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


It is currently Oct Mon 21, 2019 3:29 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 26 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Mon 10, 2019 12:18 am 
Member

Joined: Jul Sun 15, 2007 9:34 pm
Posts: 255
Location: PA
In a thread regarding the HQ 170 concerning the oscillator circuit used in HQ 140, 150, 160, 170, 180 and a few others, Manual Man stated:

"Sometimes it took in-circuit testing of several 6C4's before you got one where drift was minimal after 'appropriate' warm-up time which varied from receiver to receiver due to part aging/poor part quality/etc."

I found this to be true, and believe this issue is a product of either poor tube or component choice in the fairly standard circuit.

I loathe hacking any decent looking vintage set that otherwise works fine, so I'm looking into a different tube to use in that circuit to increase stability at the same injection level. Thus, it would have to be a 7 pin miniature based tube, to keep aesthetics original. Looking also to keep the circuit, but just adjusting component values to fit the operating parameters of the replacement type. Not to worried about increased filament draw, within reason.

Years ago I did this with an HQ-140X, and with new components in that circuit using a 6AB4, stability became excellent, and dial calibration held almost perfectly from tube to tube of the type. Emperically, 6AB4 is one answer to my question. (6C4 = 1/2 12AU7; 6AB4 = 1/2 12AT7)

Just wondering if any BA tinkerers on the forum have had good results using another type, like one of the 7 pin miniature based triodes used in VHF or UHF TV tuner circuits, or even a triode connected pentode like 6AU6 or 6AH6 for stability and repeatability.

Always looking to make a "better mousetrap" and keeping the BA receiver or transmitter stock in outward appearance.

All input is welcome !

-O.B.


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Mon 10, 2019 2:02 am 
Member

Joined: Dec Sun 14, 2008 3:33 pm
Posts: 865
Location: Tokyo
Tangential response to your question. I've built a lot of regens and regens are oscillators. I have found that low mu, high transconductance triodes, empirically, work the best. By best, I mean they go into oscillation more smoothly and reliably, and at lower plate voltage. If I had to choose from a 12AX7, 12AT7 and 12AU7 triode, my first choice would the 12AU7, followed by the 12AT7. I have made good regenerative detectors from the 12AX7, but it's easier to do with a 12AU7.

I've also noticed connecting a double triode in parallel makes for a better regenerative detector. The 6J6 works much better with both sections in parallel rather than singly.

The 6BE6 pentagrid, triode connected, is an excellent regenerative detector and I suppose my favorite. Triode connected, the 6BE6 mu is about the same as a 6C4/12AU7, but the transconductance is twice as large. In my experience, this makes it a better choice. Hallicrafters chose a triode connected 6SA7 for the SX-28 LO. IIRC, the SX-28 manual makes note of the high transconductance of this tube in triode connection. The 6SA7 and 6BE6 are similar in construction.

Again, all this is an empirical observation, made from building regens. It may not be entirely relevant for your superhet needs. While regenerative detectors are of course oscillators, they are oscillators that need to move from near onset, into oscillation, and back again, smoothly and without hysteresis. A superhet LO simply needs to go into oscillation. There is also the matter of different interelectrode capacitance: of no importance in a regen, but it could affect the calibration in a superhet.

Low mu and high transconductance: in other words, low plate resistance. Comparing the published Rp of some popular triodes:

12AX7/6DR4: 80K

12AT7/6AB4: 15K

12AU7/6C4: 6.5K

6BE6, triode connection: 2.7K (estimated)

6J6, both sections in parallel: 3.5K

6AK5, triode connection: 5.2K

6AU6, triode connection: 7.5K


I've made triode regens with all the above. My first choice is the 6BE6, triode connection (OK, actually I use the 6BY6 in triode mode; the 6BY6 is a $1 tube and I have lots of them; if anything, it works better than the 6BE6). The 12AU7, both sections in parallel, also works very well but given the price of that tube, I don't usually choose it. There are also some low mu, extremely high transconductance VHF/UHF triodes. They would seem to be excellent choices but I've never tried them.

So, to sum up, if I were you, I'd try a triode connected 6BE6 or 6BY6 in place of the 6C4.

One more thing: pentode regen detectors are a completely different thing.

