The Hazeltine FreModyne FM Receiver (Part2).

< back to part 1

Practical Design of the Fremodyne.
.

This spectrum display shows where a Fremodyne radiates, and also the importance of choosing the IF. The 4th and 5th harmonics are just outside the FM band. It is not clear in this picture as the station tuned to is on 107.7Mc, but due to the mixing occurring, there is actually radiation, modulated by the quench, at the received frequency. Note the local oscillator gives a clean signal compared to the super-regenerative detector. The 'filled in' curves are a result of being modulated at the quench frequency. Incidentally, a calibrated spectrum analyser gives an immediate indication of the bandwidth and operating frequency of a super-regenerative detector.

In January 1948 (ref. 5), there were 125 licencees to Hazeltine, but only five manufacturers in production at the time. Meck with their CX500 FM converter, Gilfillan with their 68F AM/FM mantel radio, and Howard with their 474 AM/FM mantel radio were three of these. I do not know if kitset manufacturers had to pay royalties or not, but if so, then Perco would be another.
Sensitivity is quoted at 200uV. Although weaker signals can be heard, the signal to noise ratio deteriorates. The circuit is designed to feed a typical audio amplifier consisting of a triode or pentode voltage amplifier and pentode output stage. (Typically, a 12SQ7 and 35L6 in an American mantel radio). Usually, the load presented to the Fremodyne will be the 500K volume pot. Obviously, the lower the value of volume control resistance, the lower the audio output by virtue of the 100K filter resistor. Although Hazeltine don't state it, Hi-Z headphones can be driven at a level similar to that of a crystal set.
Instructions for Fremodyne receivers suggest using a piece of wire 3 to 8 feet long as the aerial where the "power line" aerial is insufficient (i.e. the aerial terminal is connected to the power line via an isolating condenser). Failing that, a proper dipole is suggested in weaker signal areas. No hints are given as to what kind of transmission line should be used to suit the unbalanced input, or if 300 ohm ribbon can be used. In reality we can guess that in most "dipole" connections would have been with balanced line. This may seem crude as most Fremodynes have an unbalanced input, but it saves the cost of a matching transformer. As interference is not usually a problem in the FM band, the unbalancing of the aerial system does not cause any practical ill effects.
Various manufacturers of the Fremodyne performed slight alterations to the circuit but the super-regenerative IF stage remained essentially the same.

The question may now be asked, what if the FM RFC was simply replaced by the input tuned circuit?  The DC conditions on the grid would remain identical, and the VHF signal is still fed in. This would save one condenser and one RFC. Indeed this can be done, and about half the Fremodyne circuits took this approach. The circuit of such a receiver is here:


 

This circuit  was used by Meck as a stand alone converter, and is discussed later. This set incorporates dangerous features such as audio and aerial connections connected straight to the mains supply.


Applications of the Fremodyne.

The Fremodyne circuit was available in the form of an "FM Converter" either made commercially, such as the Meck CX500, or as a kitset such as those produced by Perco and Heathkit. The circuit was also used in low cost mantel radios and was essentially just tacked on to an existing AM design to produce a low cost AM/FM mantel radio. The Howard 474 (AM/FM) and 901-A (AM only) are an example of this. No evidence exists of the circuit being used in any form of console set, radiogram, TV sound receiver, or car radio (except for possible inclusion in the Australian "Telecond" referred to in Part 1). The Electronics Australia "Fremodyne Four" was an adaption of the design into a VHF communications receiver, for home construction. Kits were available for the EA design.
The Fremodyne received much attention in electronics publications at the time, being hailed as being a new way to receive FM at low cost, and that big things were in store for it. The way these articles were written seems to imply that the circuit was going to become almost as common as conventional FM circuits, and that service technicians were going to see a lot of it in the future. No doubt, Edwin Armstrong would have been displeased with the Fremodyne making its debut. Despite being the inventor of the super-regenerative receiver, he did not approve of it. or other "not true FM" receivers being used for FM reception. This was understandable since his FM service was intended for hi-fi reproduction with good interference immunity, and he did not want it receiving bad publicity as a result of inferior receivers being sold to the public.

Examples of Fremodyne Receivers.
Late 1947 saw a number of manufacturers jump on the Fremodyne bandwagon, only to withdraw from using the design a few years later. Here's some examples; links are given to full articles with circuit diagrams for these sets.

Meck FM Converter.


Inside the Meck. Valves are 7F8 and 6H6. The safety aspects were upgraded in later versions. Additionally, valves were changed to 14F8 and 35W4 to reduce current consumption and line cord resistor dissipation. One model also included a 12AT6 as and audio stage.
This is a Meck AM radio which uses the identical cabinet to the FM converters. This was undoubtedly a further cost cutting feature.


Here's one of my Meck FM converters.

A similar FM converter to the Meck which appears to be a clone was made by Telvar but housed in a wood and leatherette cabinet.
Article on the Meck FM converter here.


Osmor FM Tuner.

From the UK came this variation, the Osmor "Mini Magic".

This is a British version of the Fremodyne, sold in kit form by Osmor. Note the C1/L1 trap to reduce the super regen radiation. Also note the RF input is not tuned. This was acceptable as the three BBC VHF stations were always grouped close in frequency in each locality, thus image rejection was not an issue. No pics of the finished product are available; only this circuit.


Perco FM Tuner.

This was a kit version from Perco described in May 1948 "Radio News".

This circuit, like the Heathkit, uses a tapping on the oscillator coil to eliminate an RFC. The local oscillator in this model runs on the low side. Again, note the lethal output connections not isolated from the mains. The aerial input is via a separate primary coil which is the method shown in the Hazeltine patent. Incidentally, the pin numbering on the 35W4 heater is incorrect. Pins 3&6 must be swapped.

Article on the Perco here


Heathkit FM-1.

This Heathkit FM converter, model FM-1, came on the market in 1949. It was quickly removed when the full superhet FM receiver, the FM-2 replaced it in 1950. In today's Aussie dollar, this kit would be about $150. "Pulls in stations far beyond normal expectations"....well yes, if you compare it to a crystal set.
This circuit actually uses a power transformer to supply it. The use of a 12A6 as a rectifier is novel and would have been done as the 12A6 was a WW2 surplus item and thus would have been available at low cost. The other unusual aspect is the design of the RF coils; being what are large single turns rather than conventional 3 or 4 turn coils of smaller diameter. The receiver uses a 14F8.

Article on the Heathkit here


Howard 474 AM/FM mantel radio.

This is the well known Howard 474; a complete AM/FM mantel set. This was the most common of
the Fremodynes and examples of it still turn up. Like the Meck converter, the Howard 474 uses the same cabinet as an AM set of the same period; the 901-A. I acquired one of these sets in September 2004.

Article on the Howard 474 here.


Gilfillan 68F AM/FM mantel radio.
The Gilfillan 68F was another AM/FM mantel set, similar to the Howard 474.


Article on the Gilfillan 68F here


Sentinel 315-W and 315-I mantel radio.


Depending on a walnut or ivory cabinet, this was another AM/FM mantel radio. This has a  more elaborate cabinet and dial than the other sets.

Article on the Sentinel  here.


Olympic 7-532V and 7-532W.
Other AM/FM Fremodyne mantel sets were the Olympic models 7-532W (walnut), 7-532V (ivory), and 7-537W and 7-537V. Interestingly, both these models use octal valves throughout, except the Fremodyne circuit. In the 7-532, a 14F8 is used, while the the 7-537 uses a 12AT7.

Olympic 7-532V.

The 7-537 is obviously the cheaper of the two. It has no tone control and no external FM aerial connection. A power line aerial is permanently connected. Both models use the three RF choke design. The circuit also implies that the RF choke for the super regen detector is wound on the same coil former as the 21.75Mc/s IF coil. But we know from other circuits that to do this is not essential; it's merely convenient.

Article on the Olympic 7-532V here.


There was also another kit from the Rollin company (ref. 8) which used a super-regen superhet design, but it was not of the Fremodyne type. Instead, a 6BE6 converter was used to feed a 6J6 operating as a super regenerative detector tuned to 31Mc/s. An IF of that frequency would cause problems in areas where there is a station around 93Mc/s.
Home made Fremodyne (1996).

This is my home made Fremodyne which I'd built late 1996. If you wish to build your own, this is a good place to start. It's a very good performer that gets a lot of use. It has even been used in a moving car running off an inverter.

Article on home made Fremodyne here.


Magazine Projects.

Low Cost FM receiver.


This amateur constructed set was described in "Radio & Television News" for August 1951, as "Build This Low-Cost FM receiver". It is an attempt to simplify the original Hazeltine design, but unfortunately some performance is lost by so doing.

Article on the "Low-Cost FM Receiver" described here


FM and TV Sound Tuner.

This project from Radio & TV News, November 1957, is a VHF receiver based on a modified TV booster. The proviso was that the booster had at least two variable tuned circuits. By using a booster incorporating an "Inductuner" type of tuner, it was possible to have continuous tuning right across the VHF band.
No illustrations of the finished project were shown. UHF coverage was suggested simply by using a standard UHF TV converter to feed the modified VHF TV booster. The article is available at worldradiohistory.com


A One Tube 2-Meter Superhet.

From Radio & TV News, February 1949, this adaptation of the Fremodyne was for a 144Mc/s amateur band receiver. The IF was 38Mc/s.
The relatively low sensitivity of the Fremodyne would make it a poor choice as a communications receiver, except for local signals, and since this circuit does not include a proper coupling capacitor between the local oscillator and detector, sensitivity would be further reduced.
Suggestions were made in the article to make some of the resistors (R2 and R3) variable to optimise the performance. The article is available at worldradiohistory.com

The Electronics Australia Fremodyne Four.

A few notes regarding this well known Electronics Australia design first described in September 1962, then updated in March 1967. Constructors who have built this set in recent times, hoping to use it for FM reception may be put off the Fremodyne concept, if this is the only example of the circuit they have tried. It must be remembered this project was created when there were no FM transmissions in Australia, except for TV sound and two-way communication. TV sound deviates +/-50Kc/s as opposed to +/-75Kc/s for FM broadcasting. Incidentally, the audio amplifier is a good example of one suited to the Fremodyne.


Go to Part 3 of the Fremodyne >

Home