National SW-54 Thoughts - Videokarma.org TV - Video

Dubis7 · #1 04-03-2020, 04:40 PM

I think I'm actually on the right track with this one as is, but I wanted to get some thoughts from anyone who has more experience than me.

I'm rebuilding a National SW-54 shortwave radio. It's a pretty standard restoration so far, but this set has a few major issues. The cheater cord on the back was physically melted, and the shunt resistor across the rectifier tube was burned in half. As near as I could tell, this was due to the set being run with dead electrolytics for long enough to cause a meltdown. There's a scorch mark on the underside of the cabinet where it failed.

Anyway, I've replaced most of the caps and rebuilt the power supply. I picked up a kit from Hayseed Hamfest and apparently my set is a later revision with some additional .002 uf caps between the antenna inputs and one of the coils. No biggy, I've done the rest and those won't impact a basic power up.

Well, anyway, I got it to come to life for a moment, but it died again just as quickly. Checked under the chassis, and there's two resistors in series. One goes to a ground connection through Pin 1 of the 50C5, and the other acts as a bridge between two of the filter cap sections. The one providing the bridge was smoking and hot to the touch (my finger proves it, oh well) so I measured both. Bridge resistor (looks to be a 1K from the schematic) looks to be in tolerance, but I'm getting some odd behavior from the one going to the 50C5. It measures 1K when the set is cold (should be 15K if I'm reading the schematic right), but I noticed that the resistance dropped to 0 when I applied power.

A dead short to ground would explain the smoking 1K resistor, and probably my dead reception as well. I mostly wanted to know if having a resistor become a dead short when power is applied sounded like something anyone has any experience with.

I'll attach the schematic as well, although it's not a perfect match to my set. On mine there's an additional resistor between the 1K bridge and the 60uf filter section that I can't seem to find on the the diagram, and the resistor that's marked as being in parallel with the C1D section coming off of Pin 1 on the 50C5 is in series with it instead. I'm assuming mine is a different revision since it also has a network that isn't listed here, but I'm no expert.

Since both resistors were getting hot, I have parts on order to replace them. I'm hoping removing that wonky 15K will get me back up and running. I have to believe the meltdown I mentioned previously took that out as well.

Dubis7 · #2 04-03-2020, 09:41 PM

Okay, a couple corrections.

C1 D is in parallel with that resistor. It didn't register with me that that line is just a straight bridge to ground through the filter capacitor on the other end of the chassis.

The reproduction electrolytic can from Hayseed Hamfest used stranded wire that was too thick for the holes in the tub socket terminals. I ended up cutting half of the wires off to make it fit and feeding that through. In poking around tonight, I discovered that a single strand on the 60 uF section had separated from the rest and was hiding underneath, shorting that section to ground. I've removed that, but now I have to see what damage that caused.

I did try powering up since then, and it looks like the resistors are no longer overheating. I still plan to replace them, of course, but I'm now reasonably confident that that short was the source of my issue.

Basic symptom at the moment is I have faint hum through the speaker (not filter hum so much, more like volume up too high faint distortion) but no signal besides a faint chirping every once in awhile on all bands. So I'll have to figure that out. I plan to check my B+ voltage coming off the rectifier tomorrow (I think that's going to be 115VDC coming off of pin 8 on the 35Z5, but correct me if I'm wrong on that) and trace it up to the 50C5. It looks like it has to go through those resistors that were overheating, so I still have some hope that that's the source of my issue.

I'm still trying to wrap my head around the resistor pattern coming off of pin 8 on the 35Z5. It doesn't match what I'm understanding from looking at the chassis. Pin 8 has two resistors. One is definitely the 15K ohm going to Pin 1 on the 50C5 and connecting the the D section of C1. The other (the one that was burning hot) is connected to pin 8, but goes off to a terminal strip, hooks onto one of the 40 uF sections on the electrolytic can, and then runs through an additional resistor to the 60 uF section in the can mounted to another terminal on the strip. At least that's what I think is happening, I'll retrace it tomorrow to double check.

There's also some stuff wired to pin 1 of the 35Z5. I don't even see that pin listed on the schematic. I know it's factory, as far as I can tell the underside was untouched so there must be some reason, but I'm not seeing where all that is happening. Is Pin 1 on the 35Z5 a no connect internally? The metal is present but perhaps a look at the tube diagram will clear that up.

I'm just spitballing here. Hopefully some of this makes sense. I'm sure another glance over the layout will clear some of this up, but in the meantime any thoughts are appreciated.

Dubis7 · #3 04-03-2020, 09:44 PM

Okay, yep. Pin 1 on the 35Z5 isn't connected to anything internally. That threw me off. Pins 4 and 6 are also unused as well, but they didn't include any physical pins for those. I figured that meant that any pin that was present was connected. That may very well clear up some of my confusion.

Electronic M · #4 04-06-2020, 11:33 AM

Wow that's a lot of info...

