|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
206.75.129.36
In Reply to: RE: What capacitors to buy to refurbish old RCA amp? posted by Michael Samra on February 10, 2016 at 00:31:27
Here Is the original RCA schematic - I was lucky to get the instructions with my amp. The one you have is based on reverse engineering, and isn't completely correct. There is global feedback here, it is just taken from the primary of the output transformer instead of the secondary. Enough to keep the damping factor up to about 2.5, which is no dynaco but it's not dissimilar to many triode amps. I put my 8 ohm speakers on the 4 ohm tap to help control them a bit further.As for modifying to silicon rectifiers, there's a couple of things which persuade me against it. First, one pair of GE 5R4s has lasted 15 years - through three or four sets of 807s and they still test good and put out full voltage. Second, the power transformer puts out a huge voltage 755-0-755 and I would be afraid to lose the rectifier voltage drop.
Hum is cancelled by applying hum voltage to the input stage, a trick which works very well in practice, I can't hear anything even up close to apeakers.
So far the Russian capacitors look like they are the front runners for the 0.1s, I am going to get a set.
Sprague doesn't make electrolytically of the correct ratings, but those solens sure look interesting. Are they suitable for filtering?
Edits: 02/10/16Follow Ups:
There is global feedback here, it is just taken from the primary of the output transformer instead of the secondary
I noticed now but I have never seen the GNFB come off the plate winding like that.I have seen local FB implemented like that.
.Most RCA amps did have two 5R4s,at my RCA theater amps do with 807s.
Here is the thing,the Schotkeys sound like a tube rectifier without the loss and they are quiet.I would work with Eli a bit on the 6SL7 driver stage because you can make a lot better amp with a more suitable driver stage.The 6SL7 is ok because your driving pure pentode but being you have four 807s and regulated screens,give those 807s some extra wheaties.
Nobody ever went broke underestimating the intelligence of the American public
H. L. Mencken
The feedback configuration threw me for a loop at first too (no pun intended!) Especially the little trick they used to bias the first 6SL7 section properly. DC current flows through the feedback resistor, which would upset the bias of the first 6SL7 section - but rather than let that happen, they put the grid a bit positive with a voltage divider, while C2 makes sure it's still referenced to ground AC wise.
I think that from a conservative design standpoint, a second 5R4 would be a good idea.. but I have to say that in reality, it's really not necessary. At idle the amplifier draws 210mA, which approaches but does not exceed the 250mA limit of the tube. And I have used these amplifiers a LOT, I mean in the 20,000 -30,000 hour range since I got them.. and the USED GE 5R4GA's which came with the amp (and who knows how long they were used before I got them) are still good! And pretty hard use too, I used these to rock the house! When you have it really cranked the 0D3s dim a bit during transients.
I haven't noticed that the driver is inadequate... I suppose a more robust driver MAY be able to push the 807s into AB2, but that would probably expose power supply weakness. One possible area of concern is balance in the phase inverter, I do want to investigate that once all the capacitors are replaced. I will definitely make sure that the driver doesn't clip before the output stage does, but something tells me the 1950s powerhouse of engineering which was RCA wouldn't have let something out the door which was designed that badly!
Which model of RCA do you have? Are they the rack mount style ones with built in tube tester? I'd love to get my hands on a pair of those some day!
I have several
I have a pair of 9358s,I have one of the 807 versions that you see in the photo,and i have four of the M-12182s which are the 6146 amps..I converted one pair to 6550s but I really think these amps have a lot more potential and I was see what they can do.
The 9358s and the 807 amp in the photo has round output transformers that are high frequency limited from what I saw on the sweep out of the amps.
Nobody ever went broke underestimating the intelligence of the American public
H. L. Mencken
Nice collection! I didn't realize the round output transformers were that limited, I've wanted something including them for a long time - my MI-12188a's are very wide range. I also have an MI-12202B which I have set up to do TV sound. It's very restricted above about 8k, but I like having an RCA amp from 1940 included in day to day life! If I see one cheap enough on eBay I want too get an appropriate MI series speaker to go with it.
The 6146 amp I believe is the successor to my amplifiers - I think the substitution to 6550 was an excellent idea. If they have similar transformers I would agree there's tons of potential there, especially with four of them!
Here is the poly cap you can use in place of the 30uf lytic
On the driver,there are nice high impedance CCS or even an enhancement mode mosfets you can use that will give the 6SL7 an easier time of driving the outputs.I know it works by itself but keep in mind this is 1940s technology and they weren't using it for hifi.. You're allowing the 6SL7 to drive into a higher impedance load and it won't add a sand type sound as there is no gain but you can try a few things.
Nobody ever went broke underestimating the intelligence of the American public
H. L. Mencken
Yeah, at the rated 250 mA., the forward drop in a 5R4 is a whopping 67 V.
You could go SS B+ rectification, by using what I call a "cockeyed bridge". Use series wired pairs of UF4007s to connect the ends of the rectifier winding to ground. Snub each of those pairs with a very high WVDC 0.01 muF. cap. Take the "raw" B+ off the rectifier winding's CT through a 1200 PIV Schottky diode. Insert a CL-90 inrush current limiter between the Schottky diode's cathode and the PSU filter. The thermistor will do several good things, including a slight delay in B+ rise. Choke I/P filters generate nasty inductive kickback spikes that can destroy SS diodes. Either a high WVDC 0.01 muF. cap. at the filter's I/P or an appropriate MOV across the choke disposes of the problem. The OEM oil filled caps. will remain the PSU reservoir, but they will become part of a RC section that will follow a new stack of 2X 10 muF./450 WVDC 'lytics and voltage equalizing resistors. Ripple will be lower than OEM and that new R is how you compensate for the much lower forward drop in SS diodes. A "routine" inductive wirewound resistor is the type to use. Every last bit of additonal inductance contributes to the cause.
BTW, a "cockeyed bridge" is, in fact, just another FWCT arrangement.
Eli D.
Was thinking about concerns regarding the ratings of the rectifier tube. Although the GE tubes in the amplifier continue to work well, I am thinking about what can be done to improve matters.
Looks like the military 5R4WGA-B-Y will drop in, and it is rated to 275mA and for higher voltage than the RCA original. I think I will pick up a couple of these, antique electronic supply has them for $9.95 each, they should keep me going more or less indefinitely.
I wonder if the Russians made a tube with same or higher ratings - although usually I wouldn't plan on using a Russian made tube this close to the edge and expect reliability!
The most rugged 5R4 variant I'm aware of is the so called "potato masher". That variant was used in early B-52 bomber avionics equipment. :> D
Eli D.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: