• 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.

Galaxy radio drift

Here is what I have come up with so far gentleman. Immediately upon installing the three can unit with a ground, the eradic frequency change stopped and the frequency counters 6th digit on the radio went stable. Anyone that owns one of these radios knows that that six-digit is always going up to or down to up to and down two. Just to be safe, I went ahead and grounded the case on the frequency counter Crystal.


All of us are right on this and all of us are wrong. Each one of us has a little piece of what is required to make this radio stable. Not just me, not just Sue, not just anyone. But together we all have a cure.
 

Attachments

  • IMG_20180905_213611~2.jpg
    IMG_20180905_213611~2.jpg
    166.8 KB · Views: 36
  • IMG_20180905_213601.jpg
    IMG_20180905_213601.jpg
    7.8 MB · Views: 36
With legs like that, grounding the case probably didn't do much good. Maybe a little bit.
 

Attachments

  • IMG_20180905_225705~2.jpg
    IMG_20180905_225705~2.jpg
    375.7 KB · Views: 30
Well, shoot just as the conversation starts to condense into something can write about...

We are looking at drift, and finding a better method to use for those whom can't wait or are in an environment that doesn't work well for the typical methods...

One aspect was encasing the entire XCO in a shield - and in older radios - that was THE go-to for it.
Another was the oven method, you warm up the innards of the radio - in conjunction with the above - you got things to work well within a limited, but still a waiting, period of time until you can use the radio in a roundtable.

I was simply trying to show that many a good clarifier circuit gets ruined by poor mods- because as I fix the radios that come across my desk, I find in the Clarifier replacement column - they're replaced because the end user had a mod done to the circuit to maximize slide or several attempts at mods to make the Coarse and Fine track as one to help them tune others in. Another problem was due to the mod itself damaging the substrate or the pot to make it unusable because the user modded it in a way it popped like a fuse and in some instances - took out parts of the VCO circuit too - at least a foil trace or two.

If you filter out just my messages in this thread, I wasn't bashing the temperature effects or even downplaying them - I'm learning right along with the rest of you. I'm looking for options...I have several but I need the rest of those jumping into the thread - don't add so much foam to it - that we can't see the efforts others are making.

With that being said...

How To Build A Better Mouse....

1. The Mayhem, The Animal - The Beast...

For Clarity here's how - or what I do - for someone when they want to unlock the Galaxy Clarifier - I'm using 98 for this - I can show this also on a 949 as needed...my own stock radio...

But because more documentation is available for the 98 - this is shown Graphically to help others with older style radios - work alongside as you may want to investigate these as I go...

Since there has been issues before - I will add that the graphics contained have logos and originations and sources for their specific design and all copyrights are held by the respective companies ...
- used for Educational Purposes Only

First though...
ssBoardForGalaxyConversion.jpg

Schematically...
QUIKGalacyClarifierConversion.jpg

What the front panel looks like ...

FineCoarse98VHP.jpg



2. Select the Right Cheese...

When the conversion is complete - there is little if any retuning that is necessary - for the Diodes, R113 and all parts are kept intact, the only thing that has changed in this, is a constant source from the 8 volt PLL power supply is used.

The mod listed above unlocks BOTH Fine and Coarse and ties both together to use for VCO to track SSB modes and AM FM and any mode in-between. RX and TX are sourced to a single power supply reference.

The cheese is in the use of two diodes and this is how I found out about the COMPENSATION the diodes can provide was by accident. D68 and D38 are the two I'm referring to in this.

The drift the galaxy would have - and after this conversion - still does occur - is slightly different - but still inherited by the design and parts used. So we don't have all our ducks in a row yet.

What I want you to pay attention to is the Voltage Divider design - for it isn't all the way up to source rail voltage or all the way down to ground buss - it floats - not towards the ground buss, more in the upper range of voltage but not all the way to the top.

A lot of people, techs or otherwise have always complained that the Galaxy radios have this "drift" problem - it's from the design itself - it's not all to blame on thermal though - it's the sheer range of tuning both the Fine and Coarse provide the person when it comes to tuning in to someone.

