How does output transformer primary impedance impact the circuit? Princeton Reverb

[HowardMoon]

I'm planning a Princeton Reverb build and since all of my cabs are 16Ω I'm looking at this multi-tap output transformer, which the manufacturer confirmed is the best match for a BF Princeton Reverb. Thing is, the classic Fender/standard 8Ω Princeton Reverb output transformer reflects an 8.5kΩ primary impedance, while this one's primary with all 3 4/8/16Ω secondaries is 6.6kΩ.

My question, not having a ton of electronics knowledge (but more each day thanks to forums like this):

What's the impact on the amp circuit to have a differing primary impedance than it was designed for? I understand I can double or halve the primary impedance by using speaker mismatches, but the starting difference isn't large enough to make this useful. Is there something in the circuit that needs to be done to compensate for a nonstandard output transformer primary impedance? Different resistors? Bias adjustment?

My main concern is that none of the components will be damaged/overly stressed in the long term, but I'm also interested if there's a widely accepted impact on the tone, since I definitely want this amp to sound how a Princeton Reverb's supposed to sound.

 

[Lowerleftcoast]

The Deluxe Reverb uses the 6k6. The Tweed Deluxe and the Princeton Reverb use the 8k.

The difference... the 6k6 tends to hang on to the clean longer.

The circuits all use 6V6 valves. There is no difference in resistors or bias to compensate for the OT.

The 6k6 may technically be (from Hi-Fi standards) more of a correct value but neither will damage/overly stress the circuit.

Your choice..Do you want it to stay clean like a DR or... No wrong answer. Just a choice.

 

[printer2]

And I thought the higher impedance would be cleaner, as long as the voltage is higher. The 6.6k tap will ask more of the tube and your power supply to make more power. The difference may not be that much audibly. An amplifier is the sum of its parts. If you have a lower impedance transformer it will load your power supply more, if the PS does not have the current capacity the voltage will sag. (not a bad thing). How much power you make depends on the impedance and the voltage the PS is able to deliver.


Found this, page 8-9 goes through the distortion and impedance.

http://rfcec.com/RFCEC/Section-3 - ...er Output and Distortion (By Steve Bench).pdf

 

[Jsnwhite619]

6.6k will be a bit cleaner. But as stated, it will tax the PT considerably more. What brand are you looking at?

Sent from my moto g(6) using Tapatalk

 

[King Fan]

And which PT are you using?

 

[HowardMoon]

Thanks for the replies so far, folks. Glad to hear at least some of you say it won't be a huge deal!

I linked to the OT in the first post--ClassicTone is the brand. I've also eyed the more-expensive Mercury Magnetics FB-FPOM, but they don't provide any info on primary impedance (and neither does their standard Princeton Reverb OT). Seems kind of weird that this info isn't available.

For the PT I was planning to stick with the one included in MojoTone's kit. Looked at the product page and wiring diagram but I'm not sure what info there would speak to whether it would be able to handle working with a lower-impedance OT. Thoughts?

FWIW, when I reached out to the company behind ClassicTone, the response didn't indicate any alarm at using the 6.6k OT for a Princeton Reverb. He did, however say this: "Keep in mind that Fender used 8K Ohms only in the Tweed era. After that they used 6.6K Ohms. I don't know why they used 8.5K for the Princeton reverb, since the Princeton Reverb was much later than that." Hmm...

 

[schmee]

A couple things:
I've used a DR OT in my PR for a long time. It's good.
Another thought: I discovered after building a few PR's that if you want clean headroom, while sticking with the PR phase invertor and 6V6's, a larger Power Trans is the ticket to go along with a DR transformer! Some kit suppliers make a fat stack PT I think. I'm currently using a Vibrolux PT. I was amazed how much that one thing did, compared with the stock PR power trans and a Deluxe OT in both.

 

[Jsnwhite619]

Personally, if you want the 6.6k OT, I'd go with this PT. 33% higher current rating than Mojo's - 100mA vs 75mA - and a slightly lower secondary voltage - 325v vs 330v. https://www.hammfg.com/files/parts/pdf/290AX.pdf

I am no expert or authority on how the vintage stuff holds up, but if I am building my own, I want to ensure a safety net. Dropping down to a 6.6k primary OT is going to put more stress on your PT. Using this calculator - https://thesubjectmatter.com/calcptcurrent.html - the 8k in a Princeton Reverb is going to max out at 120mA, 60% higher than the Mojo's rating. The 6.6k takes that up to 139mA, 85% higher than the Mojo PT rating. The Hammond PT means those numbers would only be 20%+ and 39%+. Obviously, they aren't melting down and dying on an everyday basis because there's enough people using them. But, if I'm picking the parts out myself, 39% over spec looks a lot better than 85% over spec.

As far as sound difference, above I stated "a bit cleaner." In my experience in using the two different OT's in the same circuit - a Tweed Harvard - single coils didn't really want to break up until about 8 on the dial out of 12. It was actually a lot cleaner than with an 8k.

 

[HowardMoon]

Personally, if you want the 6.6k OT, I'd go with this PT. 33% higher current rating than Mojo's - 100mA vs 75mA - and a slightly lower secondary voltage - 325v vs 330v. https://www.hammfg.com/files/parts/pdf/290AX.pdf

Can I ask where you found the rating for the Mojotone PT? It's not on the product page as far as I can see.

 

[Jsnwhite619]

Right here on the hookup diagram.

