My F4 adventure begins

6L6

Moderator
Joined 2010
Paid Member
The 400hz riding on the com radios is largely delegated to history. Though interestingly most 757 still have it audible when they transmit. Yes, still audible in the cabin (mostly the lights), but that's rarely the loudest noise.

400hz is not where the damage is usually done to a professional pilot, it’s more like a notch somewhere in the 6-8k range, depending on what aircraft you were flying. Engines and wind noise in the cockpit are louder than the electronics, by a wide margin.

F4 can play LOUD better than pretty much all of the other Firstwatt as it has the least issue into lower ohm speakers because of parallel devices. The amount of volume you get from F4 is entirely up to your preamp. It’s one of those things that is weird until you wrap your head around it.
 
  • Like
Reactions: 1 users
Member
Joined 2010
Paid Member
The 400Hz (*) I'm referring to is not on the radio... it's the frequency at which the power transformers for the CSS (cabin sub system ) operate. There is at least one transformer per row/column. The entire aircraft is filled with that low level sound.

I worked in labs where we had up to a 100 such transformers.. so when I step into an airliner I feel like they should be paying me for working. Stuff like the PA, headphones, video, audio, over head lights, etc... distribution throughout the cabin.

My tube preamp has 31db gain in the line stage and does 20V max. So it can easily drive the 14.4v at which the F4 is rated into 8 ohms ( 25 wpc )... so the preamp is not the limiting factor with a single F4. In fact, I can easily over drive it. (**) That's the point of driving the F4 balanced +/- 14.4V. Which should yield a 28.8V differential for the 100wpc at 8 ohms (and higher into 6 ohms).

And yes, the F5 doesn't play any louder than the F4. Actually, I think it has less power.

That's why I want the balanced preamp and a second F4... AND -perhaps- an F5 Turbo V3... ;-)

Because, sometimes, you just have to play at 11. You just have to go over the cliff.

I think my end game (***) will be to drive the A2s and F4s into the Maggies ( I sure hope I can ) and the SiTs into the Audio Notes. Keep the single ended CJ into the SiTs and the SIT3 balanced preamp into the F4s. I should be able to toggle the A2s into either amp as I so desired ( might have to sell the A5s... except they were made to sound beautiful... can't sell a beautiful sounding amp, huh? ).

BTW, I got two new racks, with the optional 10" risers and four shelves each, so I should be able to fit three pairs of 5U, 400mm cases... I don't know that I can have them all powered at the same time.. that is a different issue. I could put the preamp(s) on the top shelf, I guess. Keep the turntable, digital source components and phono preamp on the Target stand. (****)

+++++

(*) I think it's 400Hz... might be 300Hz... I honestly forget, I feel like Kelly Bundy in that episode when they discover there's a finite amount of info that fits in her brain. As I get older, I forget things while new things are added.

(**) Well, the distortion I hear might also be the preamp being overdriven with a 2V signal... but in such case it's driving more than 20V. As I approach the limits, we (wife too) hear some hardening and harshness of voices.. specially female voices. The Rat Shack meters, such as they are, do show pretty high power output then, peaks just to the 50 wpc LED.

(***) That's what I tell my wife. Note that my new "amp stands" are expandable.. I could add one more shelf to each.

(****) At some point, you all can come down and take a listen. The BBQ and IPA will be on me. Cigars if you wish, too. SoCal is nice in the winter. ;-)
 
Last edited:
That’s correct, all the AC busses are 400hz on all modern jets, at least all the airliners I’ve flown. Also, modern military jets as run on the same freaks and volts: 115 and 400.

It’s three phase AC generated by an IDG (integrated drive generator which includes a CSD (constant speed drive)) and a generator. The constants speed allows the the gen to generate the non-fluctuating output. The thing has an input shaft that shears (mechanical gear box problem) and can also be electrically disconnected.
 
Member
Joined 2010
Paid Member
So there has to be at least one IDG per engine plus the APU, huh?

We used to prefer running on ground power to save fuel prior to test flights. But sometimes the APU was much more convenient, specially when we were gonna do in flight refueling... and the crews didn't care about cost, only range. Talk about E-Ticket Rides.

TCAS would go nuts.. two large jets flying less than a hundred feet from each other..

In any event, Bose headphones are prevalent in the planes... Ugh... but they seem to be the nicest around. Only times I ever enjoyed a Bose system.

Meanwhile, the F4 is cranking out beautiful tunes in the living room!
 
Member
Joined 2010
Paid Member
I've been thinking of buying a used ARC LS17SE to drive the balanced F4s (*)... but it only swings 15V and has a gain of 18db.. so I got in touch with ARC about it.