To blame it on thermal would be fantastic for then we'd have a fighting chance to force the maker to re-design the radio to suit the user needs better and save us from having to face the music from the customer for the makers bad design in the first place.

I reduce some of the drift by the use of the resistor and diode in series - limits slide but the user is able to recover the channel center and use the "detent" on both controls to obtain it - faster.

3. Use Your Mouse Like You Own It...

One thing I do see, that many others are either not paying attention for or not offering any recourse to fix is when the user transmits - it Warbles - or it's frequency is not steady. This is a separate problem dealing with load variation - the power supply and the voltage regulation on it, and from it, is not the best. Although some improvement can be made in current handling by upgrading the 473 to a greater capacity transistor - it does not offer much more than extra capacity. There is another problem and that deals with the sensitivity of the Varactor to changes in voltage from the supply.

Again the upgrading of the transistor is a good option but to reduce the effects of loading - we still have the problem of a large window of adjustment or range to tune and we have it straight across the 8 volt rail.

The mod shown above is a simple one an uses R113 as a BUFFER from the voltage supply. This helps to reduce the loading off the rail into the clarifier - but little else only other options are increase the R113 value to limit slide even further or abandon the dual concentric and settle for single range carbon composition potentiometer - with todays radios, not an option.

So limit power supply drains by reducing the sags (dropout events) by keeping limiters in place and operating normally by setting AM power, ALC, AMC and FM Power and deviation levels to operate like stock.

I guess what I'm trying to say here is there are little factors that many would not tend to think as problems until the radio is heavily modded and or limiters removed - then do we notice the failures and shortcomings in the system.

So while everyone blames drift on thermal - there are a lot of other mitigating factors that tend to cause more problems even for radios that are warmed up and on frequency - including poorly managed regulation and or heavy modulation conversions that affect the power supply and it's ability to keep up with demand.

I can also show you other methods I dabbled in for offsetting Clarifier drift by using diodes ACROSS a resistive shunt.

How that method works is by knowing how much of a voltage variance is happening when the radio is cold to when it warms up - anyone whom has used a 2950 and 2970 knows the RIT Clarifier tune WHEN COLD is not equal to the frequency displayed nor is the expected frequency of the VCO is exact to the display even if the RIT had been kept center slot - even correctly tuned AFTER it had been warned up to room temperature. This is due to the loop issue I discussed earlier - the tech can do some initial measurements of voltage using warmed up equipment on that cold radio - to see a profile of the voltage rise or fall on the wiper line or the line that goes to the XVCO VCO from the RIT and TX from PLL feed into - the line is from the voltage divider of the source line and the ground buss, again a drift that can be compensated for by using a diode across a resistor - use a simple formula to obtain the voltage drop needed when cold and sub in accordingly. The tech or user can use a FIXED resistor to provide a known drop and the diode or series of diodes (some PLL and RIT differences can get to as high as two volts from one side or another requiring the inherent voltage drop across the resistor paralleled with the set of diodes) to obtain the compensation thru the voltage drop changes the diodes provide across the fixed resistance.

So that is another thread for later. The whole effort here is to show that although thermal seems to be the paramount importance for the Drift problem, I've seen other issues of drift that are not seeming to be covered here, which is part of this problem - it's not going away so we have to face it.

:+> Andy <+:
 

Attachments

  • Galaxy33HMLVCO.jpg
    Galaxy33HMLVCO.jpg
    87.9 KB · Views: 16
  • SVC251.jpg
    SVC251.jpg
    119.6 KB · Views: 19
Larry,
now that you've done your experimenting, it's time to make sure that your design works on another previously stock radio.

There is always the chance that what you experienced with this radio was an anomaly specific to your radio, and the only way to be sure that you are fixing a common problem is to fix the same cause with the same solution.

Honestly if i was considering selling somethings like this as a product, i would probably buy five radios and test it in all of them in order to get some sort of baseline that a customer could count on.

the last thing you want are a bunch of people trying to return your product saying "it didn't work".

LC
 
Larry,
now that you've done your experimenting, it's time to make sure that your design works on another previously stock radio.

