or more until everyone zerobeat again.

These days it easy to stabilize any VFO, even a 30's design.

Carl

[w3jn]

It would depend entirely on how cranky your QSO partner is, and how selective/drfity his receiver might be.

[Steve - K4HX]

something less than 1 kHz would be good. Less than 500 Hz and almost no one would notice.

[k4kyv]

Isn't there some kind of add-on synthesis circuit that renders an analogue VFO rock stable? A varactor diode is added to the original tuned circuit to form part of a PLL to lock the oscillator onto a band of uniformly spaced reference signals distributed in small, say 100~ steps. The original free running oscillator locks onto the nearest step, so it is tuned in the normal manner with the original tuning control, but does not drift from the locked frequency.

I am rather picky about carrier frequency, and try to stay as close to exact zero-beat as possible, at most +/- 300~ error. It is simply good operating practice, although I sometimes fail to pay attention when the station I am talking to has drifted off frequency. Zero-beat is something to be appreciated by anyone receiving with a synchronous detector, since it takes the BFO a finite time to lock onto an off-frequency carrier.

[WD5JKO]

Here's an old-timer question:
Consider a VFO designed and built in the late '30s or early 40's, either ham-built or commercially-built for the ham market, and being used on an AM transmitter.

   Al, maybe we can get Walt in here since he lived through that era. I suppose many receivers were regen back then, and the few superhets had fairly wideband IF selectivity. Remember the BC-610 internal VFO? Still some of the WW2 surplus was respectable in terms of drift, even by standards of the 1960's and 70's. The BC-223 with  high level MO was stable until the antenna swung in the wind since this was a MOPA 2 X 801 RF transmitter. The ART13 wasn't too bad either with an 837 VFO. The RCA Ar88 was a late 30's design, and probably regarded very highly in the day. Then don't forget the ARC-5 command sets. Those receivers can tune in SSB today, and not drift very much. The transmitter like the BC-458 was used by Central electronics as the basis of the CE VFO for the 20A and 10B for use on SSB transmission. Pretty good engineering for the day.

Jim
WD5JKO

[vincent]

You could try a Huff and Puff stabilizer:

http://www.hanssummers.com/huffpuff/library.html

[KA0HCP]

I checked the manual for the Meissner Signal Shifter EX and it has no spec for drift.  Actually, it has no specs at all.  It refers the user to 'standard technical references" for alignment.  hah.   

A related question about VFO Drift Stabilizers came up on the Zed late last night, which I answered.  Here are links to two Huff-and-Puff boards:

here is a link to his web page. I have no idea if he is still selling them:
http://home.comcast.net/~k4dpk/pep_adapter.htm

Here is another vendor of a "Huff and Puff VFO Stabilizer"
http://www.cumbriadesigns.co.uk/index.shtml

[WD5JKO]

Just remember, Huff and puffs only work with a VFO that stays oscillating. This won't work with the Johnson 122, or heath VF1. The Cumbria will also work with a dual frequency VFO where the VFO is shifted between receive and transmit. This is all a tangent to the original question posted by Al concerning what was acceptable in the 1930's and 40's. Stabilizing a VFO today seems to be a subject for another thread...in my opinion...

Jim

[vincent]

You're right! But I was focused on what K4KYV has written on the first part of its response.
I apologize.

[KM1H]

I wonder what sort of phase noise performance that VFO Stabilizer has??

By the mid-late 30's most except the poverty types were using superhets but crystal control was still king. The Meissner Signal Shifter was the first mass marketed VFO and improved after WW2 with the EX model.  I have 2 of the early and a couple of the EX. All can be made pretty stable at least thru 20M.

Carl

[Steve - K4HX]

The original poster didn't ask for technical disserations and circuits. He asked the following.


What was a generally-acceptable specification on frequency drift for AM operation?



Please answer that and save the other stuff for another thread.

[Jim, W5JO]

A friend and I checked several vfo models in our spare time using the heath, Johnson and a national.  If they work well the drift usually was under 1 kc.  The instructions for the heath said to heatsink the TC caps to prevent damage.

If the builder did a good job and mounted everything well, they were pretty good.  Receivers built after about 1955 were much the same.  Manufacturers had an eye to the future of ssb and even national issued a bulletin for the NC 183d recommending replacement of the silver micas in the osc. Circuit with air variables.

The best of all began with the Collins pto.  I believe it it speced at about .5 kc. Over 1 hour following 15 mins. of warming, maybe better.  A Collins guru can chime in here.

[k4kyv]

This is all a tangent to the original question posted by Al concerning what was acceptable in the 1930's and 40's.

The original poster didn't ask for technical disserations and circuits. He asked the following.

What was a generally-acceptable specification on frequency drift for AM operation?

Please answer that and save the other stuff for another thread.

Purely academic, but I would venture to say there were few or no "acceptable standards" for frequency stability and drift in the 30s and 40s. Very few hams even used VFOs; nearly all transmitters were crystal controlled. One would call CQ and tune from one end of the band to the other for a response. Very rarely did the replying station come back on frequency.  If someone did use a VFO, even if it was 5 or 10 kc/s off frequency, it would still be "acceptable" since it would be closer to the frequency than 90% of the crystal controlled rigs.  Stability for AM would meet the standard of that era if it didn't drift outside the passband of the receiver over the duration of a transmission.

