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


It is currently Mar Wed 20, 2019 8:23 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 18 posts ] 
Author Message
 Post subject: S38D full wave bridge
PostPosted: Oct Thu 19, 2017 5:48 pm 
Member

Joined: Aug Sat 26, 2017 4:51 pm
Posts: 42
Location: Cincinnati, OH
Did the recap using Hayseed Hamfest kit and still some hum, even with 470 mfd cap, so decided to replace the half wave rectifier with a solid state full wave bridge. Had to delete return to ground bus at end of filament string of V3 to avoid shorting one leg of bridge. Tied pin 6 to neutral of 120VAC line. I am using polarized plug and had tied the ground bus to neutral before the modif. Now hums real bad and motorboats on signals. Since V3 is audio amp, may need fil to be at ground bus potential to avoid huim? Also Please address following questions:

1 Have you tried to use a full wave bridge successfully? Other ways to minimize hum?
2 My schematic does not match my circuit. I have John F Rider schematic for S38, early and revised. My S/N is 149763 which has noise limit switch but I don’t have that provision
3 BFO inoperative. Did not move feedback wire.
4 Don’t understand pilot light circuit. Did they use voltage drop on 35W4 rectifier filament? What is purpose of 330 ohm R19? I also have a 15 ohm in that circuit. An why tie pilot bulb to plate of rectifier and plate also tied to fil center tap. Goofy?
5 My filter resistors don’t match circuit.
6 other suggestions to improve?

Also replaced earphone connection to ¼” phone plug. Using phone/spkr switch as on/off relocated from vol control. Thought better to minimize AC hum proximity to AF.

I have added polarized plug and fuse

Receiver was sensitive except for 30 Mhz band and hum and images, of course. Calibration close too.

I am considering renovating a National NC183D but want to finish S38D first. Now there’s a real boat anchor. Must weigh 40 lb but with 2 RF, 3 IF and dual conversion must be hot receiver and has transformer with full wave and inductor filter, no AC/DC. Ever done one?

Thanks

David
AG8Z


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Oct Fri 20, 2017 12:17 am 
Member
User avatar

Joined: Jun Sun 19, 2011 2:31 pm
Posts: 6308
David,

You can find the schematic and manual for the S38D here: http://bama.edebris.com/manuals/hallicra/s38d/ If you are using a schematic for the S38 it is a far different circuit with a separate BFO tube.

The rectifier tube has a tapped filament for operating the pilot light and it is important to always replace this bulb if it goes out to avoid damage to the rectifier filament.
A properly operating S38D doesn't need a solid state full wave bridge or a bigger input filter capacitor.

As far as AC/DC "all American 5" type sets the S38 series are pretty well designed. I prefer the original S38 with its "real" BFO but the late versions (S38C and newer) have the safer isolated chassis setup although proper attention to safety is needed with all AC/DC type sets. Any of these, when properly operating, have minimal hum. The internal speaker has limited low frequency response as is typical of these types of radios and this reduces any hum issue. Using modern headphones with a vintage receiver will result in excessive hum unless you use a resistive attenuator to reduce the sensitivity of the phones.

AC/DC sets will generally have very different hum levels depending upon orientation of the AC plug.

You have so many modifications with your set at this point that it will be more difficult to troubleshoot. I presume you left the tube rectifier in place to keep the filament string voltage distribution as designed? I would back off on the modifications and get the set operating as intended and then modify if you wish. Receivers in the S38 series are good entry level receivers but will never be a great communications receivers. Any major improvements will require external equipment:

1. A good preselector will improve sensitivity and image rejection on the higher ranges, even better would be something like the Trio SM-5 "precon" also sold by Lafayette as their HE-73. This device operates as a preselector for 40 and 80 meters and also as a crystal controlled ham band converter on 20, 15, and 10 meters with output on the 80 meter range which provides better stability and a slower tuning rate.

2. An IF Q multiplier to improve selectivity and it will also work better than the built in regenerative BFO in the later model S-38 series.

3. An external audio filter to provide better selectivity and a notch filter.

Of course all of the above must be carefully connected to the receiver since it is an AC/DC type. Even with the later "safe" AC/DC series I would still permanently add an isolation transformer.

If you do all of the above you will end up with something that works far better on the ham bands than your typical S-38 but that kind of ruins the whole S-38 experience of "it takes a real master to use a S-38 to make contacts" :)

A hint for your NC-183D, expect all of the 47K screen resistors in the RF and IF sections to have drifted far higher than marked. Several of these operate pretty close to the rated 1/2 watt dissipation and any increase in resistance will cause them to heat even more. I believe National got a bad run of resistors during this time frame because I have otherwise never seen resistors drift so radically (and consistently) high with age. I have gone through 3 NC-183D receivers along with a Northern Radio diversity modified version and all had most of these 47K marked resistors in excess of 200K which kills gain and sensitivity.

Rodger WQ9E


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Oct Fri 20, 2017 1:05 am 
Member

Joined: Sep Thu 01, 2016 3:56 am
Posts: 249
Location: San Jose, Ca.
David, Reducing hum in the AA5 and other sets usually results from good functioning B+ filtering, good audio amp cathode bypassing, audio sensitive lead dressing and positioning. A leaky coupling cap can also do it (in this case C11A). In the AA5's specifically, the 1st audio (V3 in this case) filament needs to be grounded. I think that converting the rectifier to full wave bridge is going to be very messy and difficult, and really not necessary. When this set is operating properly, there is not much hum. Some times bad tubes cause hum. This results from a few issues with them. Swap out V3 and V4 to be sure.

Using a polarized plug is a good idea, so move the on-off switch to the other 117 leg.

Regards, Larry


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Oct Fri 20, 2017 2:06 pm 
Moderator
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 22657
Location: Somers, CT
Hi

I agree with Larry.

The engineers at Hallicrafters took extensive measures to ensure adequate RF bypassing, including
capacitors across the rectifier cathode and plate to minimize tunable hum problems. There are quite
a few RF bypass caps on the AC line and in the DC supply for that reason.

Once you go over to a full wave bridge, the radio chassis is always hot when referenced to either the
AC Neutral or Hot side of the line. That also means that the radio, which uses the power line as
a counterpoise when a long wire antenna is being used, may have tunable hum issues unless
extraordinary bypassing measures were used across each diode, etc. to ensure a stable RF path
that is not being modulated by the rectifiers being turned and on at the AC line frequency rate.

Pete

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


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Oct Fri 20, 2017 8:51 pm 
Member

Joined: Aug Sat 26, 2017 4:51 pm
Posts: 42
Location: Cincinnati, OH
Thanks to all for much info.
Any suggestions on getting BFO working?
Suspected V3 fil as source of hum, but cathode is grounded. May try adding cap from fil to gnd B- bus. Otherwise probably return to orig circuit.
Got the correct schematic now from BAMA, which clarifies alot.
Pilot does not light but measures a few ohms continuity??
Left the 35Z5 rect for the fil string. Anyone measure the fil and B+ currents? My rectifier will handle the fil load. May try DC on fils
That 1K resis in the B+ filter string measures 1500 ohms, may affect BFO?
Can you still get those rubber feet to isol cabinet from chassis? I have some I may try to adapt.
Heard about those 47k screen resistors in the NC183. This one has bad audio distortion, no BFO or Xtal filter function. Slop in main tuning knob and warped lid, some rust. Wish I cud find an original, matching National speaker.
I am sure now with advise on this forum I can enjoy my S38D like I did in the BSA SWL contest where I won a Heath Q-mult for my AR2. That's another project and my SB220 that is in the midst of R Measures mods, but I digress.....


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Oct Fri 20, 2017 11:41 pm 
Member
User avatar

Joined: Jun Sun 19, 2011 2:31 pm
Posts: 6308
Check the voltages on V2 (IF amp/BFO) tube to see if they are in the ballpark. If you have a spare 12SG7 substitute and see if this restore BFO action. The BFO is created by causing the V2 to oscillate via the gimmick capacitor feedback path C29/C30 from plate to grid. Take a look at these and see if they have been disturbed but before you make any adjustments to the existing take a photo for future reference. Increasing the coupling will help V2 break into oscillation but you don't want to overdo it; there should only be enough feedback to create reliable BFO action.

A word to the wise, spend some time reading up on things before you implement the Measures modifications to your SB-220. Suffice to say that article was published in QST AFTER the level of technical oversight had already seriously declined. Or to put it another way, voodoo has no place in electronic engineering.

I built my SB-220 in 1976 and it served many years in contest and DX duty before being retired to less active status in the mid 1990s; it is bone stock and works fine. The only thing close to a modification with mine is I used an external switch to select between my Drake C line and whichever Yaesu transceiver I used at the time and it had a center "neutral" position to disconnect the PTT line allowing the SB-220 to remain offline without turning the power switch off. Do be very careful with the 3-500Z tubes, never rock them getting them out of the socket. I believe all but the earliest 3-500Z tubes (and maybe some of the Chinese replicas) have a small insulating plate sitting beneath the tube envelope. This serves the dual purpose of providing better pin support lessening the chance of damaging a seal when inserting/removing the tube and it also allows better cooling of the base seals with the typical fan that is used instead of a blower and chimney.

A common issue with the Heathkit SB-220 is poor soldering of the filament choke connections and in particular the connections to the tube socket. Preferably a large iron like the old "American Beauty" is used for this but a good gun will also work but often you find that the builder used the only iron he had and the connections aren't properly soldered. This will cause a significant voltage drop and excess heating at the connection. Also make sure that the socket contacts for the filamentary cathode haven't lost tension because this will create very high localized heat which can damage the seal and will cause the solder to come out of the pin.

The SB-220 is one of the better Heathkit products and the only major oversight I have seen is the lack of a standby switch so that you don't have to power the amp on/off to bypass it. The filament transformer is properly sized so that it creates a reasonable soft start for the tube filaments but it is still desirable not to constantly cycle the power switch when changing bands and tuning up a tube type exciter.

If your SB-220 hasn't been in use for awhile the tubes may have developed gas. Although not optimal a generally good enough way to degas them is to put it in the CW/tune position (lower plate voltage) and short the PTT line to keep it in transmit letting the tubes cook at idle bias for at least 30 minutes. Optimal degassing occurs with the plates glowing red because these are the "gettering" element in the 3-500Z but generally some action will occur at the lower plate voltage but it takes awhile to get it hot enough. Going to full voltage (SSB position) too soon greatly increases the chance of a flashover. Resist the temptation to run CW in the high power position. The power supply and cooling system of the SB-220 hold up well at 1 KW CW input or 2 KW PEP with a reasonable duty cycle but don't push it in a high duty cycle mode. Heathkit rated it for RTTY at 1 KW input at a 50% duty cycle and it should do fine there. The only thing many of us did to help it when running in extended contest service was to remove the outer case which improves cooling.

Rodger WQ9E


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Mar Thu 22, 2018 8:46 pm 
Member

Joined: Aug Sat 26, 2017 4:51 pm
Posts: 42
Location: Cincinnati, OH
Wow, just now reviewed your comprehensive reply. Thanks Rodger.

Just now getting back to projects from remote job assignment.

My S38D is 5 tube version, so still need good schematic. Manual on BAMA site has page split right at CW/AM switch which obscures circuit info.

Returned to orig power supply circuit but 12SQ7 filament open, probably due to cathode to fil short.
Will repl tube and chk others this WE also for NC183 and resume speed. Will advise.

Thanks again.


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Mar Sat 24, 2018 8:56 pm 
Member

Joined: Aug Sat 26, 2017 4:51 pm
Posts: 42
Location: Cincinnati, OH
Replaced 12SQ7, all others tested OK (thanks Dan the Radio Man) and S38D plays real well with no hum. BFO regen not adequate for CW note but does add gain and selectivity due to regenerative effect.

Thought I saw solid state BFO oscillator circuit somewhere. May add that or just use on AM for SWL listening on foreign broadcast.

Still need schematic


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Mar Sat 24, 2018 9:30 pm 
Member
User avatar

Joined: Jun Sun 19, 2011 2:31 pm
Posts: 6308
Take a look at the S-38C schematic on the Nostalgia Air site, they have the Run 2 schematic version which I believe is very close electronically to the S-38D and should prove helpful until you find a better S38D schematic.

A solid state BFO would be better for CW usage or you could use a safely connected external Q multiplier also.

The original S-38 model had a separate BFO tube but later models were built using a regenerative IF stage which wasn't as effective. Shortly after I built my Knight Kit Star Roamer as a kid I added a solid state BFO from a Popular Electronics article. It didn't make it into an excellent receiver but it definitely made it better for CW.

Rodger WQ9E


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Mar Sun 25, 2018 2:16 am 
Member

Joined: Dec Sun 14, 2008 3:33 pm
Posts: 794
Location: Tokyo
>BFO regen not adequate for CW note but does add gain and selectivity due to regenerative effect<

This sounds like there isn't enough feedback from the 12SG7 plate to control grid. The IF amp isn't quite going into oscillation. While you haven't moved the feedback cable, someone else may have. Take a look here:

http://www.the-s38-guy.com/Page_63.html

If moving the cable doesn't help, try putting a 1K in parallel with the 270R cathode resistor. The lower resistance should push it into oscillation. If it is still not strong enough, try disconnecting the cathode bypass cap.

If none of this works, take a look at the S38E Mk2 schematic. The Mk2 did away with the cable-gimmick feedback, and instead simply inserted a resistor between the suppressor grid and ground to create a TGTP oscillator within the tube.

In my opinion, either of these methods are good enough for occasional CW or even SSB listening. However, very strong signals can block oscillation and I think the best remedy is simply to attenuate the signal between the antenna and the receiver. If this were my main rx 50 years ago, I'd build a separate BFO. Today you can build or buy very simple 455KHz BFOs using ceramic resonators. But since you have much better receivers, I wouldn't go beyond just getting it to work as originally intended.

Rodger mentioned the Star Roamer. One small but appreciated feature of that receiver's regenerative IF amp is the front panel control: you can set regeneration just before oscillation to improve selectivity, and you can set it well beyond oscillation onset to tame very strong CW signals.

Rob


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Mar Sun 25, 2018 3:29 am 
Member
User avatar

Joined: Jun Sun 19, 2011 2:31 pm
Posts: 6308
And to add to what Rob stated about the Star Roamer the variable regeneration control works even better with a separate BFO added so that you can adjust the built in control for maximum selectivity and then use the separate BFO for CW. If I had to choose today between a Star Roamer and one of the S-38 receivers for a basic "novice" ham receiver today I think I would go with the Star Roamer. The tuning mechanism works well in the Star Roamer and I think its front end coils are a little better because it goes a bit higher in frequency before images become objectionable. It is also easy and safe for adding accessories with its transformer operated power supply.

I have a couple of the follow-up Star Roamer II replacement receivers that I hope to turn into one good example. It is an interesting solid state replacement for the tube type original with a FET mixer (but no RF stage), two stages of IF, a separate product detector with BFO for CW/SSB and has different AGC time constants for CW and SSB. It is a nice looking receiver with some interesting features like calibrated band spread (unlike the original) but I suspect it is one of those oddball designs that has some features that would provide high performance if not for the required overall cost constraints.

