Warning: Creating default object from empty value in /homepages/11/d132647312/htdocs/Amfone/mkportal/include/SMF/smf_out.php on line 47
Introduction, and more Gonset G50 Stuff.




 
The AM Forum
September 19, 2018, 01:49:30 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: Introduction, and more Gonset G50 Stuff.  (Read 918 times)
0 Members and 1 Guest are viewing this topic.
KD1SH
Member

Offline Offline

Posts: 23



« on: June 23, 2018, 03:24:03 PM »

Greetings, everyone,
     New member, here, so a little intro before I get to my questions:  Name is Bill, licensed since 1993 but active in radio and electronics in general since I was a kid.  Worked in electronics all my life, and currently employed as an Electronics Engineering Technician in an R&D lab.  Though the job doesn't specifically involve RF/radio work, I consider myself reasonably knowledgeable in the field.  Not nearly - by any stretch - as knowledgeable as some of the members posting here, so I'm really looking forward to learning from you guys.
    New to AM - always spent most of my time on VHF/UHF sideband, with a bit of HF here and there.  But, I was recently infected with the AM bug; picked up an old Lafayette HA-460 at a hamfest, re-tubed and re-capped it, and liked 6 meter AM so much I went crazy at Nearfest and picked up a Heath Seneca, Johnson Ranger with 6N2, Ameco TX-62, and Gonset G50.  Well, if you're going to jump into something, might was well jump in with both feet.  The Ameco and the Gonset are both re-capped and working fine, with only a few small issues.
    Which brings me to my question:  in the Gonset literature, they say it uses "integral speech clipping".  I'm not seeing anything that looks like what I'd expect for clipping circuitry, though - no diodes, feedback paths, or any such.  I do note, though, that the 6L6's, in a parallel single ended class A arrangement, have their grids tied together with 47 ohm resistors, so that the operating conditions would appear to be slightly different for each tube.  Wondering why they did this, and what effect it actually has on the modulator's operation. 

Thanks
73 -
Logged
PA0NVD
Member

Offline Offline

Posts: 285


Nico and Chappie (Chappie is the dog...)


« Reply #1 on: June 23, 2018, 03:54:48 PM »

Hi Bill
Welcome, I am sure you will feel at home here, just as I do.
The 47 Ohms in the Gonset are just stoppers for parasitic oscillation and do not affect the bias. I think that, because the modulation uses a Heising choke, (which is close to saturation ) the clipping occurs due to that saturation and sales used this design flaw to a positive phrase like :clipping".
If you want to improve the modulation quality, the best thing is to start finding a balanced output transformer for approx 20 - 40 Watts. One side to the 6L6 modulators and the other side to the 6146. Than the DC currents of the 6L6 and the 4146  will almost cancel and you modulation quality improves. In the tread of the gonset you will find more very nice mods to add negative clipping etc.
Logged
KD1SH
Member

Offline Offline

Posts: 23



« Reply #2 on: June 23, 2018, 06:51:31 PM »

   Thanks for the reply - the mod sounds interesting.  Trying to picture it in my head, but I'm thinking you mean to replace the Heising choke with a conventional modulation transformer, but instead of feeding the primary from a push-pull stage we'd run the output of the Gonset's single-ended modulator into the primary; one end to the 6L6 plates and the other end to B+, and the secondary with one end to the RF PA and the other end to B+?  This mod is mentioned in another G-50 thread, but it's quite old and the link to the schematic doesn't work anymore.
    When I first went on the local net with the rig, the consensus was pretty much that I had too much base.  Mic is a D104; not typically given to heavy base, so that surprised me.  At first I blamed the heavy base on my re-capping; I replaced the triple-section metal can, 30/30/30 uf, with three individual caps, 38 uf each, figuring (foolishly not checking the schematic first) that these were power supply filter caps, and more - within reason - is better.  But then I found out that these are the cathode bypasses in the audio section.  To my thinking, the original 30 uf caps are pretty huge for cathode bypass, and now I've gone and changed them to 38 uf (20%).  But, I tried disconnecting C36, a 0.0068 uf disc across the Heising choke.  The only purpose I could see this cap having is to shunt the high freq's.  It made, from what I'm told, a great improvement; much sharper, brighter audio.  My voice characteristics are such that it tends benefit from more upper-mid's and highs than from the lower end.
    All in all, I love the rig; the receive audio is fine, and the VFO stability is great compared to my Lafayette HA-460.  The Lafayette - though I do like the little rig - doesn't just drift, it goes over the falls in a barrel.

73


Hi Bill
Welcome, I am sure you will feel at home here, just as I do.
The 47 Ohms in the Gonset are just stoppers for parasitic oscillation and do not affect the bias. I think that, because the modulation uses a Heising choke, (which is close to saturation ) the clipping occurs due to that saturation and sales used this design flaw to a positive phrase like :clipping".
If you want to improve the modulation quality, the best thing is to start finding a balanced output transformer for approx 20 - 40 Watts. One side to the 6L6 modulators and the other side to the 6146. Than the DC currents of the 6L6 and the 4146  will almost cancel and you modulation quality improves. In the tread of the gonset you will find more very nice mods to add negative clipping etc.
Logged
PA0NVD
Member

Offline Offline

Posts: 285


Nico and Chappie (Chappie is the dog...)


