• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Transformer help for 6V6 - RCA SP10

I'm DIY building the RCA SP10 push pull 6V6 10 watt amp and I have a question about what output transformer to use for best sound.
Wondering if some of you will more experience can help out here. I understand some of the design rules - but then there seems to be a lot of "bending" of said rules.

The RCA Tube Manual RC19 has the schematic in the "circuits" section - calls out 9k-10k output transformer impedance.
https://mysite.du.edu/~etuttle/electron/circ287.gif

I have a hard copy of the the 1954(?) SAMS photofacts for the RCA SP10 and the transformer is listed at 8.7K - close to the 9k.
I've seen other DIY schematics using the 8k on the internet.

My selection I have for transformers are from EDCORUSA and are 8K or 10k (10 watt) version
CXPP10-10k (10K) EDCOR - CXPP10-10K
CXPP10-8k (8K) EDCOR - CXPP10-8K

So which one? Or does it really matter? Three IS a 20watt EDCOR that's 8.6k but it's more $$ and larger and seems way overkill.

If it matters - while both EDCOR transformers have 4 ohm and 8 ohm taps - I'll be using (mainly) the 4 ohm taps for my speakers.
 
Look at the last chart here:

https://frank.pocnet.net/sheets/127/6/6V6.pdf

Notice that second harmonic is minimum between 6K and 7K for a single tube. As you drop the load impedance, second harmonic increases, third harmonic decreases, and power out begins to drop. The 5K recommended load (Dashed vertical line) is an arbitrary point chosen for some trade off between the three.

Since you are building a PP amp, and second harmonic cancels (to a great extent), you can trade off second harmonic increase for third harmonic decrease and a slight loss of power. Dropping below 8K p-p your power will start dropping faster.
 
You can use anywhere from say 6k to 12k pdrimaries depending on plate, screen, and bias voltage selected. I just finished a 6V6 PP amp with 7.7k primaries and I used 380v plate, 325v screens, and -28v bias. Very good sound. 14 watts output, both channels driven.

All else being equal, I'd lean to 9k or 10k primaries. I wasn't able to do that on that amp though, because I had the outputs already, and designed the rest of the amp around them.
 
I would give the more expensive 9K 20W a 2nd thought. Two reasons. First, since your push pull 6V6 should yield around 10W, a 20W transformer will give you more base (low frequency) headroom. Second, should you change speakers or move the system to a larger room down the line, it will be a simple tube (ie: 6L6) and a resistors switch.
 
I would give the more expensive 9K 20W a 2nd thought. Two reasons. First, since your push pull 6V6 should yield around 10W, a 20W transformer will give you more base (low frequency) headroom. Second, should you change speakers or move the system to a larger room down the line, it will be a simple tube (ie: 6L6) and a resistors switch.

I'll take that under consideration - always good for future tweaking. I've heard that the original RCA SP transformers were way over-designed as well.
 
If OPT impedance is too low, the load line may end up going to positive grid voltage with lots of Vp left on the table. aka: The output signal clip way before it is near 0V. You can end up losing power this way. Unless your design handle positive grid current, I would go for north of 8K (Ra-a). So go low but not too low :)
 
Last edited:
There have been several numbers recommended here, but nobody has looked at the important criteria yet, the speakers.

The OPT is a ratio device. For example an OPT that is rated for 8K to 4 ohms, will be 6K into 3 ohms, and 10K into 5 ohms.

Take a look at the impedance VS frequency curves of the speaker system being used. It will be close to 4 ohms at 1KHz, but all over the place at other frequencies.

Keep in mind that those curves were made using static single frequency sine waves. The real dynamic impedance of a speaker system varies a lot more than the curves show, especially in the bass region with dynamic music. If a huge kick drum hit tries to instantaneously reverse the cone's travel while a singer is singing in the midrange frequencies, the counter EMF generated by the woofer will be fighting the amp's output and possibly cause ringing in the crossover network. These effects are hard to measure, and highly dependent on the volume level and amplifier damping factor (increases with higher impedance OPT's, and improved with feedback, especially the type of feedback used in this amp).

