Sidebar Sidebar
  • You can now help support WorldwideDX when you shop on Amazon at no additional cost to you! Simply follow this Shop on Amazon link first and a portion of any purchase is sent to WorldwideDX to help with site costs.

Superstar 3900 problem

Robb said:
Found the fault . . .

The 2sa934 had a very small split on the side. Noticed this after I pulled it out. It is at TR50. Replaced it with an NTE-294. Cost was $2.

Radio temps and AM operation are all back to normal.

Thanks for all of the insights from you all that taught me so much.
Still need to learn more!

:)
Click to expand...

glad you got it sorted,

we all have much to learn;)

make sure AM is set around 4w carrier, the final will last so much longer, as will the voltage regulation/ am modulator circuits, both are common failures in this chassis (and its predecessor) when people start cranking the power up.

these are highly underrated radios, like the 148 gtl dx mk2/ss360fm they are cloned from, when not abused they are very reliable, especially earlier models. i've never lost a ssb contact yet due to the supposed drift on them, which generally settles down once they reach operating temperature.
 
Of course I don't need to tell you that this radio/chassis is common enough. Sure that you know that already.

This isn't my radio, as it belongs to a local I am helping out. Aligned the VCO voltage to 1.6v on A band/AM CH.1. Peaked the VCO input/output to .8v and 2.0v/p-p as per alignment specs (yep; I use a scope!). To tune the individual first oscillators (AM/LSB/USB) is no walk in the park. They are very sensitive and change all too quickly. That reason for the 'drift' you wrote of. They are too sensitive to temperature and go all over the place until everything gets stable first.

Receive is OK after peaking. The TX coils came up a bit too. The driver/final bias were set at 25mA and 50mA respectively.

This radio still has a few other problems. It doesn't quite have a clean bill of health as yet. The channel LED readout has been hacked so that the first number no longer reads correctly. Don't ask me why they did this as it makes no sense; the owner of this radio got it in this condition. At least it now TX's well on AM, SSB, and FM and receives well enough to use.

The other problem seems to be that the meter movement isn't reading correctly on AM. Also when on SSB, when the NB is turned on the meter reads at S9 without any incoming signal present. On AM , it won't read more than S9 on receive either. Gonna have to go some rounds with these issues yet.

Anyone have any ideas as to what to look for these other two problems? No doubt something in the meter circuit and the NB circuit. Any experience with these problems?

Comments, ideas, and thoughts welcome . . .
 
Last edited:
tr52 gets hot to your finger for a small transistor rob, too hot for its own good imo, seen a few fail, never actually measured the temperature,
its good to hear you fixed it(y)
 
Working on the S/RF meter circuit now.

Questions for the gurus . .


What function does the KIA324 IC-1 do?

Where is the AGC circuit and how can it be recognized; diodes?


Replaced TR 13 & 14; diodes D8, 9, 12, 13, & 14 for the meter circuit.
The only part I didn't replace in that circuit is KIA324.
 
Last edited:
The KIA324 is the same as the LM324 that you can get at RS. It's a quad op amp. 2 of the amps are used in the AGC circuit, The other is tha squelch amp, I think. It looks like the 4th one isn't used. Get a 324 data sheet and you can figure out what's what.

- 399
 
jazzsinger said:
the KIA is a makers mark, its for the Korean Electronics Corporation if i recall, MB is Fujitsu, MC & MM is Motorola, NJM is Japan Radio Co, the 324 is the device number,

here's a link to some of the makers datasheets on cb tricks:
Datasheet NJM324D


a link to many of the manufacturers prefixs:
Manufacturers Part Number Prefix for Integrated Circuit Company

might help you source parts/equivalents.

some general ic info:
IC's
Click to expand...

Thanks Jazz. I did buy four of these op amps. They are a very generic, common, and cheap. Paid 60 cents apiece for them. An LM324 made by ST Micro. Popped one in.

unit_399 said:
Robb -

What is the problem with the metering circuit??

BTW what happened to this radio that it has so many problems ??

- 399
Click to expand...
The metering circuit in the SS 3900 uses a 1N4148 diode in the chain that isn't sensitive enough to keep the meter linear and accurate. I put in a 1N60P that has a turn-on voltage that is half of what the 1N4148 will work at (.27v vs .58v - respectively). This makes the meter far more responsive. Could and might pop in a 1N6263 as an experiment; since this Schottky is even more sensitive than the germanium/1N60P is. Meter works fine now in all modes; but I will experiment to learn.