Rob


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Mon 10, 2019 4:03 am 
Member

Joined: Jul Sun 15, 2007 9:34 pm
Posts: 255
Location: PA
Rob;

Thank you for your detailed response to my question. Low mu; high Gm; low Rp it shall be.

These seem to be similar to properties considered for an audio driver stage. I should do some reading in my fourth edition Radiotron Designer's Handbook on the subject.

A 6J6 with both sections in parallel was considered as I have a large number of these, but wasn't quite sure about using one. It seems to be a good second choice, behind triode connected 6BE6, which are also on hand.

The key problem with the use of 6C4 in the Hammarlund receivers is performance consistency among tubes of the same type, with drift an adjunct issue. They may test good on a TV-7 instrument, and not oscillate in the circuit; or work for some time and just cease oscillating while still testing good. That's why I used 6AB4 25 years ago as a stopgap fix, and it seemed to work. That receiver, though, is long gone.

I appreciate the sharing of your experiences with this community.

Take care.

-O.B.


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Mon 10, 2019 6:46 am 
Member

Joined: May Sun 13, 2007 5:22 am
Posts: 215
Location: United Kingdom
Back in the 1940s Canadian Marconi built a HF receiver using a 7 pin 9002 as the local oscillator. This is a vhf triode and provided good stability at the top end (30 MHz). Schematic is here: http://www.jproc.ca/marconi/csr5a_schematic.bmp


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Mon 10, 2019 11:17 am 
Member

Joined: Dec Sun 14, 2008 3:33 pm
Posts: 865
Location: Tokyo
OB, just an idea about using the 6BE6 as a triode. The usual practice is to tie grids 2/4 and 3 to the plate, using grid 1 as the control grid. I suppose it would be possible to use grids 2/4 and 3 as the functioning plate, and ground the actual physical plate. Why? Well, the 6BE6 plate surrounds all the inner grids and the cathode, so it could be used as a shield if grounded. Would that be beneficial? Perhaps. Exterior tube shields are often good practice but they do keep the heat in. This is something I have not yet tried.

If you have to buy a 6BE6, I'd go with the 6BY6 instead. They are similar, but the 6BY6 appears to be a more robust and consistent tube. And it can be purchased NOS for $1.

Rob


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Mon 10, 2019 11:52 am 
Moderator

Joined: Jan Thu 01, 1970 1:00 am
Posts: 27941
Location: Detroit, MI USA
Also note that leaving the oscillator tube filament powered all of the time while the set is off makes a huge difference in drift in that circuit. Hammarlund has also chosen to leave the first mixer 6BE6 filament powered all the time. They likely felt they couldn't do much more and left it at that.

While there are valid reasons not to, such as the possibility of cathode poisoning with no B+ applied, enough commercial designs did exactly that so it is a viable option that they chose. Even doing that did not fully eliminate the drift, probably due to other thermal changes in the components surrounding the 6C4 in normal operation.

I haven't tried it, but have often wondered whether adding a small cooling fan directed at that area of the chassis would help stability.

Having owned Hammarlund receivers in that series, and serviced others over the years, I have really never been all that impressed by them. While they are good for what they are, and a bit better than some of the competitors in their price range, I'd suggest looking at high end commercial and military receiver designs of the same era to see what differences might exist in that circuit to improve stability.

_________________
Dennis

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


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Mon 10, 2019 5:01 pm 
Moderator
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 23084
Location: Somers, CT
In some early Trio receivers they used a 6C4 for the local oscillator and eventually changed the design to a triode connected 6BE6.

Pete

_________________
Just because it can be done doesn't mean it should be done.


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Mon 10, 2019 7:03 pm 
Member

Joined: Apr Thu 20, 2017 2:35 pm
Posts: 107
Location: Bethany Ontario
Anyone ever tried using a paralleled 6J6 double triode in place of the 6C4? it could work and drop in depending on how things are wired at the tube socket. It has higher gain that could be advantageous, and draws a lot more current than the 6C4, but the power transformer should be able to handle this.

Al


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Tue 11, 2019 1:36 am 
Member

Joined: Dec Sun 14, 2008 3:33 pm
Posts: 865
Location: Tokyo
The June 1957 Radio TV News had an article about a 6J6 oscillator design. As you can see, it uses both sections in parallel.

The 6J6 seems to have been an extremely popular tube for homebrew projects in the 1950s. Shows up all the time in CQ and QST.

