I just cracked open my first 148GTL and noticed the final has a bias driver, tr37. I'm not sure why. Could someone key me in please? I see plenty of am/ssb, 2166/1969 rigs that don't have one, so what is the deal?
-
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.
-
A Winner has been selected for the 2025 Radioddity Cyber Monday giveaway! Click Here to see who won!
Cobra 148GTL, why bias driver?
- Thread starter Cable Guy
- Start date
Found this from nomad...
Post in thread 'Pix Request, 148 GTL / Grant XL' https://www.worldwidedx.com/threads/pix-request-148-gtl-grant-xl.209400/post-578269
Excerpt
"The final bias diode type MV13Y is a bit different. It has TWO diodes in series inside it. Forward voltage drop is about twice what the driver's diode will be. The reason for this is that the diode is not connected directly to the final's base circuit. The MV13Y sets the voltage on the base of a NPN transistor TR37. The emitter voltage coming out of TR37 is what feeds bias current into the base leg of the final transistor. TR37 provides a current boost, and isolates the MV13Y bias diode from the final's drive signal. The voltage drop feeding into the base of TR37 and out from its emitter leg is the same 6/10 of a Volt as a normal single diode, and gets us the same remaining 6/10 of a Volt feeding into the final's base terminal.
The MV13Y on the final ALMOST never goes bad, unless one of the wire leads falls off. TR37 also isolates it from transient voltages that could happen if the final breaks down. Other types of radio that connect the bias diode directly to the final will frequently blow out the bias diode when the final fails. THIS radio almost never does that."
That's great, but now I'm wondering why this isn't implemented in other rigs. Maybe cost?
Post in thread 'Pix Request, 148 GTL / Grant XL' https://www.worldwidedx.com/threads/pix-request-148-gtl-grant-xl.209400/post-578269
Excerpt
"The final bias diode type MV13Y is a bit different. It has TWO diodes in series inside it. Forward voltage drop is about twice what the driver's diode will be. The reason for this is that the diode is not connected directly to the final's base circuit. The MV13Y sets the voltage on the base of a NPN transistor TR37. The emitter voltage coming out of TR37 is what feeds bias current into the base leg of the final transistor. TR37 provides a current boost, and isolates the MV13Y bias diode from the final's drive signal. The voltage drop feeding into the base of TR37 and out from its emitter leg is the same 6/10 of a Volt as a normal single diode, and gets us the same remaining 6/10 of a Volt feeding into the final's base terminal.
The MV13Y on the final ALMOST never goes bad, unless one of the wire leads falls off. TR37 also isolates it from transient voltages that could happen if the final breaks down. Other types of radio that connect the bias diode directly to the final will frequently blow out the bias diode when the final fails. THIS radio almost never does that."
That's great, but now I'm wondering why this isn't implemented in other rigs. Maybe cost?
Thanks. Just one more question, maybe. This one has a 2sc3419 in place of tr37 whereas the schematic calls for a 2sc496. Now there is also a 68 ohm resistor strapped from base to ground of tr37. I don't have a driver or final installed yet, and it seems to adjust correctly with .680 VDC about midway on the bias trimmer. Voltages at this transistor are very near spec. I pulled the added 68 ohm resistor off and it adjusts from about 1.2 to 1.6vdc, way too much. I feel someone stuffed a quasi-equivalent transistor in there and added the resistor to compensate for the added gain of the wrong transistor. Should I leave it as I found it or find the correct transistor? Everything else looks good, and I tacked in a temp driver/final just to check current and wattage.
This rig came to me with 2 fake ebay 1969's installed as the final AND DRIVER with no output. I didn't have a spare 2166 for the driver, but found a nte235 in the parts bin and an eleflow 1969(cb-20). They got it back up. I see no other mods, just the tr37 and 68 ohm stuff.
This rig came to me with 2 fake ebay 1969's installed as the final AND DRIVER with no output. I didn't have a spare 2166 for the driver, but found a nte235 in the parts bin and an eleflow 1969(cb-20). They got it back up. I see no other mods, just the tr37 and 68 ohm stuff.
Last edited:
The 68Ω resistor, is it from the base of TR37 to ground, or from the base of TR36 to ground? And where are you measuring those voltages, the base of TR37 or the base of TR36?
If the pot is set halfway (2.5k), having 68Ω to ground there would drop the voltages well below what you are reading.
If the pot is set halfway (2.5k), having 68Ω to ground there would drop the voltages well below what you are reading.