« Reply #3 on: June 23, 2018, 07:37:16 PM »

I don't think you got it right. You use the primary as the modulation choke. But the center tap to the +HT, one side to the anode of the 6L6 tubes in parallel and the other side of the primary to the 6146. The secondary you use just like it is used now, to drive the speaker
Doing so, the DC current of the 6146 is opposite to the DC current of the two 6L6 tubes. If they draw the same current, the DC magnetization is canceled just as it is in a balanced PA. But there is a difference, so a small DC magnetization will remain which is no problem. Doing so, you can modulate with less iron and less distortion because there is no core saturation.
In the other G50 thread running now, there are more nice mods to improve the modulation. Even a negative clipping mod.!!
Nico
Logged
KD1SH
Member

Offline Offline

Posts: 23



« Reply #4 on: June 23, 2018, 08:19:11 PM »

    Okay, thanks, Nico; now it makes more sense.  I was trying to figure out how, with my interpretation, the induced magnetic fields would cancel, how the impedance transformation would play out, or how the speaker would get driven.  Now it all comes together.  I wonder, now, why Gonset or Johnson or any of the others never thought of a center-tapped Heising choke?
    I do have a few mod transformers: one's a huge UTC VM-4 - way too big for this purpose - and a 40 watt Thordarson which might be just right but probably a bit bulky to find a home in the Gonset.  I've also got a 30 watt Stancor, which might just be the right physical size.  Only problem is, it was designed for a solid state modulator, so the insulation of the windings might not be sufficient for the high voltage.
Logged
KD1SH
Member

Offline Offline

Posts: 23



« Reply #5 on: June 23, 2018, 09:13:21 PM »

   And thinking on it now, my comments about modulation transformers are misguided.  Secondary impedance would be all wrong for driving a speaker.  Output transformer is what I'd need.  I'll be looking for one at the next hamfest.  Indeed, suggestions and schematics in the previous G-50 thread are awesome!
Logged
PA0NVD
Member

Offline Offline

Posts: 285


Nico and Chappie (Chappie is the dog...)


« Reply #6 on: June 23, 2018, 10:40:21 PM »

balanced output transformers are quite plentiful, eg for 2 x EL34, 2 x KT77 or KT88, etc. I even used a 6 Volts filament transformer with a 110 / 220 Volt primary winding with success. I didn't really notice a high end roll-off as I expected. It worked nice up to at least 5 - 6 kHz.
The nice thing is you can try a transformer outside the transmitter connected with long wires, so you don't have to take the old one out
Logged
KD1SH
Member

Offline Offline

Posts: 23



« Reply #7 on: June 24, 2018, 08:16:20 AM »

    Yeah, being able to work out my mods externally before doing any permanent stuff would be  great. The radio is in beautiful shape: the chassis shines like a mirror - I can literally see my reflection in it - and the Heising choke is riveted on.  I try to walk a line between being a "restoration purist" and an actual user.  It's nice to leave classic gear as original as possible, but I'm a user of the gear rather than a collector.
Logged
PA0NVD
Member

Offline Offline

Posts: 285


Nico and Chappie (Chappie is the dog...)


« Reply #8 on: June 24, 2018, 12:48:53 PM »

I am quite similar, I like to keep it original when possible. When the saturation is a problem for the audio quality, you can also lower the power and the 6L6 current a little too. 20% less power will keep the choke out of saturation and will hardly be noticed.
You can do that by increasing the G2 resistors for the 6146 and for the 6L6. But, also that is a change, and you have less power.
Operating with a little less load, a deeper dip, also lowers the power and current of the 6146, but you are prone to over-modulation. The balanced transformer mod is to my opinion the technical most clean modification. I do not know how the mod choke looks like. Hammond has nice output transformers of all kinds.
Logged
KD1SH
Member

Offline Offline

Posts: 23



« Reply #9 on: June 24, 2018, 04:09:17 PM »

    Well, at this point I'm planning on making the output transformer mod.  I've looked at the Hammond site, and like you say, great selection there.  I'll be headed to a hamfest in a few weeks, where I'll scout around for some used stuff before I spend bigger bucks on new Hammond parts.
    I've also considered going solid state with the rectifier - replacing the 5U4 - save a little filament heat and get a tad more power out.  Makes me just a bit uncomfortable, though: the 5U4 is easy on tired old circuitry - the ultimate soft-start - bringing up the DC nice and slow.  I'd put some resistance in series with the diodes to soften the surge a bit.  Maybe some NTC resistance - start off with higher resistance when cold and gradually decrease as the filter charging current and bleeder resistor current heats it up.
Logged
WA1LGQ
Contributing
Member
*
Offline Offline

Posts: 297



« Reply #10 on: June 26, 2018, 12:34:52 PM »

