RCA console amp RS-199C

[HBrown]

Wow! Thanks Tubeactive! That info is more than I could have hoped for.

I will definitely try those mods as it sounds exactly like the results I was envisioning that I wanted. I have several little SET amps so I had a good idea what this one should be capable of, and it just wasn't there. This little amp fits really nicely into a convenient spot in my secondary listening area where a larger chassis like my ST-70 or even the Magnavox 9302 just won't fit. I will be stoked to get it sorted and sounding good.

I am not well versed enough with tubes to be able to recognize what was going on, but I have tried several different things suggested previously with only slight to no impact on the drive power of the amp. Yes, I did end up having to put resistors in my heater drive circuit to tame a really high heater voltage. Now that you have given me a better understanding of the gain situation, I will begin putting together a parts order and get this project moving.

 

[BinaryMike]

... for simplicity at first, disconnect each 2W 47 Ohm from the connection point of the yellow speaker leads, which is also the white wire feedback connection. Simply bend back the 47 Ohm leads and safe them off. Then, we can now imagine that each 100pf appears as the feedback connection. Install a 33K or 22K resistor across each 100 pf. Therefore, each 6CG7/6FQ7 cathode will now have a 3300 Ohm cathode R. The white wires for feedback will have a 100pf in parallel with 22K or 33K (33K for less feedback and very open sounding IMHO, 22K for a bit mellower sounding...

The 47R resistor simply insures an adequate load on the amplifier for stability with its original feedback loop intact. You will often see this resistor in series with a small cap so that the necessary loading appears only at ultrasonic frequencies, where speaker impedance rises dramatically, but omitting the cap does little harm in terms of power dissipation at low frequencies.

The suggested revisions reduce feedback to an almost insignificant level, which radically increases output impedance and distortion along with gain. RCA's original design had the necessary gain upstream of the power amp, so an extra gain stage is the 'correct' way to remedy this chassis' gain shortcomings. In this way, the original feedback network's gain/phase characteristics can be preserved, along with the engineer's designed output impedance (damping factor), low distortion, and frequency response flatness. GordonW's solution also makes a great deal of sense, engineering-wise. Now I do realize that some folks prefer the sound of pentode amplifiers with little to no feedback and their various idiosyncrasies. I just think the OP should make his decision in the light of as much understanding as possible.

 

[HBrown]

I still have the 6DT8 input tube in it right now, with the mods that were suggested for it.

I assume the feedback mods and resistor values tubeactive gives are for the stock tube set-up.

I've changed stuff around on this amp enough to where I have cut parts out, shortened leads on components and 'Jerry-rigged' a bunch of stuff. I need to strip it all down and start over with good new passive parts so I can make room to incorporate the upgraded caps when they come. I'll get the power supply squared away first then we can have some fun.

 

[HBrown]

I'm not adverse to adding another gainstage to it by any means- in fact I'd prefer it- but I'm not astute enough at this point in my journey to properly design it. The outlay of cash for basically whatever tube I would need and res/caps is trivial.

At this point, I'm not necessarily concerned about keeping the original circuit intact at all- but since it is working, albeit at reduced volume, it does sound pretty nice as-is.

 

[BinaryMike]

I'm not adverse to adding another gainstage to it by any means- in fact I'd prefer it- but I'm not astute enough at this point in my journey to properly design it.

Simple voltage gain stages don't have to be designed --- it's already done for you, cookbook style. Did you check out the information in post #59?

 

[HBrown]

BinaryMike, yes, I did spend some time perusing that section in the tube manual- I downloaded it for reference- haven't printed it out yet. Good stuff in there- a big thanks for making it available. I do understand there are 'cookbook' circuits in there- I just have to get in there and learn the rules of how stages are interfaced together.

I do have several tube amps- but I didn't build them from scratch- I just replaced parts here and there to get them going. I'm not new to electronics, but I am fairly new to the tube side of the hobby.

 

[gadget73]

I'd probably just find the schematic for the original tuner, and just clone the last stage onto the amp chassis. No design work needed, you're simply building it exactly as RCA did electrically, but physically a little different. The original feedback connections and resistor values would be correct for this particular setup, no fiddling needed.

 

[HBrown]

Ok everyone, so, I just got in the shop and began trying stuff.
I first tried a 12AX7- but found it had too much gain (like folks told me it would, but I had to try it). I then ended up with a 12AU7 driving into the stock 6FQ7. I basically tried a bunch of values (limited to what I had on hand) of the cathode resistors, coupling caps and feedback scenarios.
I arrived at the configuration shown on the attached schematic I drew up. With this circuit, I'm getting plenty of gain with a CD player or an MP3 player and it sounds pretty darn decent.

