$3 For Better DX-959 Stability?

[Lil'Yeshua]

I'll root around my junk boxes and see if I can find a couple of 949's and I'll send them to you just for the cost of shipping. The more people working to fix this problem just might catch the attention of the manufacturer and they'll fix them at the factory. If I find even one 949 I'll send it to you.
I'll send you a PM if I find one that might be useable for your testing.
I'm glad there are so many Super Techs on this board.

The bean counters would probably dictate the cost what goes in these radios in order to stay competitive in the market however,I would pay $250 or more if Galaxy would produce a line of precision crafted radios that are frequency stable and would at the same time include better quality parts.
Such a radio would have to be stable on frequency,have superior audio quality,have a true VFO(I hate clunky channel changer switches),LED displays that don't cut out into partial segments,have an auto setting modulation circuit to keep modulation from going over 100%,have channel scan capability. Have dual channel scan capability,and come in a slightly smaller size,have superior receive capabilities,a good S/N ratio. Back lit controls. A DX-959S so to speak.

 

[ExitThirteen]

Tallman,

I would be most appreciative if you did manage to find a 949 or two laying around, and if they are complete (It can be just the radio, don't have to have mics, cords, brackets, etc unless you find them too) I'll be more than happy to pay for shipping if you'd just drop them in a USPS Flat Rate box or two. Let me know. Thanks!




~Cheers~

 

[Robb]

Today, C136 and C137 were changed out for NP0 caps of the same capacitance values. Used Multi Layer Ceramic Capacitors ('MLCC') that were purchased from Mouser Electronics. They are really very small in size compared to a standard NP0 ceramic cap, but pack all of the same attributes/qualities. Also replaced C281/100pf with a 100pf NP0 cap and took out the X7R rated cap that had been placed there until an NP0 was available.

Just one problem with that. Didn't get a 100pf/NP0 MLCC for that location. Ha; forgot to order it when the order was placed (got overlooked)! Once again, I had to substitute; but this time I picked up some ceramic NP0 caps locally. So using a 68pf/NP0 in its place on the top side of the board and then put a 33pf NP0 cap on the trace side of the board. When putting capacitors in parallel, you add the two values together and that will be the capacitance value. So I tried them both together first before putting them in the C281 spot with a capacitor tester and they measured 101pf. Which is close enough to 100pf. And - Bob's your uncle - they worked spot-on in the radio.

Will run it a few days before I put the 1S2687 varactor diode in at D49/SVC251 to test it against the OEM part for stability/drift.

It hasn't been cold or cool here in the last few weeks, so it is hard to say how the radio will behave in a cold environment. But it does warm up quickly when I turn it on first thing in the morning. So far, it looks like this may work out; but I will hold my breath and see if there is any problems that may develop. This radio had a lot of other problems when first purchased used; but the serious problems have already been sorted out before this experiment began.

The UJ caps (those being replaced by NP0) were chosen by Galaxy's management to compensate freq drift in the radio. If that is the case, it didn't work as the engineers calculated. I suppose the engineers designed this radio correctly from the git-go, then the decision to build the radio by Galaxy management to the vendors was made. Of course, all vendors wanted to underbid each competitor to get the job, and of course Galaxy wanted a better profit margin. In that process, they both compromised a dollar per unit and chose lesser parts; the UJ's. We get to inherit that mistake. That too is all theory, if it is in fact true - one can only speculate.

If you have ever bought a LCD monitor for your computer in the last three years and it failed on you in that time (regardless if it was a cheap brand or the best), I can tell you that from first-hand experience (having repaired three of my own) that the brand of caps they chose to use in the power supply were far less than stellar. That too was a management decision. Crap caps plague so many consumer items we buy. Not always the cause of all failures - true - but more often than not . . .

 

[Lil'Yeshua]

I bought a Westinghouse flat screen LCD tv several years ago. It lasted exactly seven months before the display started messing up so I sent it in for a replacement which lasted exactly seven months. On the second set I discovered if I smacked the top if the tv with my hand in a certain way the tv display would act right for a while. That lasted until one day when I didn't want to keep getting up and smacking it but chose my superior throwing arm aim and threw something at it. I hit the screen and 'CRACK'! I since bought an Emerson flat screen LCD tv and had no problems with it.
The Westinghouses were in my estimate,Yugos and $700 wasted.
Kinda makes me think about Galaxy radios and the channel display going bad on them.

