Help troubleshoot mosfet install in a Uniden 78/Cobra 29?

[ScurvyPete]

Allow me to start off by saying that I know there are many of y'all out there who prefer the bipolar transistors to the mosfets, but I'm not interested in that argument right now as much as I am trying to work out the problems I've created for myself in the name of learning new things.

I've been in the process of swapping a 13N10 mosfet into my Uniden 78; per the CB Tricks/EKL instructions and using the EN1230 companion part.
I removed R55 R56 and C61, soldered in the 1230 part in R56's place (in the correct direction), and left R55 and C61 "open". I did not solder in the 68pf cap on the backside of C59 (mainly because I didn't have one on hand, and because it said that the older-model Cobra 29's and 78's may not need them).
After install, my radio was putting out a 1W deadkey and swinging 7-8 PEP, and around 2W avg power.
I went back to check over my work again, and noticed I had been a little sloppy with the solder, and had bridged the gate with one leg of C60 and the drain with one leg of the L15 coil. I desoldered these connections and resoldered cleanly. However, in the process, the little foil piece that the drain pin solders to on the PCB came up, and now I can't get it to connect to the board. The radio will turn on, will receive, and the TX/RX light changes when the mic is keyed, but I have no output power at all.

To sum this up, here are my questions:
1.) How do I reattach the foil to the PCB so I can connect the drain pin of the mosfet?
2.) Should I have left that solder bridge between the mosfet and C60 and L15 alone?
3.) Surely the absence of the 68pf cap wasn't keeping the radio to only 1W DK and 7-8W PEP/2W avg?

Sorry for the wall of text, and thank you for any help

 

[2FB327]

First question, what did the radio do before, did it work and is the wattmeter accurate? The reason I ask is that when I first started I could only get my Dosy to read about 25pep, which wasn't even close to the 38 they were doing on a bird.

I have done a few of these, I have never had any luck with the "companion component" I have experimented with others also and what they don't tell you is that you have to run a resistor to the drain pin to get the voltage right. If you run a search on this site, you'll find a lot of old post, mainly mine trying to get it right.

As far as lifting the foil, I'm good at that, the only fix that I know of that lasts is to take a piece of wire or the wire end from a resistor and "bridge" to another pin in that trace.

 

[sonoma]

you should be getting about 3.5 to 3.8 volt at the gate for the mosfet to work right. see what you are getting.

 

[Robb]

Isn't the max gate voltage for the 13n1o lower than the irf-520?
IIRC, it is 3v/max, while the irf-520 is 5v/max..

 

[ScurvyPete]

First question, what did the radio do before, did it work and is the wattmeter accurate? The reason I ask is that when I first started I could only get my Dosy to read about 25pep, which wasn't even close to the 38 they were doing on a bird.

Before was about 3.5-4W deadkey swinging to around 15PEP and somewhere between 4-5W RMS.
As far as meter accuracy, it's a crapshoot. I've been using a Dosy I bought off Ebay, but I'm getting pretty damn close to the same numbers off my little RadioShack SWR/Wattmeter that I used to reference against. So while it may not be "Bird watts", two different meters showing just about the same number tells me that the Dosy isn't greatly under or over estimating the wattage.
Also, I'm aware of everyone's general feelings of discontent with the Dosy meters, and I don't think they're the greatest thing ever or super accurate. I just wanted something to test with to get me "in the ballpark" and with a scale that I could see better than that little RadioShack meter.

I have done a few of these, I have never had any luck with the "companion component" I have experimented with others also and what they don't tell you is that you have to run a resistor to the drain pin to get the voltage right. If you run a search on this site, you'll find a lot of old post, mainly mine trying to get it right.

I'll run another search shortly. I feel like the mosfet mod in general is one of those topics that has been somewhat beat to death and I hate to add to the clutter, but I also seem to always be dredging up a ton of information that isn't exactly pertinent to what I want to know, either.
What do you mean by not having any luck with the companion part? And any tips for running the resistor to the drain pin for correct voltage?
Like I said, I'll run some more searches. I'm just trying to get some help to point myself in the right direction.

 

[ScurvyPete]

you should be getting about 3.5 to 3.8 volt at the gate for the mosfet to work right. see what you are getting.

I'm assuming that I need to be transmitting to check this?