Their reply was based on Nelson's FW write up...and only referred to the single ended set up:

"Virtually all active line stages achieve adequate gain and voltage swing to drive the above combination. Basically you need a linestage with about 12+ dB gain and 10+ volt maximum output."

So based upon that, ARC "thinks" the preamp will work.

However, I noted that in the same write up, Nelson states:

"...If your loudspeaker is less efficient, then you need more gain from your source and/or more gain and voltage swing from your linestage. As you go in sensitivity below 87 dB, you will want to consider 100 watt balanced mono-block use, with a preamp capable of swinging 14 volts per balanced output and having a gain of 20+ dB"

So, I'm still waiting for ARC to reply to that, because it would seem that their preamp doesn't have enough gain and is very close to the 14V swing ( should be 14.4V )...

So, that means that the BA3 preamp with balanced outputs will need to swing +/- 14.4V and a gain of 20db.

This is in essence what my current single ended tubed preamp is putting out. Which drives the F4 to full rated output... So, now I will need the balanced version.

In regards to the LS15SE... I'm not quite sure what to do. I figured I'd offer the fellow 1900 bucks... which is definitely not chump change and will force me to do some financial prestidigitation... but I am not going to put out that money just to get another single ended preamp since I'm very happy with my current tubed set up. If worse comes to worse, I'll use the B1K in SW mode... (**)

(*) This is in addition to the BA3 FET baby. The main stereo in the living room sounds awesome when sitting out at the porch!

(**) Has anyone run a B1K in balanced mode? Could I use a B1K -> ACA -> transformer combination to drive the F4 in balanced mode? I guess it won't quite achieve +/- 14.4V, but how close will it get to a20db gain? Actually, do I even need the transformer here since the ACA is a stereo amp if I don't need the full 100wpc into 8 ohms ( about 200 wpc into 4 ohms...).
 
Last edited:
  • Like
Reactions: 1 user
BA-3 with specified parts and JFET IDSS will give you approx 30db gain. Run the rails @ 30 volts, and you will have surplus voltage swing. ZM has previously stated that although +/-24v will swing the OS’es to max output, having more than that in the preamp/FE will ensure low distortion all the way.

This is for FE use, I have not done the balanced math. Sorry for the quick and dirty answer, but anyways: the BA-3 was made for swinging the rails of the F4 to full output, but I know you allready know that :)
 
  • Like
Reactions: 1 users
Member
Joined 2010
Paid Member
Yes, in single ended mode it indeed does... in fact, many preamps do. Any preamp that drives at least a nice 20V and a good 30db of gain will safely run the F4 with plenty of overhead... (*)

My plan, however quixotic it may be (**) is to drive a pair of F4 amps in balanced mode. Hence I will need that +/- 14.4V. And yes, I know the BA3 will do it. I'm just thinking of a 3rd preamp in the mix.

Hence, my question about the B1K. I love how it sounds stand alone driving the other amps. I wonder if I could so something with it, to up its voltage output. I got an ACA and ACA mini at hand, so I wonder if I could use them to up the voltage.. or if I could use a transformer with the B1K directly. For sure it would be a LOT cheaper than going out and spending bucks on the ARC preamp.

(*) I think the ARC LS17SE is simply too close to the rails at a rated 15V ( I think it's 18V in balanced mode.) .. but still has a lowish gain specification ) It will likely run the F4 but at Wide Open Throttle with no overhead to speak of. Which really defeats the purpose.

(**) But then, no one has ever accused me of being a Sancho Panza. The point of all of this is to drive Maggies... 200 wpc at 4 ohms... so I can compare the sound of the A2 ( sweet ) to the F4 ( transparent ).
 
Member
Joined 2010
Paid Member
It was brought to my attention that this is not +/- 14.4V.... that is true and I'm guilty of being somewhat sloppy here.

My read on how Nelson spec's driving the F4 in balanced mode, is that one channel will be "push" and the other "pull" so the differential voltage will be 28.8V.

Also, when I write "V", I mean Vrms, just got sloppy and short handed. So I believe the true number is a differential 28.8Vrms across the inputs of the F4. That makes sense because that will yield 100/200 wpc at 8/4 ohms.

I hope I got this right this time. Thanks.

I any event, I'm currently driving the F4 with the CJ PV9 with teflon tubes and it drives the baby Audio Notes. It makes beautiful sounds... so much that I have not powered up the A2/Maggies for a over a week... The interesting thing in this set up is the the preamp has two sets of SE outputs, so the only difference in sound here is the amp/speaker combo... I also have the F5 on top of the F4 so that's yet another option... The F5 sounds unengaging in comparison to the F4. I have to try the ACA mini with the Audio Notes... tomorrow maybe?