If the impedance goes below 4 ohms for a significant portion of the audio spectrum, or has a severe dip below 4 ohms in an important range, then go with a higher impedance OPT (10K in this case). Conversely, if the speaker is above 4 ohms for a large or important range of frequencies, then go for the 8K or lower OPT.

The 6V6 has a rather small cathode with limited peak current capability, so if you like music with large transients, especially in the bass region, pick the higher impedance choice if waffling between two OPT's. In reality the difference between 8K and 10K is small compared to what most speakers do, and may not even be noticeable.

Despite all of what I said, I have a 6600 ohm OPT in my 6V6 amp, which makes 25 watts per channel. I like the way it sounds, but as another member posted, my OPT choice was made long before the amp was built.....I have a bunch of them. My Yamaha NS-10's also spend most of their time well above their rated 8 ohms. The 25 watts come from bending the ratings a bunch on the 6V6GT's. I have 425 volts on the plates.
 
Hmmm...

Since the Dyna ST35 OPT - the Z565 - supposedly has a 7k plate-plate primary with 8 ohm/16 ohm secondary, that means that with a 4 ohm speaker load the Z565 primary will appear as 3.5k ohms plate-plate, with the 4 ohm speaker connected to the 8 ohm secondary tap. (!!)

With today's low impedance speakers that often dip down to 3 ohms, perhaps that Z565 OPT would be better served by a pair of EL86 rather than EL84!

Food for thought there...
 
Last edited:
Ok - after some design layout work I figure I can pretty easily accommodate the slightly larger 21 watt edcor (vs the 10watt), with a 8.6k impedance and 4,8,16 taps. All for total cost of about +$25 difference.

CXPP21-8.6K
EDCOR - CXPP21-8.6K

So I get the almost exact RCA 8.7k spec'd value, and based on most of the comments here - a pretty decent balance. Not sure if they measure up to the original OE iron but from what I've heard about the RCA SP10/SP20 they also used "overkill" size transformers.
 
If I had "4" Ohm Loudspeakers that had a DCR of 2 to 3 Ohms . . .
I would pick the 10k instead of the 8k output transformer.
Get your DMM out, calibrate to see what it reads with the probes shorted; then with a 4 Ohm load resistor; then measure the loudspeaker terminals.
Surprise . . . Surprise!

"A Hundred Million Measurements Are Happening Every Day". I changed one word of Oscar Hammerstein's lyrics in "Flower Drum Song"

We do not Listen to 4 Ohm Load Resistors.
We listen to vibrating speaker drivers.

Just my $0.03
 
  • Like
Reactions: 1 user
I would apply that idea for triodes, absolutely. But I think pentodes are more load dependent, and work differently.

I saw this post in an old thread about choosing the load impedance for 7027A (a higher dissipation 6L6GC).
https://www.diyaudio.com/community/threads/pentode-push-pull.78712/#post-906924

SY:​

There are some interesting aspects to the loading questions. Take a look at the RCA datasheet for the 7027A (my favorite medium-power pentode). There's a chart labelled "Operation Characteristics Push-Pull Class AB1". The chart shows changes in power and distortion with load for a pentode stage with 540V B+ and a 400V screen.

Note the distortion curves. There is indeed a distinct and sharp null at about 6700 ohms p-p. Naively, one might thing to run the stage that way, and indeed, it will result in a very high 76 watts output. But... note what happens if you reduce the plate load to 4500 ohms or so. The max power drops to 60W (not significant), the 3rd rises to 4%, but look at the 5th and 7th! They've dropped out almost entirely. And, open loop, that 4% is not too bad. And note that variation in load either have little effect on the third (load lower) or actually take it toward the null (higher). So using a 4500 ohm nominal loading drops the power slightly, but greeatly reduces load sensitivity.

PS - BTW, the speakers I'm likely to be driving will be Snell Type E-III (8" 2-way, nominal impedance 6 ohms, nominal 91dB sensitivity) or JBL Studio 530 (5.5" 2-way, nominal impedance 6 ohms, 86dB nominal sensitivity). Intend to treat both of those as '4 ohm' speakers. I don't need a lot of power, as I'm in an apartment, my listening room is small and I have intolerant neighbors (unfortunately).
 
Last edited: