Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives
Support This Site: Contributors :: Advertise


It is currently Mar Sat 28, 2020 3:57 pm


All times are UTC [ DST ]





Post New Topic Post Reply  [ 30 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: High Fidelity AM Tuner Project - mulling over possibilities
PostPosted: Feb Wed 13, 2019 9:57 pm 
Member

Joined: Apr Tue 17, 2012 7:58 pm
Posts: 114
Location: Metamora MI, 48455
Hello all,

I'm currently toying with the idea of building a simple tube MW AM tuner for my Hi-Fi setup. Unlike in many areas of the country, we still have quite a few AM stations worth listening to in the Detroit area (CKWW, WKYO, WWJ for the news, WJR, etc) and in addition, my absolute favorite station, CFZM, comes in nearly as strong as a local station throughout the day.

I haven't put pen to paper and started drawing anything up yet and calculating component values, but my current plan would be to run the input from the antenna coil directly into a 6SG7 run as a mixer, with a 6J5 local oscillator, followed by a variable selectivity/bandwidth IF input transformer, coupled to a 6SG7 IF stage. The variable selectivity IF transformer seems like a nice option to have; leave it in the wider position during the day, or when listening to a station with wider audio bandwidth, yet I'd maintain the option of sharpening up the selectivity at night when the band is more crowded.

The 6SG7 IF would be followed by the IF output transformer. The IF output transformer I have on hand has a center tapped secondary and was designed for use with full wave diode detectors. Is there any real disadvantage to running say a 6H6 in the full wave configuration compared to a simple half-wave detector? Finally, I'd like to add a 6J5 cathode follower after the detector so I can drive a reasonable length of cable.

The power supply would be something fairly typical using a 5Y3 or 6AX5, etc.

Does this seem like a reasonable approach to decent sensitivity and selectivity on the AM broadcast band?

In particular, would there be any tremendous advantage in using the 6SG7/6J5 combination as opposed to say a 6SB7Y? A 6SB7Y has a conversion trans-conductance of about 950 micromhos, whereas a 6SG7 as a mixer should have a conversion trans-conductance of approximately 0.25*4100 micromhos (1025 micromhos), which is comparable to the 6SB7Y. But, the 6SG7 when run as a mixer has a much lower equivalent noise resistance of about 12 kilohms when compared to the 6SB7Y at 62 kilohms. With the noise levels in the broadcast band, would this difference in conversion noise even matter?

Would I be better served by a traditional RF amplifier stage ahead of a noisy converter like a 6SB7Y or 6SA7?

Any thoughts or critiques are welcome.


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Wed 13, 2019 10:19 pm 
Member
User avatar

Joined: Sep Mon 16, 2013 2:42 am
Posts: 3763
Location: Tucson, Arizona U.S.A.
Definitely include a tuned RF stage or you are likely to have image problems especially in the lower part of the band.

Add a 10KHz trap in the audio circuit to eliminate interstation whistle.

Use a large coupling capacitor between the cathode follower and the output jack so that it can also drive a solid state amplifier.

_________________
Jim Mueller


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Wed 13, 2019 10:39 pm 
Member

Joined: Apr Tue 17, 2012 7:58 pm
Posts: 114
Location: Metamora MI, 48455
Jim Mueller wrote:
Definitely include a tuned RF stage or you are likely to have image problems especially in the lower part of the band.

Add a 10KHz trap in the audio circuit to eliminate interstation whistle.

Use a large coupling capacitor between the cathode follower and the output jack so that it can also drive a solid state amplifier.


The 10 kHz trap is an excellent idea. I had forgotten how annoying the 10 kHz adjacent channel whistling is, especially at night. I don't have the issue on my car radio, but then it has poorer frequency response when switched to AM than a POTS telephone. I think I have a 10 kHz trap (somewhere in my boxes of random parts and junk) I swiped out of a shot Heathkit BC-1A. That ought to work.

A cap on the output of the cathode follower is a must, as I have some tube amps with no cap on the input to the first AF stage.


