I came across this old GE J-805 radio at a local goodwill and am in the process of restoring it. Someone has replaced the power transormer with a TINY thordarson, rated for only 260-0-260 @ 60m ma! 
way too small for push pull 6v6's!
Anyway, I am going to be doing extensive mods to the power output section to obtain more power.
here is the original schematic:
http://www.nostalgiaair.org/PagesByModel/115/M0008115.pdf
I am so confused as to how the phase inverter works on this! Wouldn't a more conventional cathodyne work better, and still be able to utilize the 6J5? I could always go to something like a 6SN7 or something instead.
Also, the output transformer is laughably tiny..the size of a fender champs OPT! I have a hammond 1650F laying around begging to be put in there, as well as a 270fx for the missing power trany. This will be a franken-modded radio that at least appears it came this way from the factory. The radio was in verypoor shape when I got it, so no guilty consience modifying it. I consider it "giving it new life"
way too small for push pull 6v6's!Anyway, I am going to be doing extensive mods to the power output section to obtain more power.
here is the original schematic:
http://www.nostalgiaair.org/PagesByModel/115/M0008115.pdf
I am so confused as to how the phase inverter works on this! Wouldn't a more conventional cathodyne work better, and still be able to utilize the 6J5? I could always go to something like a 6SN7 or something instead.
Also, the output transformer is laughably tiny..the size of a fender champs OPT! I have a hammond 1650F laying around begging to be put in there, as well as a 270fx for the missing power trany. This will be a franken-modded radio that at least appears it came this way from the factory. The radio was in verypoor shape when I got it, so no guilty consience modifying it. I consider it "giving it new life"
Hi ThSpeakerDude88,
The 6J5 is an excellent tube. You really must get a copy of Morgan Jones' "Valve Amplifiers", and possibly the companion book "Building Valve Amplifiers". They are an easy read and full of great information. At least get the first book. Available at Amazon.
Now for your circuit. Try the original, then try your idea. This is a great opportunity to learn first hand. The first part of your experiment is already built, just restore it.
Now, are you planning to rebuild it and improve performance while you are at it? That's what it sounds like to me. If this is the case, the caps get replaced. Use normal plastic foil types, there is no need to go crazy here. You may find that many resistors have increased their value a fair amount. Besides, they are carbon composition. They go too. This will probably get you to most of where you want to go. Next, listen to it. Take distortion measurements if you can. Then rebuild the phase splitter and compare the performance, both by measurements and listening (do this first so you have no expectations).
-Chris
The 6J5 is an excellent tube. You really must get a copy of Morgan Jones' "Valve Amplifiers", and possibly the companion book "Building Valve Amplifiers". They are an easy read and full of great information. At least get the first book. Available at Amazon.
Now for your circuit. Try the original, then try your idea. This is a great opportunity to learn first hand. The first part of your experiment is already built, just restore it.
Now, are you planning to rebuild it and improve performance while you are at it? That's what it sounds like to me. If this is the case, the caps get replaced. Use normal plastic foil types, there is no need to go crazy here. You may find that many resistors have increased their value a fair amount. Besides, they are carbon composition. They go too. This will probably get you to most of where you want to go. Next, listen to it. Take distortion measurements if you can. Then rebuild the phase splitter and compare the performance, both by measurements and listening (do this first so you have no expectations).
-Chris
BTB, thanks for making sence of it for me. I tried redoing the schematic but I didnt recognize it as a paraphase until I realized that the 6SQ7 triode drives the top half of it. Will a cathodyne have better capabilities to drive the tubes to greater output power?
Anatech, when I rebuild it I will replace all coupling caps and OOT resistors , etc. Standard restoration stuff. I am not greatly concerned with ultra-hifi specs, rather something that sounds good and powerful through the primative short throw 12" 70 year old loudspeaker. I think that 10 watts from the 6v6's is doable with good distortion specs.
As for the book, yes I do need to order that. I currently have TUT1, and while it is for guitar amps, it does cover a lot of usefull information about PI's and other essentials of building a tube amp.
Anatech, when I rebuild it I will replace all coupling caps and OOT resistors , etc. Standard restoration stuff. I am not greatly concerned with ultra-hifi specs, rather something that sounds good and powerful through the primative short throw 12" 70 year old loudspeaker. I think that 10 watts from the 6v6's is doable with good distortion specs.