 

[sonoma]

the 13n10 takes the same voltage as the 520 . the 13n10L only takes 1.8 volts to run.

scurvypete you just need a dead key to check the voltage. and most guys build their own companion part. at least I do to be able to play with the voltages.

 

[Robb]

the 13n10 takes the same voltage as the 520 . the 13n10L only takes 1.8 volts to run.

Thanks, sonoma.

 

[Robb]

On the 'non-L' 13n10 MOSFET, the gate has a minimum of 2v and a max of 4v.

https://www.fairchildsemi.com/datasheets/FQ/FQP13N10.pdf

So, if you set the gate voltage at the typical 3.8v, it should be just fine.

 

[sonoma]

if the gate is too low it will not turn on as it should and have a low output. that is why I just say 3.5 to 3.8 will do just fine. the 13n10L is shipped to people as a 13n10 and it will pop with over 1.8 volts.

 

[ScurvyPete]

Well, I have a solder jumper between the drain and L15 (again) and I'm transmitting. I put the red lead of the multimeter on the gate pin, and the black lead on a screwhead showing on the solder side of the board (I wasn't really sure where I was supposed to ground to). Doing this with the deadkey gave me a reading of around 7.5 on the 20 DCV scale.
From what I've been reading, this seems way too high, but I don't know that it's an accurate reading because I'm not sure I used the correct procedure to measure the gate voltage, and the RF output I'm seeing is still way low (but with a lot of swing).
Right now I'm working with a deadkey somewhere between .5 and 1W and swinging to almost 10W PEP and 2-3W RMS with a decently loud "AHHHHHH". So my situation is pretty much the same, but the numbers are just slightly different.

I studied on it for a good while, and it seemed like the leg of the L15 coil was on the same trace as the drain pin, so that was what I soldered to (and it was the closest component to the pin). I'm wondering if this was the wrong thing to do or not.

Thank you guys for all the good info and help you've provided me so far. I really appreciate it.

 

[Robb]

MOSFETs are pretty unforgiving. If there is just a little too much voltage on the gate, then it will pop without hesitation. You might want to examine the spec sheet link I posted on post #9 above. It will show you which leg of the MOSFET is the gate. All you do is put your '+' lead of your voltmeter there, and the '-' lead to the chassis ground. It is probably OK; just so long as it is not above 4v you will be OK,

1w swinging 10w?
Sounds like you have a swing mod installed . . .

 

[ScurvyPete]

MOSFETs are pretty unforgiving. If there is just a little too much voltage on the gate, then it will pop without hesitation. You might want to examine the spec sheet link I posted on post #9 above. It will show you which leg of the MOSFET is the gate. All you do is put your '+' lead of your voltmeter there, and the '-' lead to the chassis ground. It is probably OK; just so long as it is not above 4v you will be OK,

1w swinging 10w means you have a swing mod installed . . .

According to this, when looking from the front of the radio to the back, and with it upside down (because we're on the solder side), the gate would be the leg on your right, correct? That's the one that I checked.
Your comment about the lack of "wiggle room" with mosfets is another reason that I don't believe I'm getting the correct voltage reading. Surely this thing would have popped with 7.5V at the gate.

As far as the swing kit, it's possible. My girlfriend was at a flea market and saw the radio and texted me to see if I wanted it. I told her to jump on it, and she managed to snag the radio & power cord, a cheap truckstop power mic, magnet mount antenna, and external speaker for $30.
When I looked inside, the tuning slugs still had the wax seal on them, nothing looked clipped, and the solder side looked pretty factory to me. Of course, I'm by no means an expert, so there is a definite possibility that I missed something.
As far as I know there was nothing wrong with the radio, and this was a mod I just wanted to try for the sake of learning something new

 

[sonoma]

you want to put the ground on one of the tuning cans or the ground of the circuit board. not a screw head or the metal case of the radio. you can even hook it to the ground wire coming into the radio. this way you will get a true voltage reading.

 

[ScurvyPete]

I've been reading this thread: IRF-510 in a Cobra 29 Impressions

To quote the original post from CTStallion, and compare my steps (in blue)-

So, here is the procedure I used (some of it follows my old 2SC1969 days:

1. Took the radio out of the box, connected it to a 13.8V supply, Wawasse Catalyzer JBC-1000-SM & dummy load, and commenced taking initial readings. Then I hooked it up to a peak reading wattmeter & modulation meter and finished initial tests.