I will have to give it a good test once I have it in working order. The Star Roamer II was sold at the same time as the Knight Kit R-195 which for $30 more ($99.95) and it is a pretty good solid state set. It has ceramic filters in the IF strip and a product detector and it also has an RF stage for better image rejection. I am impressed with the R-195 and it was probably a pretty good deal at its price however you could also get the assembled and good performing DX-150A for $119 and customers obviously approved because it is far easier to find the DX-150/160 series of sets than either the Star Roamer II or R-195. During this time period Allied and Radio Shack were in business together so the Realistic shortwave receivers along with the Science Fair kit line could be bought through the combined Allied/Radio Shack catalog along with the Knight Kit products.

Rodger WQ9E


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Mar Mon 26, 2018 2:22 am 
Member

Joined: Dec Sun 14, 2008 3:33 pm
Posts: 794
Location: Tokyo
Thanks, Rodger. Good to hear a positive word about the much maligned Star Roamer. It was an excellent buy for forty, 1960 dollars. I also like the antenna trimmer control on that receiver, it lets you fix up tracking errors on the fly.

Rob


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Feb Sat 16, 2019 3:08 am 
New Member

Joined: Aug Tue 22, 2017 4:06 am
Posts: 21
Hi: I'm not sure if this is right thread for this question: What is the function of the oscillator coil winding that runs from the bypassed B+ line to V1 2nd and 4th grid through 22 ohm resistor? On the Rider schematic, its at the bottom of the oscillator coil stack.
WA2BXY


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Feb Sat 16, 2019 1:53 pm 
Member

Joined: Dec Sun 14, 2008 3:33 pm
Posts: 794
Location: Tokyo
It's a tickler feedback coil. Pentagrid converters ordinarily use a Hartley type oscillator with a tap on the coil, but on the S38D, that tickler coil provides additional feedback for that one oscillator coil. It must be for the top SW band, where LO output often falls. The 22 ohm resistor helps keep the LO output more consistent over the entire tuning range.

Rob


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Feb Sat 16, 2019 11:27 pm 
New Member

Joined: Aug Tue 22, 2017 4:06 am
Posts: 21
Mine is an S38C. The oscillator doesn't start on the top range. I have a B that functions on all 4 bands to compare it with. On that one, I replaced the 22 Ohm in that leg: I found it aged to 33 ohms. That seemed to help sensitivity a little on band 4.


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Feb Sun 17, 2019 12:55 am 
Member
User avatar

Joined: Jun Sun 19, 2011 2:31 pm
Posts: 6308
Are you B+ voltage readings OK (weak rectifier maybe)? If those are OK and if you have a spare 12SA7 tube try subbing it in; a weak converter tube will start to fail from the top range down.

Also take a close look at the HFO trimmer cap for the oscillator on that range; some types are very prone to developing leakage from debris buildup which can kill the oscillator.

Rodger WQ9E


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Feb Mon 18, 2019 2:50 am 
New Member

Joined: Aug Tue 22, 2017 4:06 am
Posts: 21
I suspected V1 12SA7 might be weak, so I've got a couple on order; be here this week hopefully. I'll check B+ against the 38C. Since the band 4 oscillator wont start, all i can do is a visual inspection on the oscillator cap, but it doesn't look nasty.


Top
 Profile  
 
 Post subject: Re: S38D full wave bridge
PostPosted: Mar Wed 20, 2019 4:30 am 
New Member

Joined: Aug Tue 22, 2017 4:06 am
Posts: 21
Discovered an open tickler winding: not even much of a winding. Just an inch of #30 crossing under 1 turn of the band 4 winding.


Top
 Profile  
 
Post New Topic Post Reply  [ 18 posts ]  Moderator: Sandy

All times are UTC [ DST ]


Who is online

Users browsing this forum: W8ZV and 6 guests



Search for:
Jump to:  


























Privacy Policy :: Powered by phpBB