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Wed 13, 2019 10:46 pm 
Member

Joined: Apr Thu 14, 2016 8:25 pm
Posts: 467
Location: pensacola fl
There are several choices here. In the older rca tube manuals there is a trf hifi tuner but it is for strong stations only. You can get several tuners from the past such as the heathkit aj30 or pt1 tuner as well as eico hft94. These are wide banded am tuners. I have a pt1 and an eico hft94. I converted both to take an am stereo decoder and receive what I broadcast with my part15 am stereo transmitter. The eico can bet tuned a little wider bandwidth if you want full frequency response beyond fm stereo without circuit changes. The heathkit will need loading resistors to take it out to 20khz but that is not needed if you want to go out to 15khz. You just have to have propper pre emph at the trasnmitter and they both can offer very good results. Both have a wideband mode. There are other receivers such as fisher tuners and scott tuners that also had wide band settings as well. I hope you find one and restore it and enjoy the fruits of your labor. The two that I have each have a tuned rf stage. The heathkit has 2 i-f stages and this has more sensitivity.The heathkit aj30 is the same as the pt1 internally just different styling. heathkit had the bc1 that can be widebanded by adding a resistor. Both of them have other models as well that had wideband as well. You can find these on ebay. These all have a trap for 10 khz.


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Wed 13, 2019 10:52 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1489
Location: Jackson, TN
Have you considered a double tuned crystal receiver for picking up locals?

These can provide superb fidelity with tunable sensitivity and selectivity. (via adjustment of coupling between the antenna and detector tanks)

If desired, a tube preamp stage could be added to boost the detector signal, although it may well not be needed.

Tim


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Wed 13, 2019 10:58 pm 
Member
User avatar

Joined: Feb Sun 02, 2014 9:13 pm
Posts: 2117
Location: Roanoke, VA
Bear in mind that US AM stations are limited to the NRSC 10 kHz frequency response if they are analog, and 5 kHz if they are also transmitting AM digital. Most analog AM stations also use audio high frequency boost to improve the sound on the typical narrow-band AM receivers of recent decades. There isn't really anything to listen to on a high-fidelity AM receiver, as the days of AM stations approaching high-fidelity response (some with 12 to 15 kHz response) have been gone for decades.

Dale H. Cook, Radio Contract Engineer, Roanoke/Lynchburg, VA
https://plymouthcolony.net/starcityeng/index.html


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Wed 13, 2019 11:22 pm 
Member

Joined: Apr Tue 17, 2012 7:58 pm
Posts: 114
Location: Metamora MI, 48455
Dale H. Cook wrote:
Bear in mind that US AM stations are limited to the NRSC 10 kHz frequency response if they are analog, and 5 kHz if they are also transmitting AM digital. Most analog AM stations also use audio high frequency boost to improve the sound on the typical narrow-band AM receivers of recent decades. There isn't really anything to listen to on a high-fidelity AM receiver, as the days of AM stations approaching high-fidelity response (some with 12 to 15 kHz response) have been gone for decades.

Dale H. Cook, Radio Contract Engineer, Roanoke/Lynchburg, VA
https://plymouthcolony.net/starcityeng/index.html


I guess I should clarify a bit. I consider the 10 kHz theoretical limit for any station to be "High Fidelity" when compared to the typical frequency response of a cheap AA5 or the typical car radio, which often top out around ~3 kHz or 4 kHz, or worse.

CFZM and CKWW sound great on a higher quality tuner or radio. Actually, I had the pleasure of listening to a Zenith 12-A-58 at a fellow collector's home with CFZM on. It sounded *almost* as nice as a typical FM station, which is a testament both to the Zenith and to what the folks at CFZM are doing. CKWW sounded similar. When I'm in northern Michigan on vacation, I listen to WDBC and WQXO, both of which have better than average audio quality as well. Most stations do artificially restrict their bandwidth though, or use the IBOC garbage. All that said, I still feel justified in putting in the effort into building a higher quality AM tuner. Perhaps not "High Fidelity" in the strictest sense, but offering somewhat wider bandwidth with the option to sharpen it up using the switchable IF as I alluded to earlier.


Last edited by benman94 on Feb Thu 14, 2019 3:06 pm, edited 2 times in total.

Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Thu 14, 2019 3:01 am 
Member

Joined: Dec Sun 14, 2008 3:33 pm
Posts: 929
Location: Tokyo
>Would I be better served by a traditional RF amplifier stage ahead of a noisy converter like a 6SB7Y or 6SA7?<

Pentagrid converter noise is well below existing MW band noise, it's not a concern. However, an RF amp, as has already been noted, improves image rejection. It provides another stage for AGC. The increased front end selectivity may improve protection from cross-modulation. And it also helps with inefficient antennas. Speaking of antennas, have you tried a loop? They have to be tuned and turned, but they worked great on MW.