Right now the amp exists in practically a 'breadboard' state as I attached a metal plate sticking off the top surface to add the extra tube onto for testing. This design has a lot of potential as-is. I have decided to take all of my parts off of this old chassis and build it up with a symmetrical layout on a new, slightly larger Hammond chassis.

I used a mixture of everyone's suggestions in putting this together. I realize my separate cathode resistors on the 6BQ5's should be 130-140 Ohms- I had a 150 Ohm on hand and tried it- and it works. I have ordered some 130 & 140 Ohms to try in it. I have ordered several values of coupling caps to go between the 12AU7 and the 6FQ7 to try as well. I would like to continue to optimize it. If anyone could look over my schematic and see if anything jumps out at you I would appreciate it. I'm willing to try whatever I need to improve it. All suggestions are welcome.

 

[HBrown]

A pic of the very rough "breadboard" style added gain stage so far. I will transplant in a new larger chassis when I get it optimized.

 

[gadget73]

How much feedback are you getting? What sort of square wave are you getting?

 

[HBrown]

Gadget- I wish I could answer those questions. I have no idea about 'how much' feedback. I do own a scope, but I have only begun to learn how to use it. So far, this effort had been strictly seat of the pants and just 'trying parts' based on other circuits and others suggestions. I am willing to learn. By some stroke of good fortune, this thing sounds really decent.

 

[BinaryMike]

I would urge you to restore the feedback circuit to its original configuration before making any decisions concerning required gain in the new input stage. This will insure that your damping, distortion, squarewave results, etc are at least as good as the engineer achieved.

 

[HBrown]

That's a good observation. I have not tried the original feedback arrangement since I added the extra tube. I am just ecstatic to finally get some volume out of this amp. At this point, it sounds better and has more output than my other SE amps.

 

[gadget73]

Feedback level calcs are fairly easy. All you need is a decently accurate voltmeter. How I'd do it is to unhook the feedback and set the output to some voltage across your dummy load. 2 volts should do it, but record the exact voltage. It just needs to not be so high that it clips. Connect the feedback and note the new voltage. usually I do this at 1khz, but if your meter is not accurate at that frequency, 200 hz should be fine. Just don't go too low in case the transformers or tubes saturate and mess with the readings.

from there its just some math, or cheat and use the calculator here

http://www.sengpielaudio.com/calculator-gainloss.htm

You want the "voltage and gain" one. Reference voltage is your measurement with no feedback, measured is the one with. If you reverse it, you get the same number but its positive instead of negative.

 

[HBrown]

Did the feedback measurements and used the calculator Gadget linked above- it came back with -12 db.

Can anyone point me out a reference, or very basic instruction for using a scope to get a square wave up to view? I have a tektronix 2235A. I can get the scope set to square wave etc.. just how do I specifically connect to the amp?

 

[gadget73]

12 db is a reasonable amount of feedback. Usually enough to get the job done without too much risk of stability problems. Some things will take more, I have a console amp that runs 19 db, and another one that if you even think about more than 11.8 it goes absolutely bonkers.

You need a generator to make square waves first. Basically you connect that to the input, and the scope to the output. 1khz and 10 khz should tell all there is to really know about it. Some scopes do have a calibration lug that will give you square wave output at 1 khz, just need some way of controlling the level so it doesn't overdrive the amp. Don't need more than about a volt of output at the speaker terminals.

 

[HBrown]

Thanks Gadget- I'll have to scare up a generator. I've needed one several times. Now's as good a time as any to find one. Off to the auction site I go.
I remembered I have the Kewlsoft function gen app on my phone I downloaded a couple years ago. May try that in the meantime.

 

[gadget73]

as long as it makes a good clean square wave it should work. You need good input to be able to judge the output though.

 

[HBrown]

Got around to messing with it a bit more. Hooked the smartphone function generator up and scoped the output and got this- then checked the input signal from phone and the phone's input signal has the fuzzyness in it but not the sloped part.

 

[gadget73]

Input is suitably flat-topped?

If so, the slant shows its bass limited. How much voltage output is that, and at what frequency? Too much output will saturate the transformers and cause this, too low will do the same. Its a console amp, its not going to have earth shaking bass capabilities. I would not expect to see a lot of slope at say 1khz, but if this was 100hz it probably will fall off.

Its got a bit of overshoot, but honestly thats not terrible looking. A small value cap or cap+resistor from one of the triode plates to ground would likely take care of that. Maybe 200pf + 1.5K or so. Experiment with the values as needed.