Cobra 29LTD/GTL 10 KC mod????????

[brandon7861]

If you give up on that 5.12 crystal, send it my way, I'd love to take a shot at bending it. If I get it where you need it on a different PCB, I'll send it back. I think it just needs a little inductance to stretch it out a bit further. If you got the time, I got the postage.

 

[dss56]

PM me your address and I will send a few i got 50 of them for $5.00 from china.

 

[dss56]

I just finished up making my copy of the oscillator (with a 6.1704MHz crystal, its what I had) to see what affect my suggestion should have had.
View attachment 72431
Instead of a trimmer, I used a 10pF and a 30pF in parallel. Lifting the leg of the 30pF gives me the ability to go from 40pF to 10pF, which is the extremes of your trimmer. Also, from the emitter to the base, I have two 22pF caps. Lifting the leg of one allows me to go to/from 22pF to 44pF for C2.


Results:
*counter not warmed up and no crystal is the same, just seeing the direction and ballpark amount of change*

Trimmer 10pF, C2 22pF = 6.174354MHz
Trimmer 40pF, C2 22pF = 6.172502MHz
Trimmer 10pF, C2 44pF = 6.174215MHz
Trimmer 40pF, C2 44pF = 6.172262MHz

Doubling C2 lowered the freq 139Hz (22ppm) with the trimmer at minimum, and lowered it 240Hz (39ppm) with the trimmer maxed. The effect of increasing the trimmer is a decrease in frequency about 2.2kHz (356ppm).

Edit: I need to go to bed now, my brain is scrambled when I stay up this long. I see I was looking at a JFET circuit, not a more loaded down BJT oscillator. Once again, I made myself look like a total ass. I was thinking, OK, if 10% gets 30ppm, 100% will get about 300ppm. Being you said you got it to move 6kc (about 2/3 of 10kc), I figured, ok, about 3kc/3, or 1kc to go. 300ppm on 5.12MHz is 1536Hz, thought we had it in the bag. My apologies, I owe you a beer! I didn't realize I was looking at a JFET circuit with this particular chart.
View attachment 72433

here is a picture of the oscillator connected to my crystal/oscillator reader and it showes 5.1221mhz i will check it when its back in the circuit. Just gotta bend it a bit more for the full 10kc jump.

If 5.1221 is giving me about 6kc i figure if I can get it to read 5.1242 might do the job

 

[brandon7861]

Send a couple. Paralleling them is a common way to extend their tuning range. It would be one extra thing to try if they don't go willingly.

You could try that, to or three in parallel, before you send them to me if you had the time to experiment. It might be just enough.

 

[dss56]

Send a couple. Paralleling them is a common way to extend their tuning range. It would be one extra thing to try if they don't go willingly.

You could try that, to or three in parallel, before you send them to me if you had the time to experiment. It might be just enough.

I will give that a shot in a few BRB

 

[dss56]

Brandon thanks for suggesting to parallel the 5.120 as that works perfect made the 10kc dead nuts and all the channels 3-7-11-15-19 all become the a channels and if I add an expo kit all those missed will there.
Also because it shifted up it gives channel 41 27.415

Still would like to try the 5.000mhz xtal and see how that will work.

Using 2 xtals is a second choice but it works great nice and stable and no drifting
STILL SENDING THE XTALS IN 1 HOUR HAVE TO GO TO PO and are all packaged and labeled.

thanks dss56/Bob

 

[dss56]

Brandon7861

Well some good news and bad news.
Looking at the schematic pin 10 going to L22 and then if you add an Expo 100A kit to C87 they lead to the same L22.
So I tried a 100 Kit 40 up and down work great connected to c87.
Then the kit to the cut trace of pin 10 for the A channels. works great.
Problem is it will not shift the 40 up and 40 down to get the missed channels.
Would of been nice for that to happen.

thanks dss56

 

[brandon7861]

Brandon thanks for suggesting to parallel the 5.120 as that works perfect made the 10kc dead nuts and all the channels 3-7-11-15-19 all become the a channels and if I add an expo kit all those missed will there.
Also because it shifted up it gives channel 41 27.415

Still would like to try the 5.000mhz xtal and see how that will work.

Using 2 xtals is a second choice but it works great nice and stable and no drifting
STILL SENDING THE XTALS IN 1 HOUR HAVE TO GO TO PO and are all packaged and labeled.

thanks dss56/Bob

Cool!

I am fairly sure I can make it work with just one crystal, looking forward to trying. I have a few parts radios with SSB so I should have a decent selection of tuning cans.

With all the pulling tricks thrown at the 5MHz ones, I would be surprised if we get more than 4 or 5 ppt (20 or so kHz pull). Even that is asking a lot and it flirts with stability issues. Going from 5 to 5.1xx just isn't possible.

If you plan to operate them around their extreme of 5.020, that would put you in 25MHz territory and all your IF cans will be WAY off. The output network would be out of whack at those frequencies too and your final might switch to sending smoke signals (truth is, I don't know, not modeling it unless a need arises). Be careful trying those.

 

[dss56]

Well thanks for your help and suggestions on this post.

Let me know what you come up with.

Bummer that i cant get the missed channels with the expo 100A kit also connected to c87. Can only get the 3-7-11-15-19A channels.

Well good luck again thanks and I will try the 5 mhz xtals when I get them

 

[brandon7861]

Crystals arrived, just picked them up on my way home from work.