I would include a RF attenuator and/or manual gain control in the design to deal with very strong signals.

I think the choice of detector is very important. I've had good results with a cathode follower and a halfwave voltage doubler combination. Similar to an infinite impedance detector, but the doubler has the advantage of providing AGC voltage. The Heathkit BC-1 AM tuner used a voltage doubler detector, btw, but without a cathode follower.

Rob


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Thu 14, 2019 3:09 am 
Moderator

Joined: Jan Thu 01, 1970 1:00 am
Posts: 28414
Location: Detroit, MI USA
I agree that adding a tuned RF stage, with a 3-gang variable capacitor is going to be the best choice.

I'm totally amazed you can even pick up CKWW as far north as Metamora, as I live less than 10 miles from the station and can't reliably pick it up clearly here, not clearly enough to listen to. I have seen only one or two radios capable of receiving it well in my location. It's only a 500 watt transmitter and I'm apparently smack in the middle of one of the nulls of their pattern.

Why not just find an old Meissner with tuned RF stage, or modify a good quality postwar radio that doesn't have any collector interest or value to use as a tuner, something like the Detrola 572 comes to mind.

_________________
Dennis

Experience is what you gain when the results aren't what you were expecting.


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Thu 14, 2019 5:13 am 
Member

Joined: Apr Tue 17, 2012 7:58 pm
Posts: 114
Location: Metamora MI, 48455
Mr. Detrola wrote:
I agree that adding a tuned RF stage, with a 3-gang variable capacitor is going to be the best choice.

I'm totally amazed you can even pick up CKWW as far north as Metamora, as I live less than 10 miles from the station and can't reliably pick it up clearly here, not clearly enough to listen to. I have seen only one or two radios capable of receiving it well in my location. It's only a 500 watt transmitter and I'm apparently smack in the middle of one of the nulls of their pattern.

Why not just find an old Meissner with tuned RF stage, or modify a good quality postwar radio that doesn't have any collector interest or value to use as a tuner, something like the Detrola 572 comes to mind.


CKWW comes in fine for me, but then I'm in a rural area. The background noise is noticeably lower out here, and creeps up as I drive into the city each day. By the time I reach Rochester, CKWW is bordering on unlistenable, and by the time I hit 8 Mile and 75 there is nothing but hash.

There are spots in the Metro-Detroit area where WWJ is swamped by noise and they're broadcasting at 50 kW if I remember correctly.

As to why I want to homebrew it... why not? I've restored plenty of radios, moved onto television sets (I'm currently digging into a Transvision, and a CBS 205 color set is next on the list). I'm starting to tire of televisions though, and thought perhaps actually building something would be fun for a change. I've built a handful of Hi-Fi tube amplifiers, but never anything dealing with RF. It'll be a fun change of pace.


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Thu 14, 2019 6:26 am 
Member

Joined: Apr Thu 14, 2016 8:25 pm
Posts: 467
Location: pensacola fl
The heathkit aj30 and pt1 has an rf amp converter 2 i-f stages and a separate if amp infront of a n agc detector as well as a full wave detector which like a full wave power supply means smaller filtering and a better performing detector. It has a center tapped i-f transformer to feed it. The Eico is a good source of coils and a 3 gang tuning cap. You can fix it and hot rod/modify it to give you the performance you want at a good cost. I have one modified for cquam am stereo and 15khz audio as I said to play my transmitter through so U.S. standards do not apply here as it is part 15 carrier current. It has a single diode detector and I used it for agc only. The detector is modified so it does not give audio out just dc and it is intentionally slow so the agc doe not impact the level going to the cquam board as it can demodulate mono and stereo. That helped keep the cquam decoder in sync with the incoming carrier better. You may find if you get one you have a great supply of the essential r-f coils and tuning cap you need and do not have to go trying to get those that will track.


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Thu 14, 2019 1:58 pm 
Member
User avatar

Joined: Feb Sun 02, 2014 9:13 pm
Posts: 2117
Location: Roanoke, VA
benman94 wrote:
I consider the 10 kHz theoretical limit for any US station ...