One thing that jumps out to me is your statement that resistance of a resistor changes when power is applied. How did you determine that? If you left your DMM in resistance mode connected across the resistor and powered up the set then you are making a BIG mistake....DMMs measure resistance by either sending a fixed current into the resistor and measuring voltage drop or sending a fixed voltage into the resistor and measuring current flow....If the radio is powered and sending power through the resistor the resistor then has an additional power source across it that changes current flow and voltage across the resistor (which the DMM don't expect)....Best case scenario that power the meter doesn't expect makes your reading substantially inaccurate and worst case scenario the voltage and current it sends into your meter destroys the DMM...
I've had the later happen by accident multiple times...I tend to switch between measuring resistance with the set off and voltage with the set on and sometimes forget to change mode on the DMM.
If you like your DMM and want accurate readings NEVER measure resistance of a powered resistor!

Dubis7 · #5 04-06-2020, 12:29 PM

Okay, thanks for the heads up. I think that may be my issue. The multimeter is fine, at least. I've tested it since then and everything is behaving as far as I can tell.

I'm not seeing any voltage on my B+ (Pin 8 of the 35Z5 if I'm correct on that) and I think a few of my electrolytic sections have opened. I read that you can check electrolytics with a multimeter by reading resistance across them. If they spike once and go down to OL very quickly, they're good. If they retain value, they're bad. Two of my sections are retaining value and two are dropping, so I'll just go ahead and replace the whole thing and those two resistors.

Of course, if anyone else has any thoughts or that all sounds completely off the mark, please let me know.

jr_tech · #6 04-06-2020, 02:47 PM

I would also check continuity (measure the ohms resistance) of the pilot lamp, the 330ohm resistor and the heater pins of the 35Z5 2-3, 2-7 and 3-7.

jr

init4fun · #7 04-07-2020, 07:00 PM

Quote:

Originally Posted by Dubis7

I read that you can check electrolytics with a multimeter by reading resistance across them. If they spike once and go down to OL very quickly, they're good. If they retain value, they're bad.

Absolutely 101% not true ....

The only infallible test for an electrolytic capacitor is to either see it work in it's circuit without it or it's other associated components getting hot , or to test it using a capacitor tester that tests it at the actual voltage it's gonna see when it's in use . A cap that won't break down at the tiny (less than 9 volts) voltage your meter applies to it for the test could very well break down at the 100 or so volts your 35Z5 is gonna feed it .

My personal policy is "take no prisoners

with electrolytics , if I'm gonna get a piece of vintage tube electronics running , the first thing I do is to replace them before the first power up .

Dubis7 · #8 04-07-2020, 09:58 PM

Quote:

Originally Posted by init4fun

My personal policy is "take no prisoners

with electrolytics , if I'm gonna get a piece of vintage tube electronics running , the first thing I do is to replace them before the first power up .

These are replacements. The concern is that the positive of the 60 uF section was shorted to ground, which also has a direct connection to the negative of the same side. I know it caused the 1K and 15K resistors to overheat, and I'm concerned it may have also damaged the electrolytics.

I've ordered a replacement set for safety's sake. I'll try substituting that so I know those and the new resistors I ordered are good. I'll have to start doing continuity tests from there.

I know the pilot lamp and heaters are good because the lamp is still lighting and the series string is all coming on. So the problem is somewhere further into the set.

Does anyone have any specific ideas on checking the B+? Am I accurate in measuring between pin 8 and ground to see if that's coming off of the 35Z5?

Dubis7 · #9 04-08-2020, 03:32 PM

Okay, I've got no B+ coming off of pin 8 on the 35Z5. For sake of testing, I've gone ahead and ordered a spare tube. My heater should be good since my tubes are all lighting and the filament tested with continuity, but I checked it in my tester and it came back as no emission so perhaps that's my problem?

I want to verify that my voltmeter is good, though. I know the resistance checker part is fine, but to be safe I'm going to check the voltage on the heater pins. That should give me a good indication on what's going on.

Any other suggestions? I have a new filter cap set on the way to see if I burned out the old one, and the new resistors are installed and not getting hot. This has to be a power supply problem, whatever it is.

Dubis7 · #10 04-08-2020, 03:41 PM

Okay, now I'm really confused.

I used the voltmeter function on my multimeter (on AC setting) with the positive lead on Pin 7 of the rectifier, and the negative lead on ground. 0 volts. I then moved it to the negative lead to pin 3 on the 50C5 and measured voltage between pin 7 on the rectifier to pin 3 on the 50C5. That should be filament. I know it's lighting. I get hum from the speaker and there's a pop when I turn the bandswitch, plus the dial light is lighting. But 0 volts on the filament!

Am I just using this wrong? Maybe my meter is dead after all? I can't believe I'm not getting any voltage there.

Dubis7 · #11 04-08-2020, 03:51 PM

Quote:

Originally Posted by jr_tech

I would also check continuity (measure the ohms resistance) of the pilot lamp, the 330ohm resistor and the heater pins of the 35Z5 2-3, 2-7 and 3-7.

jr

Oh, and I almost forgot. I did check this. Each pin set showed continuity.