Yes, 68 from ground to base of tr37, and voltages at tr37 are spec. Not exactly half way. Yes, at that position, I get ~.689 VDC at ~50ma on the base of the final, tr36.
The radio is back together and performing well, this was more of an academic question regarding the 68 ohm. I have seen a post here (https://www.worldwidedx.com/threads/cobra-148gtl-a-k-a-texas-star-tr-296.233631/) referencing a 68 ohm on tr37 and wasn't sure why this one had one, then I realized there was a different transistor in that spot, a part rated for faster turn on, higher frequency and higher gain than the general purpose transistor specified.
I will open it back up tomorrow if you want to dig in and figure it out.
I will open it back up tomorrow if you want to dig in and figure it out.
In the post you linked to, there is mention of the Ranger TK-296GK. In that radio, the pot is 1k and the resistor is 47kΩ (instead of 68Ω). In your case, I think 68Ω is appropriate.
From your picture above, it appears your 1k pot is set to about 30% from the R179 (100Ω) leg, so if you have 8.3v from the regulator going into a divider made of 100Ω, 300Ω and 68Ω, the voltage across the 68Ω resistor would be 8.3*(68/468)=1.2v. If the MV13Y ever fails, the voltage cannot go higher than this.
Having no resistor there would cause the voltage at the base (and thus the finals base) to skyrocket if the MV13Y fails, burning out the final. This resistor prevents that as it ensures a failed diode cannot cause a sharp increase in voltage.
If I had to guess, i'd say the 47kΩ shown in the 296GK schematic is one of those intentional errors to see who copies their circuit. That is one of those components that can be wrong and the radio still work normally (as long as that diode holds up).
I think the reason you see so much more voltage with the 68Ω removed was due to two things, the higher current going through the diode forcing it's Vf up and no load being on the collector of TR37 (IIRC, this was when the final was out) forcing its Vbe down.
Andy also mentioned in that post to verify R181 wasn't burnt. You should do that if you haven't already.
From your picture above, it appears your 1k pot is set to about 30% from the R179 (100Ω) leg, so if you have 8.3v from the regulator going into a divider made of 100Ω, 300Ω and 68Ω, the voltage across the 68Ω resistor would be 8.3*(68/468)=1.2v. If the MV13Y ever fails, the voltage cannot go higher than this.
Having no resistor there would cause the voltage at the base (and thus the finals base) to skyrocket if the MV13Y fails, burning out the final. This resistor prevents that as it ensures a failed diode cannot cause a sharp increase in voltage.
If I had to guess, i'd say the 47kΩ shown in the 296GK schematic is one of those intentional errors to see who copies their circuit. That is one of those components that can be wrong and the radio still work normally (as long as that diode holds up).
I think the reason you see so much more voltage with the 68Ω removed was due to two things, the higher current going through the diode forcing it's Vf up and no load being on the collector of TR37 (IIRC, this was when the final was out) forcing its Vbe down.
Andy also mentioned in that post to verify R181 wasn't burnt. You should do that if you haven't already.
Last edited:
Ok, thanks. I did verify the resistors and capacitors referenced in Andy's post and all are ok, no shorts, burns or opens, and are as expected. This radio doesn't match either schematic. So, the divider formed between the 8v source, the 100 ohm, the 1k VR, and the 68 ohm provide a voltage that varies from .5 to 3.3 volts and the mv13y provides a dual diode junction to gnd not allowing the voltage to actually reach anything above ~1.3 volts. I was assuming a 5k vr because I didn't actually look, just went by the schematic.
Had it been a 5k pot, a 68 ohm resistor would hold the voltage too low to turn anything on.
I had a 148GTL that I put inside a Stalker 2 for a friend. I remember needing to use the 296GK alignment procedure because the VCO frequencies didn't match the 148GTL service manual. Now I am wondering, were your VCO frequencies 34.76xx or 34.98xx?
This vco runs at 34.985. I feel like the different revisions were to keep up with the competition, could explain why this rig is a mashup of both those schematics. Maybe ranger did it a little better, so cobras next revision included those changes and the official schematic didn't get changed.
Sorry, I'm ignorant at times, overlooking obvious things. I could do better if I could quiet my mind and focus, but with a little time it all comes back. Some days my mind is buzzing with unnecessary thoughts and I couldn't even fit together a 10 piece puzzle, some days I can do anything. I have a saying, "My idle thoughts always lead me on a journey through fascinating obscurity."
I wonder which xtal it had in it, i think the mb8719 could have had 2 different xtals to get the needed frequencies. I think either a 11.1125 Mhz or 11.3258 MHz. This particular rig had the latter. Idk if it matters.