There is nothing "theoretical" about it. It is a hard and fast limit mandated by FCC Rules and Regulations, and every AM broadcast processor sold in the US in recent decades implements it in hardware or firmware. In addition, all US AM stations are required by FCC Rules and Regulations to have annual emissions bandwidth measurements made with a spectrum analyzer to insure that they are within the NRSC limits.

Dale H. Cook, Radio Contract Engineer, Roanoke/Lynchburg, VA
https://plymouthcolony.net/starcityeng/index.html


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Thu 14, 2019 3:05 pm 
Member

Joined: Apr Tue 17, 2012 7:58 pm
Posts: 114
Location: Metamora MI, 48455
Dale H. Cook wrote:
benman94 wrote:
I consider the 10 kHz theoretical limit for any US station ...

There is nothing "theoretical" about it. It is a hard and fast limit mandated by FCC Rules and Regulations, and every AM broadcast processor sold in the US in recent decades implements it in hardware or firmware. In addition, all US AM stations are required by FCC Rules and Regulations to have annual emissions bandwidth measurements made with a spectrum analyzer to insure that they are within the NRSC limits.

Dale H. Cook, Radio Contract Engineer, Roanoke/Lynchburg, VA
https://plymouthcolony.net/starcityeng/index.html


It was my understanding that the Canucks never adopted the same limitations that the US did, which would render your entire point moot for me as I live in a border area. I've seen claims that CHML out of Hamilton tops out around 14 kHz, and this measurement was supposedly done as you suggest with a spectrum analyzer.


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Thu 14, 2019 3:49 pm 
Member

Joined: Nov Mon 06, 2017 2:35 pm
Posts: 120
Location: Texas, U.S.A.
1. Wouldn't a rotatable loop antenna, as Rob suggests, of the largest enclosed area practical, be desirable since it would provide directionality and largest signal voltage?
2. Wouldn't some sort of tuned circuit(s) prior to the mixer or RF amplifier be desirable, in order to limit the bandwidth and thus noise? Would it be desirable for the tuned circuit's bandwidth to be large enough to pass the desired modulation sidebands? Can a loop antenna be part of this circuit?
3. Wouldn't use of an infinite impedance detector be desirable, due to its negative feedback which reduces the distortion created by the detector? (ref. W. N. Weeden, "New Detector Circuit", Wireless World, no. 905, vol. XL, no. 1, Jan. 1st 1937, page 6). Is this what Rob is suggesting?
4. Wouldn't provision of an separate AVC channel be desirable, so it can be optimized for that purpose?
5. The tuned circuits, not amplifiers, provide selectivity. Amplifiers can be used to counteract tuned circuit losses.

WB5HDF


Last edited by infzqi on Feb Thu 14, 2019 4:15 pm, edited 2 times in total.

Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Thu 14, 2019 4:12 pm 
Member
User avatar

Joined: Feb Sun 02, 2014 9:13 pm
Posts: 2117
Location: Roanoke, VA
benman94 wrote:
It was my understanding that the Canucks never adopted the same limitations that the US did, which would render your entire point moot for me as I live in a border area. I've seen claims that CHML out of Hamilton tops out around 14 kHz, and this measurement was supposedly done as you suggest with a spectrum analyzer.

I wouldn't know about the CRTC regulations as I am not a licensed commercial radio operator in Canada. They could very well be different as the AM band is not as crowded in Canada. I live too far from Canada to measure their AM transmitters' audio fidelity.

Dale H. Cook, Radio Contract Engineer, Roanoke/Lynchburg, VA
https://plymouthcolony.net/starcityeng/index.html


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Thu 14, 2019 4:37 pm 
Member

Joined: Nov Mon 06, 2017 2:35 pm
Posts: 120
Location: Texas, U.S.A.
High fidelity is more than just the amplitude versus frequency characteristic of the receiver's output, in my opinion.

My opinion is that if the receiver is able to reproduce the modulating signal with minimum added distortion, it is high fidelity.

WB5HDF


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Thu 14, 2019 5:07 pm 
Member
User avatar

Joined: Feb Sun 02, 2014 9:13 pm
Posts: 2117
Location: Roanoke, VA
infzqi wrote:
My opinion is that if the receiver is able to reproduce the modulating signal with minimum added distortion, it is high fidelity.