 

[sdmahr]

Bad electrolytics were all the plague. Some second rate manufacturer, (Hong Kong / Chinese variety), thought they had successfully mastered the electrolyte formula from the Japanese and went into full production. (Little did anybody know, they did NOT have the whole recipe) They were manufactured into millions of devices, (of which the consumer was SO FORTUNATE to buy), which had a terrible mortality rate. Flat screen TV's, Computer Monitors, Power Supplies, and millions of other devices were made and subsequently thrown away as they failed.

I've seen LOTS and LOTS of these failed caps. Some puffed up, some blown, but the result is all the same.

 

[CbRadiosPRONTO]

Great info Robb. Only problem...I'll get in my pickup in February and its 7 below. Tomorrow it'll be 107 so I just open the clarifier. But, I do see how fluctuation will be less with the to change once you zero it in.


CbRadiosPRONTO

 

[Robb]

OK, now it is time for me to eat some well-deserved crow here. Been using this radio with the mods installed for the last two weeks. Thought it would be best to do it this way to give it an honest evaluation.

First off, since the weather has been pretty warm here, I cannot say just how much these mods have helped - if at all - if you live in a climate that has colder temps. My best guess - based upon what I've observed so far by putting the radio in the freezer for 10 minutes - is that it will still be way low from center freq at initial startup. If it is better; then I failed to check the before/after numbers. That is a 'fail' in terms of what I had hoped to accomplish.
But nothing ventured/nothing gained; right?

It is true that the radio seems to warm up faster and seems to drift less; but part of this too can be attributed by using it in a warm room.

The mods used in this analysis was the substitution of eight caps, five resistors, and one varactor diode. Two caps in the ref osc; three in the loop osc, and three in the carrier osc. Four precision resistors in the loop osc and one in the carrier osc. The varactor diode is in the loop osc.

I had to fix the radio before making this experiment.
There was a bad loop osc xtal and a leaky electrolytic cap in the VCO circuit.
The radio osc circuit was re-aligned each time a change was made to that circuit.

I had gone as far as to replace the loop osc switching diodes with JAN/milspec 1n4148 diodes. That did not help whatsoever. Also tested the heat on all of the resistors in the osc circuits. My thought was that if one of these 1/8w resistors were getting hot; they they might have become unstable and changed resistance. One would think so with such small, cheap resistors used in these radios; but not the case.

My best guess as to why this radio drifts is still staring me in the face. The loop osc tuning chokes are far too sensitive. Making the comparison to a Cobra 148GTL loop coils - which are also sensitive to freq change when adjusted - the 148's coils seem tame in comparison. One can quite literally accidentally breathe on one of these Galaxy/loop osc coils when adjusting them and watch it move 5-10hz on a freq counter - or more. This is true for most Galaxy radios as well, as I can say that after aligning just about every model they've made. Seem that every radio that I've aligned seems to be drift prone when/if the coil adjustments are too sensitive - this includes Uniden exports as well.

If anyone wants a little more stability from the Galaxy line of radios; then replacing two caps in the ref osc and three in the loop osc with NP0/C0G caps can be done. But don't expect too much here! Using a 1S2687 diode to replace the OEM SVC251 did little to improve anything either - but it is still a better part than the OEM part.

IMO - at best there may be 10-20% improvement with these mods - if that.

Sure wish another batter would step up to the plate and knock a homer here.
Batter up?

EDIT/FINAL THOUGHTS:

If the loop osc coils were replaced with coils that didn't have quite so much freq change; that would be my best bet for a reasonably stable radio. Please prove me wrong. Have no way of experimenting/knowing what the min/max inductance value of these coils are. It is not shown on the schematic or the part list. Not to mention, what would be the mean range needed to reach proper adjustment value/freq? Suppose one could take out one of the loop coils and test it with an accurate LCR meter. Test it for where it was set at when taken out of the radio first; then turn/test it for min/max inductance. Finding a test substitute that will fit the form factor of the board will also be a challenge. Not sure if it needs more inductance or less.