Rob

https://www.americanradiohistory.com/Ar ... 7-06-R.pdf


Attachments:
6J6 R-TV 6.57.png
6J6 R-TV 6.57.png [ 87.55 KiB | Viewed 1104 times ]
Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Tue 11, 2019 1:40 am 
Member

Joined: Dec Sun 14, 2008 3:33 pm
Posts: 865
Location: Tokyo
Another 6J6 oscillator, from QST, July 1955:


Attachments:
6J6 QST 7.55.png
6J6 QST 7.55.png [ 32.62 KiB | Viewed 1103 times ]
Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Tue 11, 2019 1:45 am 
Member

Joined: Dec Sun 14, 2008 3:33 pm
Posts: 865
Location: Tokyo
Another thought: how about a double triode like the 12AU7, one section for the oscillator, the other section as a cathode follower.

Rob


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Tue 11, 2019 5:05 pm 
Member

Joined: Jul Sun 15, 2007 9:34 pm
Posts: 255
Location: PA
Thanks to all for the points discussed on the topic. Given the tube socket and existing circuit, triode connected 6BE6 requires the least intrusion for the most benefit in the HQ 1XX oscillator application. The thought of using that tube for this application never entered my mind!

Tom Schlangen has published extensively regarding the use of triode connected pentodes, even providing tables with relative values of transconductance, plate resistance, and mu for a variety of both low level and output tubes. His work seems focused on their use in audio applications, rather than RF.

The latter is one reason I posted my question here.

As Pete Bertini mentioned, Trio-Kenwood made a running change to improve receivers in this regard. In such an instance, manufacturers generally are cost-conscious, looking for best performance improvement at least change cost.

Modifying for 6BE6 seems very simple, with little tweaking needed. Can't understand the guys griping (for decades) about the poor performance of 6C4 in that application, causing them to lose patience with those receivers. Why not expend less energy assessing the problem and attempt positive improvement?

-O.B.


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Sat 15, 2019 11:50 am 
Member
User avatar

Joined: Oct Fri 09, 2009 9:25 pm
Posts: 326
Location: Cleveland, Ohio
Could the 6C4 design in the HQ-170 be responsible for 10 to 15+ kHz startup drift on 40 meters only? ...or is this moreover a problem in that particular band switched switched stage/components? The other bands track fine and drift very little.

Todd
ka8gef

_________________
'Itoshiki oshieo idaki'
-Let us never lose the lessons we have learned.


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Sat 15, 2019 9:15 pm 
Member
User avatar

Joined: Jun Sun 19, 2011 2:31 pm
Posts: 6715
KA8GEF wrote:
Could the 6C4 design in the HQ-170 be responsible for 10 to 15+ kHz startup drift on 40 meters only? ...or is this moreover a problem in that particular band switched switched stage/components? The other bands track fine and drift very little.

Todd
ka8gef


That much drift on one band only is going to be from a component (or components) specific to that band. Probably one of the three fixed capacitors in the oscillator circuit for that band are at fault.

The only issue I have run into with having to hand select 6C4 tubes for this family of receivers is to avoid the problem of the oscillator rapidly jumping a few kilohertz in frequency.

Rodger WQ9E


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Sun 16, 2019 12:47 pm 
Member
User avatar

Joined: Oct Fri 09, 2009 9:25 pm
Posts: 326
Location: Cleveland, Ohio
Thanks Rodger. I certainly don't remember much of an issue going back to the '80's, when I first picked up this one. It looks like 3-4 caps could be involved in the 40m oscillator circuit.

I did find that a JAN/milspec 6C4 lasted longer vs. others that I seemed to replace every 1-2 years.

Thanks again.

Todd
ka8gef

_________________
'Itoshiki oshieo idaki'
-Let us never lose the lessons we have learned.


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Sun 16, 2019 4:04 pm 
Member
User avatar

Joined: Jun Sun 19, 2011 2:31 pm
Posts: 6715
It is kind of a pain but I have found through experience if there are several suspect caps it is easiest to carefully remove them all, refrigerate them for a few minutes, and then measure them on a capacitor test meter as they warm up to chart their capacitance change. Trying to test them in place is difficult because the undesired capacitance change is so small that it is difficult to measure with confidence with any other circuit changes taking place. I learned this lesson from hard experience working on a few transmitter VFOs.