2. Removed radio covers and adjusted AMC for maximum modulation.
I didn't adjust to max, but I did turn the VR4 pot up to about 3/4 (versus the 1/2 setting that it was on)

3. Removed TR14 (the stock 2078 final) and the following additional components: JP36, R55, R56, C61, & L14'S tuning slug.
Removed the final, R55, R56, and C61. Did not remove JP36 or the tuning slug in L14

4. Fabricated a 'discrete' EN-1230 component part using 1 X 1N4148/1N914 diode + 1 X 470 ohm 1/2 watt carbon resistor + 1 X 1/2 watt 3.3K ohm resistor. Placed same (with positive lead facing toward R123) in place of the removed R56.
I just used an EN-1230 component

5. Installed a 47 ohm 1 watt carbon resistor across a 220 uF 35VDC electrolytic capacitor, with negative facing final in place of the removed JP36.
This is really the only step in the actual "parts change" that I didn't do

6. Powered up the radio, turned the dynamike control to MINIMUM, and made note of the dead key wattage.

7. Depowered radio and removed the 33 pF factory cap installed across C62 & replaced it with a 100 pF cap (to see if it would raise my dead key... it DIDN'T....) so I reinstalled the 33 pF cap across C62
I didn't do this, but apparently it was pointless anyway so I'm not worried about trying it

8. Experimented with different capacitor settings across C59 (mod calls for 68 pF and I didn't have one.) Settled for a 39 pF Mica which provides approx 309 pF of total capacitance... (higher values reduced my peaks and made my audio with the stock mike sound crappy.) But you can completely skip this addition on many radios, or merely add the called for 68 pF of additional capacitance.
I did not do the 68pF cap (or any other cap) across C59, and I hadn't really planned on it either since it said that most radios don't need it. Since CTStallion seems to agree, I think I'm going to leave this step alone

9. Connected the radio (sans covers) back to the test equipment & dummy load and turned the Dynamike control back up to MAX.

10. While saying: 'HEEEEEEEEEEELLLLLLLLLLLLLLOOOOOOOOOOOO, & AUUUUUUUUUUDDDDDDDDDDDIIIIIIIIIOOOOOOOOOOOO, and/or whistling into the mike, adjusted forward power for maximum via L12, 13, 17 & 20.
I've been waiting to "fine tune" until I really get this mod sorted out correctly, so this hasn't been done yet

11. Checked for harmonics in 54 MHz band with a separate receiver... minimal.
Not really sure how I'm going to check for this

Wanna know if it worked?????

Here's the BEFORE and AFTER RESULTS: (Before was a STOCK radio)

BEFORE: RF Output Deadkey Approx 3.2 watts AFTER: Approx 2.5 watts
BEFORE: RMS Swing Watts Approx 3.5 watts AFTER: Approx 12.5 watts
BEFORE: Percent AM Modulation 80% dynamike at max AFTER: 100+++ Percent... Pins meter on whistles
BEFORE: Peak (PEP) Swing Watts Approx 9.1 PEP AFTER: Almost 40 Watts!!!

Some other thoughts/comments on this-
I ordered some 47-ohm 1W carbon resistors and 220 uF 35VDC aluminum leaded electrolytic caps from Mouser today. I'm hoping using these at JP36 will help get this mod sorted out. Other than that (and his cap experiments), I've pretty much taken the same steps he has.

Sonoma, I'm not sure if these are the same modifications you used to get this mod to work for you. When I searched for "cobra 29 mosfet 2030", it returned less than 10 results and none were threads by you. I also went through your list of threads shown on your user profile, and missed anything about a mosfet swap. I'm not saying it's not there, I just haven't found it.

Do y'all think this cap/resistor mod at JP36 will help sort me out? Or am I missing something here?

Edited to add: As far as the L14 slug goes, I've read conflicting instructions here. Some say that you should just ditch the slug (EKL's instructions). Some people say you should leave the slug in to fine-tune the mod (although you need to have 4 wraps of the coil instead of 5; mine has 4).
I'm more inclined to leave it in, but I broke the slug in half trying to adjust it and now it's pretty much stuck in position. Any tips or tricks on how to remove it?