Unless proven otherwise; then Galaxy really can 'drop the baul' for SSB.

 

[sonoma]

I had done this to a galxy radio in about 2002 and it help a little only in the mornings when it was about 18 degrees or so. before I did the mods it would not even receive until it warmed up. it does help a little but like you said it is not like a 148. I never could figure it out. was hoping you had better luck than I did. I even had international crystal cut me a new 10.240 crystal to see if it would help. no luck there either.

 

[CbRadiosPRONTO]

I'll step up with a solution.

1: Open Clarifier;
B: Place right hand on clarifier;
3: Adjust as needed

These radios really do sound good on SSB regardless of what anybody says. But yes....drift.

73's

Wait, 72.999's.




CbRadiosPRONTO

 

[Grinder74]

I thought it was 27.735?

 

[C2]

I will make an analogy here using an old school-yard teeter-totter. The leverage in this system works in opposition, when one side goes up, the other goes down. If you were to replace the center pivot with two mating gears on each side of the bench, then when one side goes up, so does the other.

A similar principle can be applied to circuit design. You are trying to stabilize one side of the circuit with better parts, but it is much easier to use that movement in a system that compensates equally and oppositely.

How has Galaxy addressed these dynamics in their circuit design? Do you suppose there is a circuit patent that they did not want to pay to use?

 

[Robb]

Think that the whole idea that the engineers used the UJ caps in various parts of different circuits was to act as a 'teeter-totter' or temperature cushion if you like. Isn't that where you are going with this? While that may have been the idea of the well-intentioned engineers, it didn't land there. In fact, would go as far to call it landing on a whoopee cushion. If it worked half as well as they planned it, this thread discussion wouldn't have even happened.

A radio chassis class that works well enough for the price point in every other regard except for this one major failing begs for some answers. Information empowers; wrong info leads astray. The once gold standard Cobra 148GTL is now also a non-competitor too. Guess we just expect better for our buck; I do.

Certainly if we stiffen up one side of that teeter-totter - as this experiment played upon; then shouldn't the other side be even far more out of tolerance?
However, that didn't happen.

Is it an issue lending to possible copyright infringement? Dunno. Doubt it. But if that were so; then Uniden would have sued Galaxy long ago for lifting the 3600 chassis design - wouldn't they? Perhaps they wanted them to have it. I still use the 'nifty-drifty nine-fifty'; but making it possibly better even if the mfr refuses to do so just makes it fair game for the consumer - IMO.

 

[Lil'Yeshua]

It means that Galaxy radios are less than perfect. Take a design and make it better.
I think that if you improve a particular section in a radio then it adds to the whole. What if you went through a Galaxy and made it perform like a quality radio? It's value would go up as a result. It's like taking a Chevy and transforming it's power train,suspension,and other systems to function like a Mercedes.

 

[C2]

When I left technician land years ago, we would call it shotgunning. It had about a 10% success rate, just for replacing a failed part. I gave up that approach quick. Another thing I learned when trying to troubleshoot closed loops is that components that affected loop lock and noise the most were typically not the problem components. if this is not understood, then use of freeze spray and heat guns can easily lead one astray.

Does this circuit even have temperature compensation feedback? How does the basic design (block diagram) differ from the references (whatever radios people are using as their holy grail of stability, what? a 148?)

not trying to mislead or misinform

 

[Lil'Yeshua]

When I left technician land years ago, we would call it shotgunning. It had about a 10% success rate, just for replacing a failed part. I gave up that approach quick. Another thing I learned when trying to troubleshoot closed loops is that components that affected loop lock and noise the most were typically not the problem components. if this is not understood, then use of freeze spray and heat guns can easily lead one astray.

Does this circuit even have temperature compensation feedback? How does the basic design (block diagram) differ from the references (whatever radios people are using as their holy grail of stability, what? a 148?)

not trying to mislead or misinform

You have to take all of the support systems into account when modifying one particular circuit loop. I for one would like to see a Galaxy that's stable on freqs. Shoot,I'd sent my 959 for modification to ROBB if he can make one perform better.
By the way Robb,those modifications you're doing could be a closely guarded secret supposing you were going into the radio mod business. Snake Radios did the same thing. I think people are willing to pay someone knowledgable who can mod a radio in a good way.