As for the book, yes I do need to order that. I currently have TUT1, and while it is for guitar amps, it does cover a lot of usefull information about PI's and other essentials of building a tube amp.
You might be able to use the 6J5 as a cathodyne splitter. The way the 6J5 is used in the present circuit, it has a gain of 1 (nominally) so that all it does is invert the phase. A concertina will not change the gain situation materially, but could give excellent balance. However, there are a couple of potential snags.
A cathodyne splitter will not work satisfactorily if the B+ for the 6J5 is not high enough, because it will constrain the available swing available to drive the 6V6s to full output. You will only have swing of amplitude about 1/4 of B+ wth a cathodyne, instead of 1/2 of B+ with the paraphase.
Another downside to a cathodyne splitter is that it is sensitive to the load, which must be balanced. It's not too happy driving tubes operating in Class AB, which is what those 6V6s are almost certainly doing.
One or other of these is more than likely the reason why the rather inferior paraphase splitter was used in the first place.
A cathodyne splitter will not work satisfactorily if the B+ for the 6J5 is not high enough, because it will constrain the available swing available to drive the 6V6s to full output. You will only have swing of amplitude about 1/4 of B+ wth a cathodyne, instead of 1/2 of B+ with the paraphase.
Another downside to a cathodyne splitter is that it is sensitive to the load, which must be balanced. It's not too happy driving tubes operating in Class AB, which is what those 6V6s are almost certainly doing.
One or other of these is more than likely the reason why the rather inferior paraphase splitter was used in the first place.
Not really
Kinda looks like the tube is inverting just one side of the wave being sent from the previous tube. It's a low gain stage (based on the cathode the plate resistor values) and the input to the tube (from previous stage) is attenuated (R20,21,22) going into the tube, to keep the drive balanced to the power tubes.
Since there is no negative feedback around the output stage you don't need much gain to get a reasonable power output. Replacing that tube with a 6SN7 or 6SL7 and making a true phase inverter might work better. You could add NFB as well. The previous stage should have enough drive for that to work. If you do this, you may wish to remove the Zobel network on the output transformer primary, which is there for stability, but may not be required with a better driver circuit.
Kinda looks like the tube is inverting just one side of the wave being sent from the previous tube. It's a low gain stage (based on the cathode the plate resistor values) and the input to the tube (from previous stage) is attenuated (R20,21,22) going into the tube, to keep the drive balanced to the power tubes.
Since there is no negative feedback around the output stage you don't need much gain to get a reasonable power output. Replacing that tube with a 6SN7 or 6SL7 and making a true phase inverter might work better. You could add NFB as well. The previous stage should have enough drive for that to work. If you do this, you may wish to remove the Zobel network on the output transformer primary, which is there for stability, but may not be required with a better driver circuit.
Hi ThSpeakerDude88,
That sounds reasonable, I was not intending to have you reach for the top. It just seems like a good opportunity for you to learn something first hand.
Good luck on your restoration!
-Chris
That sounds reasonable, I was not intending to have you reach for the top. It just seems like a good opportunity for you to learn something first hand.
Sure, but how much extra are a couple metal oxide resistors going to be? May as well get rid of those carbon things while you're there. This has nothing to do with super high fidelity at all. While you are at it, watch the voltage rating on the resistors you use. You may need to go to a 2 watt simply to get the voltage rating you need. 250V is not uncommon for a 1/2 watt resistor, and I think the older carbon ones where rated higher. You should check me on that as I'm not 100% sure that is correct.
Good luck on your restoration!
-Chris
In all of my guitar amp designs I use metal oxide and metal film resistors that are rated for 600v and have had no problems. You're probably right, I should replace all the carbon resistors (which happen to be the old carbon tube kind.) Its just a bit of a hassle though. I typically use 1w-2w metal oxide in the power supply and any high current areas, and 1/2w metal film.
Fuchs, You brought up a good point. The original circuit used no NFB, and I don't know if I should use any or not. I think I probably am going to end up replacing the 6J5 with a 6SN7. A few inverters come to mind, schmidt and LTP...
Raymoth, 1/4 of 280v is 70v, this is more than enough to drive 6v6's, which demand somewhere around 20v p-p, is it not? Or do you mean 1/4 of Plate B+? Not saying cathodyne is the best choice, but some famous amps that sound pretty good have used it. The ST35 comes to mind.
Oh, forgot to mention, the 1650F has 40% UL taps, so UL mode is definatly an option here and eliminates the need for a regulated screen node.