SSB was deemed impractical for amateur use because it was somehow believed that the frequency stability requirement was beyond the technical capability of amateur equipment. This, despite the fact that some receivers had excellent stability even when compared to present-day transceivers. One example is the 1935 era National HRO (fully warmed up). If a receiver vfo could be that stable, there is no reason why a similar VFO couldn't be used to drive a transmitter. The transmitter could also run crystal  control and maintain excellent stability. In 1933 and 1934, there appeared a series of four articles in R/9 magazine that culminated in detailed instructions for building a SSB exciter for the top end of 75m. But due to the perceived stability problem, SSB didn't catch on until after WWII.

An academic question from a historical perspective, but irrelevant to present day operation. Very few of the old VFOs that amateurs actually used would be acceptable, by to-day's standards, to put on the air... even though as cited above, excellent frequency stability was demonstratively achievable.

One example of VFOs of that era would be the built-in self-excited oscillator in a BC-610. No doubt perfectly acceptable for 1940, but by to-day's standards, closer to Timtron's SBE.

[ashart]

K4kyv's comments generated new thoughts and a new question.

I'm hoping to put my restored old AM rig on some of today's AM nets. 

The rig (see www.w8vr.org) uses a modified ARC-5 VFO.  From a cold start, the drift is about 400 hz in 2 hours.  After warmup, the drift is about 150 hz. in 2 hours, but I lazily didn't check the time-linearity of the drift.

I assume that these numbers are at least as good as, and because of several mods, probably a bit better than, those of a typical ARC-5.

The numbers intuitively seem quite acceptable to me, and were perfecly fine when I last used the rig in 1960, but that was 50 years ago.  Now though, I'm uncertain because of my own inexperience in operation on today's AM nets.  My uncertainty is compounded by the thought of using the 80M Arc-5 on 20M where the drift will be times 4, thus leading to my new question:

Will the performance of a typical ARC-5 VFO be accepted on today's AM nets?

And incidentally, while I appreciate the thoughts behind the many ideas offered, all I'm trying to do is to learn whether the rig as originally built will be acceptable on the bands.  If so, I have no interest in building any wonderful new VFOs.  Besides, many of the new circuits offered use something called transistors.  Did they ever become popular?

Tnx to all.

-al, w8vr



 

[W7TFO]

I have no interest in building any wonderful new VFOs.  Besides, many of the new circuits offered use something called transistors.  Did they ever become popular?

Gotta love it!

73DG

[k4kyv]

The ARC-5 and similar Command Sets is an example of the very few pre-WWII designs that would be acceptable by to-day's standards. More typical would be the quality of the internal VFO in the BC-375 or BC-610. Unless you have something approaching the state of perfection achieved with the ARC-5, my advice, if your goal is to put together a working, 100% authentic, vintage station, would be to do like most of the hams of the mid to late 1930s did: stick with crystal control.

If you want real authentic vintage crystal control, find some of the old round Bliley or Valpey crystals that fit into 5-pin tube sockets. Very often, the original crystal (which may lie  completely outside a ham band) is directly interchangeable with a quartz plate removed from a large rectangular WWII surplus crystal, BC-610 crystal, (although those may be valuable in and of themselves), or even a FT-243 holder.

But I would not recommend, as described in an ER article some time ago, to gut the insides of a round holder, hollow out the bakelite case and stuff it with a FT-243 or HC6 type crystal. Find a real quartz plate for the frequency or frequencies you wish to operate and mount it in the original holder.

Otherwise, adapt the "vintage" transmitter to accept a modern stable VFO or DDS generator.

[KC2ZFA]

In the 1942 article http://n4trb.com/AmateurRadio/RCA_Ham_Tips/issues/rcahamtips0501.pdf subtitled "exciter drifts less than 20 cycles per hour after 40-minute warm-up period" there's the graph attached below. I guess that gives an idea of the kind of vfo stability that one was prompted to shoot for back in those days.

Peter

[KM1H]

Will the performance of a typical ARC-5 VFO be accepted on today's AM nets?

Well, they DO accept DX-100's and Globe Chumps so Id say there are still no standards

[WD5JKO]

In the 1942 article http://n4trb.com/AmateurRadio/RCA_Ham_Tips/issues/rcahamtips0501.pdf subtitled "exciter drifts less than 20 cycles per hour after 40-minute warm-up period" there's the graph attached below. I guess that gives an idea of the kind of vfo stability that one was prompted to shoot for back in those days.

Peter

  Now take that "stable" VFO, and with frequency multipliers, put it on 20-15-10 meters. 

Jim
JKO

[k4kyv]

By 1940 or 41, TV was about to take off.  Hams were just beginning to develop VFOs (in those days called ECOs- electron coupled oscillators) that would rival crystal control.  Then the War came along, took hams off the air, stalled TV development, and the war industry developed the technology for us (TV, via RADAR). Upon return to the air, hams began in earnest working on stable VFOs.  Commercial manufacturers got a jump in the action with war technology through the Collins PTO, as in the ART-13 master oscillator. Then the 75A series receivers came out, along with the 310 series exciters.  WW2 improvements in technology were also reflected in the conventional variable-C VFOs like the EF Johnson 122, Heathkit VF-1, and numerous homebrew designs, which offered "satisfactory" stability.  Commercial TV, which for many years had been "just round the corner" finally began to rapidly develop.

The PTO by Collins WAS stable enough to multiply from 160m all the way to 10m.

[WU2D]

I agree that +/-500 Hz is certainly acceptable in an AM net situation. More and more folks are using a SDR and a waterfall display so it is easy to report on my drifty signal.
The ARC-5s and PTO rigs like the ART-13 easily beat the popular Ham VFOS made for the rigs of the 40's and 50's. Mike WU2D