• 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.
  • Click here to find out how to win free radios from Retevis!

EPT360015C Connex cx3400hp dual mosfet issue.

Just noticed something Andy. When I gave you those voltages above, the 56k ohm resistor had popped off and I didn't know it. When trying to get it down is when I realized.

Now with the 56k back inline, the driver trimmer has to be turned all the way open, and even that's not enough. Although there was no heating issue with the regulator during this time. The finals trimmers are slightly to the left now.

AM power adjustment is set at 5.45v with front control at mid way point.

I did just remove that 56k again just to see if I could get it around 50ma on the driver. I got it there at slightly to the right of halfway and had no heating issue at all on the regulator then. Only thing is, the power shortage with it like this. Also, the pre-driver is the only thing getting slightly hot after 5 minutes now, the regulator stays warm as well as everything else.
 
  • Like
Reactions: Handy Andy
I want to show you something I've seen every day - I'm sure others can chime in on this too...

A quick way to check FINAL BIAS voltage...EXCEPT DOES NOT WORK ON DRIVER

upload_2021-11-25_15-3-14.png

Are you measuring BIAS - HOW?
upload_2021-11-25_15-17-33.png
An Alternative Method....
Example: FOR EPT360014B BOARDS!
You can measure bias across the Trimmers themselves - by VOLTAGE - but does not work on DRIVER see Post 31 because it's last leg is left open - you will have to measure BIAS at the GATE at the DRIVER

IF you own a Galaxy 959 the Bias trim pots are Mirrored - meaning they are REVERSED to each other...
Close-up
upload_2021-11-25_16-22-41.png
upload_2021-11-25_16-3-11.png

Meaning their orientation is OPPOSITE of each other TO RAISE or LOWER are opposite rotations - trimming both (rotate) to the Same side - you can turn one into LARGER HIGHER GATE VOLTAGE - hence more draw from AM Regulator - while reducing the OTHER TRIMMER by LOWERING GATE VOLTAGE - in one way you're increasing the current DRAW at the FINALS by turning both pots together the SAME way...only you're saving one but at the expense of the other...

I've seen it too many times so I just wanted to point this out for you so you don't make the same mistakes others have made and wound up trashing their radios thinking they are similar to their PREVIOUS lines...
 
Last edited:
AM power adjustment is set at 5.45v with front control at mid way point.

I did just remove that 56k again just to see if I could get it around 50ma on the driver. I got it there at slightly to the right of halfway and had no heating issue at all on the regulator then. Only thing is, the power shortage with it like this. Also, the pre-driver is the only thing getting slightly hot after 5 minutes now, the regulator stays warm as well as everything else.

Ok, whew ( hesitanly)

Ok if I read the above right, you get PROPER mA draw now...

What is meant by "the Power Shortage with it like this" - in what way? LACK of AM Power?

Pre-driver uses a 5pF Disc cap ... there may be a thread or two you'll come across about a Galaxy Radio Component value missing on Schematic but they call it X...

This is one way they solve it -
upload_2021-11-25_16-16-49.png
You mentioned "heating" they use a part 2SC2314 - C275 is approximately 5pF - I say Approximately because this value can change.

Even be omitted, by specific platforms of this same chassis - so it may say Ranger or Galaxy or Connex.

All of them may use the same board, but small changes and performance curves arise from all of them because of these specific parts in specific locations are omitted, changed or otherwise altered to make the "platform" perform in a way to enhance (if that is the word for it) a given feature or type of performance.
 
Last edited:
  • Like
Reactions: NZ8N
I'm checking the bias voltages by the bare leads of R218 for driver, L50 for 1st final and L35 for last final.

I didn't mean "this way", I had something else on my mind also. What I meant was I still can't get the power up to what connex states this radio should do, not even near it yet. I'm glad to have the overheating issue on the regulator resolved. My whole problem may have been that hair line solder bridge you pointed out earlier in this thread.

C275, yes mine has that cap in place.

I got the finals trims set to 55ma each, totaling 110ma.

Everything seems right now, just low on output power.

Came a long way. Couldn't have gotten this far without your help Andy.
 
  • Like
Reactions: Handy Andy and NZ8N
Did you ever wonder why?

They "stated" 45watts, is based upon - what type of system, product, item or just plain old "Meter" to state that claim?

(or the Spectrum used to state such - might be a touch wider than many are left to believe)

Truth is, they don't - or at least show, with any sense of honesty what they used to make such a claim.

The truth hurts, but to really put this in a given perspective...

I find a truth based upon the events leading up to it's arrival on your doorstep - can be found on the side of a box of cereal...

This Product is sold by Weight, Not By Volume.
Some Settling Of Contents may have occurred
during shipping and handling.

I cannot lay claim or even give you the reasons why the power they state doesn't seem to appear on many meters - it's not a universal "gimmie" either.

Welcome to radio...

Now as far as the "drive" - take a look at the radio and begin the tune up process - if you have a scope it would be a big help. at least you can find where the weak points are...

Let's also look at current draw from the power supply to help us determine if there is any true "45 watts".

One effect you may have already noticed is the accumulated power ability is affected by the Trimmers for the Final - and you also know what the results of that effort get you closer to recreating the problem all over again - just for "gain" - Yes power will go up but the laws of diminishing returns starts to appear.

So you know the thing is widebanded - and if adjusting the trimmers up (increasing voltage) makes the radio increase power - ok let's get out the old coffee napkin while were here at the table and figure out what is occurring when you raise the voltages on the Gates...

Part of this problem is the "capacitance" you have to overcome at the same time use - to shove RF in there - on top of that we also have to tweak the voltage arriving so that just RF is in the MOSFETS Linear region. Between the two - we also have to have enough to transition from off to on - the linear region "moment".

I call it moment because you can look at this in the electrical as well as physical standpoint - being electrical you can apply power to it and make it do things - the physical being that it is a switch with a given physical characteristic of being able to switch power on and off easily enough.

But in both realms - its' not moving so we have to look at the Electrical as a atomic - physical event - but also a FIELD event to force a state from of to on as a transition but in doing so, there is these - moments - where the device appears as a linear switch. The way it does this is using a field, but also the physical presence of power as a force - not as a current but as a voltage.

So let's work on this process ...

You can read along by opening this in a separate tab on the browser if so equipped...
https://www.worldwidedx.com/threads/cobra-29-ltd-classic.260992/page-3#post-754762

upload_2021-2-14_12-50-50-png.43151


To keep this simple and so I can celebrate what is left of Thanksgiving with my Wife...

The key to this approach is two fold.

One Main one, is idealizing the MOSFET so it gives the stage the best transfer from input to output - these use open loop gain so the gain being the quality of output following the quality of input. Give the best to it and it is possible to get the best out of it.

The Second, is dealing with the Stages preceding (Upstream) and subsequent (Downstream) - that has to be optimized so the signal transfer can be made with the least amount of losses (like skewing and clipping/saturation and remembering that the previous stage is using a Bipolar Linear device.

So where's the problem? IN all of it.

So before we proceed - to optimize the design will seem a bit of substitution and perhaps a little more of when if that worked let's do this - more of an experiment than a true tested theory and making it stick.

Again, 45 watts can be the light in lumens it produces or the amount of heat it produces when you try and talk to other people with it - onto it takes 45 watts to operate the pretty lights...
 
  • Like
Reactions: Hawkeye351
To me If you got "45 watts" right out of the box, then too, we have to somehow recreate that condition.

So you changed some parts - one being the TIP-36C - so to me the trade off is the from the HFE standpoint of gain from the OEM part even if it was a typical TIP-36C - they offer a range of HFE "values" or range within the Datasheet.

So we can't cite them for misrepresentations there...

Only other thing left that we can do is to look at how much different the strip appears to their "Idealized" stated ratings...

But I do need to ask - and again a "fine print" problem...

Did they state anything along with the 45 watts? Like PEP?

Did you get this right out of the box?
(Gotta' ask the customer that - what they had to endure from the shop they bought this from)
(It also might explain the condition it arrived to your doorstep in...) :):)o_O
 
  • Like
Reactions: tecnicoloco
Andy, after looking over the trimmers, I noticed the am power trimmer looks different from all the others. It is white, whereas the others are normal. It's the one near the am regulator, it has a 502 stamped on it.

1126211027_HDR.jpg

1126211027.jpg

Also, I noticed 2 surface mount components on the trace side near the rear on the final side.
1126211032_HDR.jpg

Are those standard?
 
  • Like
Reactions: Handy Andy
IN this photo...

upload_2021-11-26_12-0-28.png
These are your RF "Protection diodes" or PIN Diodes used
to help the receive...
In this photo...
upload_2021-11-26_12-2-37.png

You have a right to be concerned...

If you can check this parts "output" make sure it is linear - for if you're having issues trying to find power - this pot looks to be pretty close to full on - so something is not right in "accepting" this into the AM Regulator.

Dug up this from an archive, it too uses MOSFET (Note diodes) Arrow points to AM Power pot...
upload_2021-11-26_12-8-42.png
So even though the board is more populated - the pot locations have not changed.
This is a stock-radio photo...
In that same radio
Note the value of R219
The Drivers GATE Bias feeder resistor...
upload_2021-11-26_12-36-50.png
IN that radios schematic...they list the WRONG value...
upload_2021-11-26_12-51-35.png


Interesting...
 
Last edited:
  • Like
Reactions: tecnicoloco
Where it's positioned at now is 5.45v with front power control at midway. It is almost wide open, which is 6.1v.

As far as where I got the 45 watt factory output figure for this radio is from the manufacturer website. The manufacturer didn't mention what equipment they used to get that figure but they did state 45-55+ watts and they do state PEP.

Also, I need to get some new glasses, hahaha, just noticed it's not 34 watts output on peak, it's only 28 watts output on peak. That's after I realign the rf stage for max. I think it's L40, L42, L43 and L44 I touched up. Driver bias at 50ma, both finals at 55ma each totaling 110ma, power trimmer set at 5.45v, nothing getting extremely hot any more.

Getting old, lol...

The owner stated he never knew how many watts it did, he never checked it to see how many watts it was doing. Although he is very happy to hear we got the heating issue resolved and we got it to transmit again. I like working with people like him, very patient and appreciative.

He did want me to tell the one helping me along (you Andy) that he highly appreciates the help you're putting in to help him.
 
  • Like
Reactions: Handy Andy
Ok, well, the readings may be what gets others to grind axes and light torches...

Speaking of that - I do have some upset people that think I spend too much time helping others - so they given me an ultimatum and so I must step aside for the day and enjoy their being here...

Hope you enjoy your weekend - I'm thinking the XYL and the Mob here wants me to spend more time with them...

Which in reality - makes life a little easier on this end...;)
 
He did want me to tell the one helping me along (you Andy) that he highly appreciates the help you're putting in to help him.
I caught that!

And tell Him - Thank Him, for his patience while we work on the Patient.

The Surgery - is going well - so far.

Putting the parts in where they should be is a lot like the old MB Game "Operation" - so if we get this to work, by our efforts, wits and luck by which it seems to do fine with - we just might be able to take this out of ICU and just place this in a less critical Recovery area.

There are minor glitches, and considering what we find in other models that have come thru the various operating rooms with different levels of disease - we have found a wide variety of different things put into the same "body" and it still works - to a degree.

Amazing. And Disgusting - at the same time - How do they get away with that? :mad::(:whistle::whistle::unsure:(y)(y)

Well, we are safe to keep the radio closer to stock so the Gentleman (men?) paying this tab, will get a radio that works and if we do this right - just might make the 45 PEP after all.
 
Well, a long time ham friend of mine came by yesterday just to chill with me. I was telling him about this radio and we got on the discussion of differences in meters. Later in the evening he invited me to his house and asked me to bring that radio and he would check it. He has a bird with a peak kit installed.

After he hooked everything up, low and behold the radio was swinging about 52 watts on his peak reading bird. Average was showing about the same as my 872. Carrier was showing about the same as my mfj meters.

After discussing further, he stated my 868 meter should show close to his bird peak and he would try to align it with his. Well he called me last night and told me the variable resistor adjustment for the 200 watt scale was bad and needed replacing, so he replaced it. He said that's what was stopping the output from being displayed properly, he said the output on the radio is fine. Now it shows close to what his bird showed.

Well, after I got home last night from his house I got to thinking about some things he told me, so I pulled out my old faithful mfj 3kw manual antenna tuner with a built in watt meter and built in 3kw dummy load. I hooked that same radio up to it and was shocked to see he was telling me the truth. On my mfj tuner watt meter it showed this radio keying about the same and swinging around 48 to 51 watts.

Again, everything is running cool to warm, even after hammering on it for 10 solid minutes, no overheating at all anymore.

Still not happy with that trimmer for am power adjustment. I did notice it has blotches or something in it cause it will stop turning in some places and displays erratic changes in dmm readings when adjusting.
 
  • Like
Reactions: Handy Andy
Thank you for letting me work alongside you to repair this...not often do we get success stories these days.

As far as that trimmer - might have to scavenge one from some other place. The trimmers used in the Galaxy radio are of low-wattage types. So to see a "beefier" (if that would be the word for it) wattage part also comes with it's own drawbacks - being linearity or scale of resistance to it's degree of rotation.

This may be one of what they call Logarithmic versus linear - only way to know is to take it out of circuit. More than likely is is the Linear type but the body color might signal that the part is of a different line of work - not necessarily precision - but that of durability - which doesn't always equate back to quality during the degrees of rotation.

That "Quality" refers to the "dropout" and dirt they show in specific spots making it difficult to trim properly. With it's composition design also raises questions about its stability of use in this application.

We do have a third "imaginary" person working here in this room with us - carrying a white elephant in with him...

Murphy...

He tends to remind us ...

"If it works - Don't Fix It."

So to me this is a stigma I will leave with you on the research, it tells me that this part may be the reason for the "odd" performance curve since there are so many others not showing this part, and work well.
 
Last edited:
  • Like
Reactions: Hawkeye351

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