Once you quickly graph the rate of capacitance change versus time as it warms up out of the refrigerator usually the problem component will become obvious.

Don't overlook the chance that the N750 may have changed its temperature compensation characteristic from desired.

Rodger WQ9E


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Mon 17, 2019 3:37 am 
Member
User avatar

Joined: Oct Fri 09, 2009 9:25 pm
Posts: 326
Location: Cleveland, Ohio
Great advice, Rodger...thanks.

We just had some damaging storms through the NE Ohio area and it took us 2 hours to make the 30 minute trip home from the Akron area this evening. The Cuyahoga Valley National Park had flooded areas throughout and numerous downed trees blocking several routes. I had to drive around barricades, under fallen trees and through fields, otherwise we would still be following endless detours.

50% of the newly planted corn in the Valley is under water.

After finally making it home, I flipped on the HQ-170A and initially thought that there was some type of solar storm in effect, only to realize that my dipole fell to the ground during the storms. It would have been a better day to work on the receiver or throw sinkers up into the maple trees and re-hang the antenna .....

Nevertheless, I hope that you had a great Fathers' Day.

Todd
ka8gef

_________________
'Itoshiki oshieo idaki'
-Let us never lose the lessons we have learned.


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jun Mon 17, 2019 4:15 am 
Member
User avatar

Joined: Jun Sun 19, 2011 2:31 pm
Posts: 6715
This has been a year of many storms here in Central IL also. Another squall line went through last night around 10 with a tornado warning briefly in effect around 10:30 but I don't think it touched down anywhere and nothing but a few branches down this time. I had a couple of trees down in late May during a heavy storm and a lot of fields still haven't been planted.

There have been a lot of power outages this Spring/Summer but I have a 40 KW diesel standby generator on order so pretty soon that won't be as big of an issue.

Rodger WQ9E


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jul Tue 02, 2019 1:50 am 
Member

Joined: Jul Thu 29, 2010 5:35 am
Posts: 1125
Location: Simonton, Texas
40 KW Roger? Do you have a hospital wing on your house?

_________________
Knowledge is useless unless you can share it. KA5OBI


Top
 Profile  
 
 Post subject: Re: Hammarlund Receivers Using 6C4 Oscillator
PostPosted: Jul Tue 02, 2019 2:22 am 
Member
User avatar

Joined: Jun Sun 19, 2011 2:31 pm
Posts: 6715
Koby wrote:
40 KW Roger? Do you have a hospital wing on your house?


Just a big house with 2 HVAC systems, 2 electric water heaters, a deep well pump, and the other usual stuff. A little smaller would work but the 40KW wasn't much more than a 30KW and it uses a more efficient direct injection turbo diesel so at typical load it is burning about a gallon per hour increasing to 2.5 GPH at its maximum 45 KW intermittent rating.

The only price that seems ridiculous is the automatic transfer switch. I could have gone with a cheaper version but I wanted one that will let the engine warm up for a few minutes before hitting it with a heavy load unlike the standard residential transfer switches that allow 10 to 20 seconds from cold start. I am installing an ASCO 300 series light commercial ATS which provides a full range of adjustable timers and I will probably go for about a 3 minute warm-up prior to transferring the load which will also give the refrigerator and freezer compressors time to equalize before restarting. This switch also does an in-phase transfer back to commercial when that power returns so you get a seamless transition coming off generator power.

And of course I have to power the Johnson Desk KW and other ham gear :)

The generator is being built and should be shipping soon and will be installed in the barn. The only hard work left is to trench in about 70 feet of conduit from the barn to the main power distribution point. 70 feet isn't really worth renting a trencher and my daughter does want to do conditioning work to stay in shape for soccer!

Rodger WQ9E


Attachments:
Mitsub 1.jpg
Mitsub 1.jpg [ 173.67 KiB | Viewed 582 times ]
Mitsub 2.jpg
Mitsub 2.jpg [ 347.89 KiB | Viewed 582 times ]
Top
 Profile  
 
Post New Topic Post Reply  [ 26 posts ]  Moderator: Sandy Go to page 1, 2  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: No registered users and 4 guests



Search for:
Jump to:  




























Privacy Policy :: Powered by phpBB