There is always the chance that what you experienced with this radio was an anomaly specific to your radio, and the only way to be sure that you are fixing a common problem is to fix the same cause with the same solution.

Honestly if i was considering selling somethings like this as a product, i would probably buy five radios and test it in all of them in order to get some sort of baseline that a customer could count on.

the last thing you want are a bunch of people trying to return your product saying "it didn't work".

LC

The statement of making a kit was just a passing thought and really not a great consideration.
 
I removed the cans from the dx2547 and install them on my dx-959 and got the same results. But this is just part of the problem. I can rectify the heat problem with fans but it's going to require some type of internal heater for the winter months. Most of you all know more than I do. I'm just trying to find a solution for the DX series of radios.
 
Somebody posted that earlier in this thread. It's a great idea for one component in the winter time, but it does nothing for the rest of the inside of the radio when it's 22 degrees outside.
I knew a guy in the early 80s that put an 1157 Automotive bulb inside his radio with a switch. He said when he went out to start his car he would turn the switch on. I don't necessarily recommend this, just sharing information.
I can stabilize the radio in normal and warm operating environments. But for the cold environments, something else is going to need to be done.
The power supply in the DX 2547 is always on. It has a core temperature of approximately 147 degrees. I put a small fan on it and it stays cool. It never produces any heat. I also put a fan in the bottom of the radio to force fresh air inside. Combine that with everything else and it's stable. But I know that when I come out to this Radio Shack in the dead of winter and it's freezing cold because the heater in the shack has not been on, it is not going to boot up on frequency.
 
Even the most expensive well designed amateur radio is going to have to warm up before it's stable. Some take more time, some take less.
All you can do is mitigate it or shorten the warm up time.
That really means keeping it warm.

That's not a radio problem but an environment problem.

That said, I think the 2547s power supply runs as long as it's plugged in.
That means you can run small heaters on the crystals and insulate around them with foam. that will give you pretty close to instant start up in any environment as long as it was plugged in and the heaters were on a constant supply.

But you need to heat all three.
Just get heaters and whittle away at a piece of foam for the top and a solid chunk for the bottom of the board and it should be fine.
 
  • Like
Reactions: tecnicoloco
Yeah; slang for picofarad. Like 330pf is said as '330 puff'. Strange - huh?

Anyway, When I did my thing with the 959, I checked the regulator voltage, returned the radio to stock (replaced the AM limiter, and had to replace two caps in the PLL along with the 10.240 xtal) just to get it reasonably stable.

But I also replaced the U/J valued disc caps (C116 & C117) with NPO's of the same value. (BTW - this is also true for C282) Which made it even more unstable. The U/J caps compensates for drift; but very inaccurate in doing this. So I put the U/J's back in. Then the loop osc xtal was also twitchy and jumped around erratically, After replacing it, that circuit was more stable once again.

These radios sure seem to require a lot of attention . . .

Like LC said, if replacing that one disc cap with that SMD device made it stable - shown in TM86's video; then it would need to be tested in a few more radios before any assessment could be made..

But I have found that these radios have a couple of unstable circuits and that the whole radio needs to be checked out and corrected in a few places even before it became anything close to stable, so I also agree with HA.
 
Last edited:
  • Like
Reactions: Shadetree Mechanic
But I have found that these radios have a couple of unstable circuits and that the whole radio needs to be checked out and corrected in a few places even before it became anything close to stable, so I also agree with HA.

So more or less the Galaxy (and similar) radios have to evaluated on a case by case basis, due to the "quality" of the original parts.

BTW, didn't see that someone had posted the same video earlier in the thread. Would have let this sleeping dog lie if I had.
 

dxChat
Help Users
  • No one is chatting at the moment.
  • dxBot:
    Tucker442 has left the room.
  • @ BJ radionut:
    LIVE 10:00 AM EST :cool:
  • @ Charles Edwards:
    I'm looking for factory settings 1 through 59 for a AT 5555 n2 or AT500 M2 I only wrote down half the values feel like a idiot I need help will be appreciated