Too bad there isnt an RH84 PP version with some sort of LFB.
Fuchs, You brought up a good point. The original circuit used no NFB, and I don't know if I should use any or not. I think I probably am going to end up replacing the 6J5 with a 6SN7. A few inverters come to mind, schmidt and LTP...
Raymoth, 1/4 of 280v is 70v, this is more than enough to drive 6v6's, which demand somewhere around 20v p-p, is it not? Or do you mean 1/4 of Plate B+? Not saying cathodyne is the best choice, but some famous amps that sound pretty good have used it. The ST35 comes to mind.
Oh, forgot to mention, the 1650F has 40% UL taps, so UL mode is definatly an option here and eliminates the need for a regulated screen node.
Too bad there isnt an RH84 PP version with some sort of LFB.
Hi ThSpeakerDude88,
Where do you get your metal oxide resistors from? The ones I get need to be 3 watt before ratings like 450 VDC are reached. The 1/2 watt units are 250 ~ 300 VDC only. Same goes for the metal film resistors.
600 VDC resistors tend to be expensive. If I need to drop that much voltage, I'll typically use a string of 2 ~ 4 resistors depending on what I'm doing.
-Chris
Where do you get your metal oxide resistors from? The ones I get need to be 3 watt before ratings like 450 VDC are reached. The 1/2 watt units are 250 ~ 300 VDC only. Same goes for the metal film resistors.
600 VDC resistors tend to be expensive. If I need to drop that much voltage, I'll typically use a string of 2 ~ 4 resistors depending on what I'm doing.
I thought you weren't interested in trying for very good quality. Those Hammond transformers are reasonably good.
-Chris
I get my resistors from either tubedepot or AES.
The metal oxide 1w resistors from AES are 350wvdc 600v overload, their metal film 1/2w ones are 350wvdc 700v overload
The hammond: I have one laying around and don't want it to go to waste.I thought why not? It will make the radio sound worlds better and the UL taps make it easy to wire up the output stage and puts less of a load on the screen dropping resistor. If you look at the schematic for the radio, there really isn't any isolation anyways because B+2 is shared by the phase inverter and other pre stages of the radio without isolation RC filters between stages. I do want it to be good quality when I plug my Samsung T10 or CD player into it. I might want to put a nice speaker in there at some point in time, but the one that is in it is darned good.
I'm still trying to figure out which power transformer to get for it. With today's modern voltages a 300-0-300 hammond is going to be more like 350v. The last 270FX I had was supposed to be 275-0-275, but when I measured it powering a SE 6V6 amp with a 5AR4 it was still up around 400v B+
A 300 series hammond might not be as much as a problem, but they are more expensive. Another option is a Weber power transformer for a fender champ. Their upgraded transformers are rated at 100ma.
How much current am I looking for this thing? Assuming worst case scenario of a ~15 watt output stage and the rest of the radio.
The metal oxide 1w resistors from AES are 350wvdc 600v overload, their metal film 1/2w ones are 350wvdc 700v overload
The hammond: I have one laying around and don't want it to go to waste.I thought why not? It will make the radio sound worlds better and the UL taps make it easy to wire up the output stage and puts less of a load on the screen dropping resistor. If you look at the schematic for the radio, there really isn't any isolation anyways because B+2 is shared by the phase inverter and other pre stages of the radio without isolation RC filters between stages. I do want it to be good quality when I plug my Samsung T10 or CD player into it.
I might want to put a nice speaker in there at some point in time, but the one that is in it is darned good. I'm still trying to figure out which power transformer to get for it. With today's modern voltages a 300-0-300 hammond is going to be more like 350v. The last 270FX I had was supposed to be 275-0-275, but when I measured it powering a SE 6V6 amp with a 5AR4 it was still up around 400v B+
A 300 series hammond might not be as much as a problem, but they are more expensive. Another option is a Weber power transformer for a fender champ. Their upgraded transformers are rated at 100ma.
How much current am I looking for this thing? Assuming worst case scenario of a ~15 watt output stage and the rest of the radio.
Digikey stocks PR01/02/03 metal film resistors with comfy voltage ratings. I just always get the 3W ones because the prices aren't much different. (3W -> 750WVDC)
http://www.vishay.com/docs/28729/28729.pdf
http://www.vishay.com/docs/28729/28729.pdf
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.