Hi Bill. Nice to see you on this board. I am located in East Hartford and I have heard you many times on the 6M net. My ant is just a halo and not up very high so I don't usually copy you as strong as some of the other guys. Maybe you have heard me, maybe not. I don't get on the net every week. I go to Ashford 2 or three times a week so maybe we will hook up from the mobile sometime. I am usually monitoring 146.52 and sometimes do 6M AM mobile. Anyway back to G50. I have one that needs a little work, but it is functional, I would like to do the audio mods that are talked about here with the push-pull transformer. I have already made some modifications to the rig, but it needs more. I would like to do something about the massive amount heat that builds up from the big power resistor. There has to be a better way. If anyone has come up with a good solution please chime in. Thanks!
Larry
Logged
WA2SQQ
Contributing
Member
*
Offline Offline

Posts: 548


« Reply #11 on: June 26, 2018, 02:56:40 PM »

When and what freq is the 6M AM net?
Logged
PA0NVD
Member

Offline Offline

Posts: 285


Nico and Chappie (Chappie is the dog...)


« Reply #12 on: June 26, 2018, 03:45:08 PM »

Hi Larry
Try to switch C1A off during reception and keep the resistor shorted or a low value. That will give you a choke input for receive and a lower voltage without generating heat. I do not know how much the voltage goes down, depends on the choke. But when you have a silicon rectifier, switching the C1 back into the circuit during transmission may give quite a surge. Than indeed you may need a limiting resistor is series with the rectifier.
Logged
KD1SH
Member

Offline Offline

Posts: 23



« Reply #13 on: June 26, 2018, 05:55:43 PM »

Hi, Larry,
    Good to see a local op on the board!  I've heard you a few times on the net, though sometimes it's tough because everyone's scattered all over the compass, and I wind up having to "split the difference" with beam headings.  Jay, W1VD, is pretty much due west, while Ken, KA1OXQ, is north, and Steve K1IIG, is southwest.  Jay is strong - can hear him pretty much anywhere the beam is pointing, but I can't hear Pete up in Vermont unless I'm aimed right at him.  Right now I'm just running whatever comes out the back of the rig - 30 watts in the case of the Gonset, but at some point I'll be running more power.  I've got a Heath SB-201 that I bought specifically for 6 meter conversion, but I haven't gotten around to that yet.  I've also got a nice home-brew 4CX400 6 meter amp that I bought at the Weak Signal Group convention, but I'm still building the power supply for that one.
    I'm on 6m sideband, 50.165, most nights of the week, with some friends, and sometimes on 50.418 AM.  I'm planning on being on the net tomorrow night - maybe talk to you then.


Hi Bill. Nice to see you on this board. I am located in East Hartford and I have heard you many times on the 6M net. My ant is just a halo and not up very high so I don't usually copy you as strong as some of the other guys. Maybe you have heard me, maybe not. I don't get on the net every week. I go to Ashford 2 or three times a week so maybe we will hook up from the mobile sometime. I am usually monitoring 146.52 and sometimes do 6M AM mobile. Anyway back to G50. I have one that needs a little work, but it is functional, I would like to do the audio mods that are talked about here with the push-pull transformer. I have already made some modifications to the rig, but it needs more. I would like to do something about the massive amount heat that builds up from the big power resistor. There has to be a better way. If anyone has come up with a good solution please chime in. Thanks!
Larry
Logged
KD1SH
Member

Offline Offline

Posts: 23



« Reply #14 on: June 26, 2018, 06:03:37 PM »

Net is Wednesday evenings on 50.400.  A somewhat informal net, starting sometime after 7:00 and running to around 8:00.  Connecticut, Massachusetts, one in Vermont.  Good guys - nice little net.

When and what freq is the 6M AM net?
Logged
KD1SH
Member

Offline Offline

Posts: 23



« Reply #15 on: June 28, 2018, 06:52:20 PM »

    So, I'm figuring on making the output transformer mod.  Checked the Hammond website; they've got a nice looking 30 watt balanced output transformer, their p/n 1645.  Primary impedance is 5K, with center tap, which I assume means 2.5K from center to each end.  Specs for the 6L6 output impedance seem to vary depending on where you look, but the consensus seems to be around 4K to 5K at 400 volts.  Figuring two 6L6's in parallel - half the impedance - then the 2.5K primary impedance of one side of the Hammond 1645's primary looks to be in the ballpark.  How important is the primary impedance, anyway, in this application?  I mean, the Heising choke isn't an impedance transforming device to begin with, so do we really care about the primary impedance in this mod?  I'm thinking basic electronics here: maximum power transfer only occurs when load impedance equals source impedance.  Hate to waste any of my modulator's power.
Logged
PA0NVD
Member

Offline Offline

Posts: 285


Nico and Chappie (Chappie is the dog...)


« Reply #16 on: June 28, 2018, 11:12:02 PM »

You are right indeed, in this case there is no impedance transformation and the specified impedance isn't too important. The output transformers are designed very broadband, so no worries about that either, they allow low modulation frequencies as well, sufficient inductance. Hammond makes very nice transformers with a very low spread inductance.
You can be sure that the frequency response is what the rest of the circuit is designed for, the transformer will not be a limiting factor
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.07 seconds with 18 queries.