How to wake up a lazy S-needle on SUPER STAR 3900 (EPT3600) - YouTube


Also put two 1N6263's in the NB and receive circuit too.

Also replaced the TR41/2SA473 voltage regulator for a TIP42 that has better specs. It is an upgrade. Bought four of these also, as they were also 60 cents apiece. Gonna put one in my Galaxy DX 99V too.

Also plan to replace the two 1000uf/25v electrolytic caps, and perhaps even the 2SB754 for a 2SB817.

Replacing D86 & 87 got the RF meter to read and adjust better on TX.

Replaced the meter/lamp too.

Radio has come a long way already. The channel display LED is going to get checked out next, as the first digit doesn't work. Registers '1' and '4' OK; but not '2' and '3'. Hopefully, it is just a bad solder joint or a short. But if it is bad, I have another to replace it.

Added a talkback switch by lifting one end of D80.

What happened to this radio? Think it was well used/bit worn out. It was hacked as well; the TR32 was pulled out and so was R249. Just abused and used up. But it has come around nicely. Checked the modulation wave forms on SSB and AM on the scope last night; they both look OK now.
 
Last edited:
with all the work you've put in should be like using a different radio, the ch displays were on hk e bay a while back, not sure if they are still there, but were very cheap, they sell for about 6 or 7 quid here, but i got 25 from hong kong for just slightly more including postage, and they work on all the old uniden exports too which is a bonus, if you compare pinouts (part # UR202) in the uniden service manual it will make it easier to source modern equivalents.

i love the 3900, cheap as chips and perform as well if not better than many much more expensive radios, drift seems to be worse on radios that have the most kc shift, suggests might be down in part to the load capacitance on the crystals being slightly different to the crystals uniden used.

one of the tightest radios i found for drift was the grant classic export, but the kc shift only swung approx 2khz.
 
  • Like
Reactions: 1 person
I know this is a bit off topic, but in regard to the video, the problem with the S-meter readings, and bearing in mind that many of the "uniden clone exports" were built this way, it helps to illustrate why we see such different reports when testing antennas using the radios S meter for the results.
Add to this the effect of the different AGC schemes used in radios, it puts things into perspective.
For Rob, and everyone involved with this thread:
Thank you, this type of troubleshooting is great for the members here to read and follow through.




How to wake up a lazy S-needle on SUPER STAR 3900 (EPT3600) - YouTube


73
Jeff
 
  • Like
Reactions: 1 person
This radio has developed more problems.

Kinda reminds me of that children's arcade game 'Whack-a-Mole'.
Ya fix one thing - and then another problem just pops right up out of nowhere!

whack-a-mole2.jpg


Not saying this is a poor radio by design. Quite the opposite is true. But this particular radio that I have here has just had the crap beat out of it from years of use and abuse. For example, I had to replace the mic gain/rf gain control pot, as the shaft was so worn that it wobbled 1/3 of an inch! It has been a learning experience for me to get it all working right as it should be.

Now for the new problem.
There is a 4w carrier on either single side band mode. SSB works; but now there is a 4w carrier present - lol! Not sure if it is the balanced modulator IC. Gonna check all of those voltages next. I sure hope I am learning something; I guess I am. We will see if I can get all of the 'Moles' with as little whacking as possible; sure hope so . . .

EDIT:

IC3/Balanced Modulator SSB voltages:
1 - 2.8v
2 - 3.2v
3 - 3.2v
4 - 0v
5 - 5.7v
6 - 7.3v
7 - 6.0v

Seems like all of the voltages are off/low by roughly .4v . . . I'm gonna guess that there isn't anything wrong with this part . . .
Gonna make another guess and say that the problem might be back in the circuit that contains TR49-54 (that is where this thread started off - BTW).
Somehow, the AM carrier is being activated by a shorted part?
 
Last edited:
Put the rig into the AM mode. Scope the output (pin 7) of the AN612. It should show a carrier of about 1 vp-p. Check the voltage on pin 1 of the AN612. It should be about 1.3vdc. Switch to ssb mode. The voltage on pin 1 should now be approx 2.8vdc, and there should be no carrier at pin 7. If there is a carrier but at a reduced level, it should be completely suppressed by adjusting the balance VR. If the balance VR has no effect, check the continuity from pin 5 to the VR, from the VR to Ground, and from VR to pin 1. If these check OK ... chances are the AN612 is bad.
 

dxChat
Help Users
  • No one is chatting at the moment.


      Top Bottom
      Back