That is indeed pertinent. AM broadcast transmitters in the US have to meet specified limits for envelope distortion, and modern solid state PDM/PWM transmitters (at least those of the major manufacturers) have significantly less envelope distortion than the old vacuum tube transmitters, so minimizing distortion in the receiver could be important.

Dale H. Cook, Radio Contract Engineer, Roanoke/Lynchburg, VA
https://plymouthcolony.net/starcityeng/index.html


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Thu 14, 2019 5:57 pm 
Member

Joined: Nov Mon 06, 2017 2:35 pm
Posts: 120
Location: Texas, U.S.A.
...
benman94 wrote:
Would I be better served by a traditional RF amplifier stage ahead of a noisy converter like a 6SB7Y or 6SA7?

According to Zepler:
1. Signal to noise is improved by use of sufficient RF amplification prior to a multigrid converter tube.
2. The impedance and bandwidth of the tuned circuits prior to the amplifier will contribute to the receiver noise.

It looks like if one of the design objectives is to minimize the noise level of the receiver, these need to be taken into account.

ref. E. E. Zepler, The Technique of Radio Design, New York: John Wiley and Sons, 1943, pages 155-161 https://archive.org/stream/Zepler_1943_ ... 3/mode/2up

Edit: Oh now I see your note about using the 6SG7 as mixer vs a multigrid converter tube. Would need tuned circuits prior to the 6SG7 to give sufficient image rejection. Would need to run noise calculations comparing the two, again see Zepler.

WB5HDF


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Thu 14, 2019 7:55 pm 
Member

Joined: Apr Tue 17, 2012 7:58 pm
Posts: 114
Location: Metamora MI, 48455
So after doing some hunting around on radiomuseum.org, it seems the 6SG7 mixer and 6J5 oscillator combination enjoyed a brief period of popularity around 1946-1947. A lot of the commercially manufactured radios using this setup didn't use a tuned RF stage out front, even on shortwave, but even a cursory examination of the setup shows that image rejection would be a nightmare with such an arrangement. Such a radio may well work fine in the day time, but at night with the crowded band, and thus a greater chance of having a powerful station 910 kHz away, many stations would be a cacophony of unintelligible nonsense.

So, I'm modifying my plans a bit. A tuned RF amplifier (6SG7, 6SD7, maybe a 717A), followed by a pentode mixer (6SG7) with 6J5 local oscilllator, followed by a pentode RF stage (6SK7), followed by the detector and cathode follower.

Conventional wisdom seems to be that the RF stage should produce just enough gain to get the signal above the noise of the mixer and not anything more, so as to avoid overloading the front end. The 6SG7 would probably be a good compromise on gain vs noise here, the 6SD7 would also be adequate. A sharp cutoff tube would give the best noise figure, but at the expense of running wide open with no AVC applied to it.

Between the gain of the RF stage, and the gain of the mixer (especially if using the 6SG7), the signal at the input to the IF will be well above the noise introduced even by a lowly 6SK7. The 6SK7 is remote cutoff, whereas the 6SG7 is semi-remote. The advantage of the higher gm of the 6SG7 would only be valid for low AVC voltages. A 6SK7 should be perfectly adequate for the IF spot with no serious penalty in terms of noise.

Detectors have been mentioned, and the full wave detector I am considering is noted for very low distortion levels, and for not seriously altering the Q of the preceding tuned circuit in any appreciable way. I'll also look at the infinite impedance detector though as it promises similar performance with just a single triode.

The coils I have on hand are J.W. Millers with iron cores, so they should offer relatively high Q, and therefore adequate selectivity.

I like the idea of a large loop antenna being used the RF input circuit, simplifying that stage a bit; using a three section cap, one section would tune the loop itself, the other the secondary of the RF transformer, and the third section the 6J5 oscillator.


Top
 Profile  
 
 Post subject: Re: High Fidelity AM Tuner Project - mulling over possibilit
PostPosted: Feb Thu 14, 2019 8:31 pm 
Member

Joined: Nov Mon 06, 2017 2:35 pm
Posts: 120
Location: Texas, U.S.A.
...
I'll bet that it will be a very fine receiver!


Top
 Profile  
 
Post New Topic Post Reply  [ 30 posts ]  Go to page 1, 2  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: No registered users and 5 guests



Search for:
Jump to:  






























-->


Privacy Policy :: Powered by phpBB