Sent from my moto g(6) using Tapatalk

 

[printer2]

You can use both impedances on the same PT, you just are not going to get the little extra the 6.6k is capable if you are throttling it back with a lighter PT. There might be a little tonal change, also early tweed amps sagged more, you might actually prefer it to a stiffer supply. It all depends on what you are looking for.

 

[Tom Kamphuys]

Here is a plot of the distortion vs. the plate-to-plate load:

 

[tubegeek]

Found this, page 8-9 goes through the distortion and impedance.

http://rfcec.com/RFCEC/Section-3 - Fundamentals of RF Communication-Electronics/04 - AMPLIFIERS - RF POWER AMPLIFIER BASICS/RFPA - Loadlines, Power Output and Distortion (By Steve Bench).pdf

Note that Bench's article is still talking about triodes until p.22 where the pentode analysis begins.

Also S. Bench always looks at things from an ideal high fidelity/sophisticated engineer perspective - you may not get much insight on "good breakup" vs. "bad breakup" from their perspective.

One of the most knowledgeable people about audio it has been my pleasure to meet.

 

[printer2]

Note that Bench's article is still talking about triodes until p.22 where the pentode analysis begins.

Also S. Bench always looks at things from an ideal high fidelity/sophisticated engineer perspective - you may not get much insight on "good breakup" vs. "bad breakup" from their perspective.

One of the most knowledgeable people about audio it has been my pleasure to meet.

I just did a quick glance at the document, seemed to be worth sharing, I have not had time to take a good look at it but it seemed like something of value. I was just talking clean output, running into clipping is a different can of worms.



On the above graph, always keep in mind that the graph is only representative of the conditions the tube was run at. At higher voltages its characteristics may be a little different. Also the distortion is the combination of the different harmonics.

Here we have a graph showing the second and third harmonics and the resultant total harmonic plot. This tube would be a good match in a push-pull circuit at 8k, the second harmonics would cancel in the output transformer allowing more power out for a given percent distortion. I remember seeing either a 6L6 or EL34 plot with the harmonics shown.

 

[tubegeek]

@printer2

Absolutely of value! Please don't think I was saying it wasn't a useful citation. Quite the opposite! People could do much worse than digesting every word Bench has ever written. I was trying to point out the pros and cons of that info as it relates to guitar amps, is all.

Guitar amps are a combination of "correct" engineering and some things that are done because they sound good, that might not be the way an EE would design them. Unbalancing a phase splitter just a bit to get some 2nd harmonics, stuff like that.

It's like learning how to play - you are well advised to learn the fundamentals of music theory. Then you go slam your guitar against a Marshall cabinet to get a sound no EE would ever design into a system. Doesn't mean Mel Bay isn't important, just that he doesn't apply in every situation. Stephie Bench is every bit as good a teacher and theorist as Mel Bay and I was very glad to see that article again - I've had it printed out and stuck in a file cabinet since last century! So thank you for putting it into the conversation. It absolutely informs this discussion in a very thorough and accurate way and should be read by all of us. Just the clear, detailed discussion of how to graph a push-pull load line alone makes it essential. I was hoping to point you and other people who might also just glance at it to the most relevant section.

 

[printer2]

@printer2

Absolutely of value! Please don't think I was saying it wasn't a useful citation. Quite the opposite! People could do much worse than digesting every word Bench has ever written. I was trying to point out the pros and cons of that info as it relates to guitar amps, is all.

Guitar amps are a combination of "correct" engineering and some things that are done because they sound good, that might not be the way an EE would design them. Unbalancing a phase splitter just a bit to get some 2nd harmonics, stuff like that.

It's like learning how to play - you are well advised to learn the fundamentals of music theory. Then you go slam your guitar against a Marshall cabinet to get a sound no EE would ever design into a system. Doesn't mean Mel Bay isn't important, just that he doesn't apply in every situation. Stephie Bench is every bit as good a teacher and theorist as Mel Bay and I was very glad to see that article again - I've had it printed out and stuck in a file cabinet since last century! So thank you for putting it into the conversation. It absolutely informs this discussion in a very thorough and accurate way and should be read by all of us. Just the clear, detailed discussion of how to graph a push-pull load line alone makes it essential. I was hoping to point you and other people who might also just glance at it to the most relevant section.

Never took it that way. I sometimes quote a post and then post using it as a taking off point rather than in response to it. I like when people brink information like this up and you can get a greater understanding in this art.

 

[tubegeek]

art.

Certainly as much art as science and also, a craft.

 

[HowardMoon]

Thanks, everyone, for all of the interesting info. I have to admit that the last posts are above and beyond my current grasp, at this point, but I did do quite a bit of reading up on "sag" as that sounds to be the primary practical concern inherent in the question of using the multitap output transformer with the stock power transformer.

As I understand it from what I've read, this matchup would potentially cause "sag"--a condition where the output transformer attempts to draw more current than the power transformer is capable of providing, which would likely manifest when the amplifier encounters large transients. The audible effect would be compression of those transients, and potentially distortion. I also read that this can be offset by the filter cap reservoir and some stuff I read recommended using larger filter caps.

So, I'm still interested in knowing if a little sag is the greatest of my worries (i.e., it's more of a "feel" thing) or if there's actually a risk of burning up the PT. I'm getting the sense that this probably isn't the case, but there are a couple posts above that make me worry. Also curious if anybody on here has any thoughts on whether filter cap values could/should be increased to compensate.

I'm learning more every day about how this stuff works, looking forward to coming back to this and understanding the info in the graphs/articles linked above!

 

[tubegeek]

I think you'll find at least as many players looking to INCREASE the "sag" compression as you will those looking to combat it.

 

[printer2]

You should not have any problem with the power transformer.