I was thinking you had HC49U's, but these little HC49S crystals are something I've used only once in my life. This may be a job for a Jfet oscillator. I will mess around with them over the weekend and see what they (I) can do. My understanding is that they are harder to pull the smaller the package, so maybe with the higher input impedance of the JFET, they might go further. I hope.

 

[dss56]

Crystals arrived, just picked them up on my way home from work.

I was thinking you had HC49U's, but these little HC49S crystals are something I've used only once in my life. This may be a job for a Jfet oscillator. I will mess around with them over the weekend and see what they (I) can do. My understanding is that they are harder to pull the smaller the package, so maybe with the higher input impedance of the JFET, they might go further. I hope.

Thanks Brandon I should have the 5.000mhz xtals next few days they are the HC49u's and will see if i can bend it enough as u are trying to do with the xtals I sent you. Good Luck have a great weekend gonna be cold and rainy here in Boston.

 

[brandon7861]

Quick update, I did manage to get it above the target frequency of 5.123333MHz, but it took using a JFET and having a very small load capacitance. Maybe with some voltage regulation and shielding we could get away with this, but I think we should stick with paralleling two and hope it lets us increase the load capacitance. I left room on this board to do so. With just one crystal, I didn't have a trimmer small enough, so gimmick it was. I will pick it up tomorrow.

 

[dss56]

Quick update, I did manage to get it above the target frequency of 5.123333MHz, but it took using a JFET and having a very small load capacitance. Maybe with some voltage regulation and shielding we could get away with this, but I think we should stick with paralleling two and hope it lets us increase the load capacitance. I left room on this board to do so. With just one crystal, I didn't have a trimmer small enough, so gimmick it was. I will pick it up tomorrow.
View attachment 72513
View attachment 72514View attachment 72515

Brandon good work im still waiting for the 5.000mhz xtals. I did try other options and seems the dual xtals is working. I tried in a few cobra 29s and all worked the same.

 

[brandon7861]

Brandon good work im still waiting for the 5.000mhz xtals. I did try other options and seems the dual xtals is working. I tried in a few cobra 29s and all worked the same.

I added a second crystal to this JFET oscillator and it raised the trim capacitance to something around 1.75pF. I then discovered that putting my hand anywhere enar it changed the frequency, so I grounded the crystal cases. That really helped with stability, but it once again reduced the required trim capacitance down to about 1pF.

The best trimmer I have only goes down to 3.5, so putting it in series with 1pF got me in the ballpark, but the instability came back. So I once again replaced the trim cap and series 1pF with another gimmick cap and I had it running at 5.123 333 for the entire night nice and solid with only couple tens of Hz shift getting my finger on the gimmick.

So today, I have plans to make a coaxial trim capacitor (to shield it) using a piece of telescopic antenna and borosilicate pipette as the dielectric, but instead of moving the metal, I plan to move the glass pipette to vary the frequency. Going from glass to air dielectric in my prototype achieved 1.3 to 4.8pF. I need to cut about 10% off its length so it fits on the board, then I will cut slots in the outer sleeve to further lower the capacitance. I figured we can wax the glass tuning rod in place once it is set.

I am curious, did you have stability issues using two crystals in your BJT oscillator? I tried the same 2 crystals and gimmick cap in my expo clone and the frequency dropped about 1kHz, couldn't get it quite high enough, so I am really surprised it is working for you. These little crystals are putting up a good fight.

Heres the cap, I plan to install it north of the crystals after I take the gimmick off.

I also need to add a buffer since changing my probe from 10x to 1x shifted the frequency a little. I was almost tempted to take advantage of that to trim it on freq.

As is, with the gimmick over the crystals, raising and lowering the gimmick from the crystal cases changes the frequency enough to not have to play with the twists, I just don't think it would be happy in a mobile environment.

 

[dss56]

I added a second crystal to this JFET oscillator and it raised the trim capacitance to something around 1.75pF. I then discovered that putting my hand anywhere enar it changed the frequency, so I grounded the crystal cases. That really helped with stability, but it once again reduced the required trim capacitance down to about 1pF.

The best trimmer I have only goes down to 3.5, so putting it in series with 1pF got me in the ballpark, but the instability came back. So I once again replaced the trim cap and series 1pF with another gimmick cap and I had it running at 5.123 333 for the entire night nice and solid with only couple tens of Hz shift getting my finger on the gimmick.

So today, I have plans to make a coaxial trim capacitor (to shield it) using a piece of telescopic antenna and borosilicate pipette as the dielectric, but instead of moving the metal, I plan to move the glass pipette to vary the frequency. Going from glass to air dielectric in my prototype achieved 1.3 to 4.8pF. I need to cut about 10% off its length so it fits on the board, then I will cut slots in the outer sleeve to further lower the capacitance. I figured we can wax the glass tuning rod in place once it is set.

I am curious, did you have stability issues using two crystals in your BJT oscillator? I tried the same 2 crystals and gimmick cap in my expo clone and the frequency dropped about 1kHz, couldn't get it quite high enough, so I am really surprised it is working for you. These little crystals are putting up a good fight.

Heres the cap, I plan to install it north of the crystals after I take the gimmick off.
View attachment 72533
I also need to add a buffer since changing my probe from 10x to 1x shifted the frequency a little. I was almost tempted to take advantage of that to trim it on freq.

As is, with the gimmick over the crystals, raising and lowering the gimmick from the crystal cases changes the frequency enough to not have to play with the twists, I just don't think it would be happy in a mobile environment.

I did not have stability issues as I removed from 1 radio and tried it in 2 other cobra 29s and still worked well right on frequency.