If I had to rate the amps ( allowing for the fact that I have to juggle speakers )... it would be F4, A2, A5, ACA Mini, F5 and ACA (v1.8).... then I got the ARC D70-II, that's tied with the A5, I think. I need to put together the NCore amp one of these days... but my experience with ICE amps and Solid State AB is such that they are on the shelf or other systems.

Hopefully I'll be able to add the SiT amps too. After that, who knows? I want to retire. Right now I joke that I work to pay for my audio toys.

(To be fair, I ought to take my wife shoe and handbag shopping for Valentine's Day - those things are expensive!!! ).

IMHO, the F4 is indeed quite a trip. An amp in search for an awesome preamp.

I just bought two four shelf, with 10" rods, audio stands... will stuff them with amps. ;-)

OP, please don't think I'm hijacking your thread... I think we're both on the same path. This also crossed our path...

https://www.diyaudio.com/community/threads/diy-front-end-2022.394339/

But I think I'm a bit ahead of that info, so who knows, we might have yet another preamp in the making? But we're a bit into this planning.. as they say:

"There comes a time when you got to shoot the engineers and start production".

We'll see. But I'd rather go into the BA3 balanced preamp and a second F4 than delay into yet another tangent. At some point I got to hear the thing and stop thinking about it.

Have fun!

BTW, I got to thank the people involved in this build... parts, time, labor... it's turning out to be a great bunch of people in this site helping out on this project, including, of course, Nelson Pass for donating his IP and humor ( Phew... no I'm not kissing a$$... I mean it...).
 
Last edited:
Member
Joined 2019
Paid Member
"My read on how Nelson spec's driving the F4 in balanced mode, is that one channel will be "push" and the other "pull" so the differential voltage will be 28.8V."

So, so close. I think you have it right in your mind, but just not quite correct in terminology. Here's how I think of it... and maybe even I have it incorrect. It's not because one is push and one is pull. Both channels push and pull. That's the topology of the amplifier. You're interchanging terminology. It's because you effectively have two signals. One is the "standard" signal, and the other is equal and opposite. One runs on one channel, one runs on the other. Because they are equal and opposite, when you hook the speaker to each of the channels, when one channel is outputting +20V, the other is outputting -20V for a Vp of 40V => Vpp of 80V. Graph below. Back home, so I have a computer. Sloppy, but whadda ya want over my first cup of coffee and a few mins of work. I can tidy it up if needed ... if you think it may be useful to anyone else...Back to work on your particular thing.

  • As you noted, when using the 'Ohm's law power equation' all AC voltages are Vrms. It's super helpful to note this.
  • Vp generally is considered "max" voltage. Some specification sheets (IMO) mistakenly will say "max voltage" and quote a Vrms or not use a suffix at all. To me, that's misleading at worst and confusing at best. Why?
    • A rated power may be at a (somewhat) arbitrary THD figure particularly for pre-amps and/or the maximum voltage output to clipping might be higher than the voltage at a level of acceptable distortion. Example - Output at X% THD might be 20Vrms. Max output could be 40V. Assuming symmetrical clipping +/-40V => Max output would be 28Vrms. Nice headroom, but who knows what the distortion looks like above 20Vrms. Either way... I just wish spec sheets would be more consistent. Slightly irrelevant, but that's what can cause some confusion.
    • Some people might quote max power based on Vp vs. Vrms calling it "peak power" particularly for power amplifiers. To me that is misleading
  • +/- Voltages are generally noted as the max swing => +/- 20Vp => 40Vpp => 14Vrms. This is why people often say you need a pre-amp that can "swing" 20V to drive an F4.
You've previously calculated that you need 14.14Vrms to drive a standard F4 to clipping into an 8ohm load. That's 25W.

You can use the same calculation to show that you need ~28Vrms to get 100W into 8ohms. That's not the same as +/- 14V or even +/- 14Vrms (which would be nonsensical).

Anyway... As you've noted... double the voltage ... 4 times the power output (assuming you've got enough current). Start there. Note... This is not the differential voltage seen at the speaker terminals. It's not the +/-. It's just the Vrms. Unless you're referring to Vp, I'd stop putting in the +/-... but that's truly just my opinion. Carry on if it makes sense to you.

The differential voltage that your speaker sees for maximum excursions one way or the other is... +/-40Vp => 80Vpp. See graph. Peak differential voltage is 80V. Like I said, I think you've likely got it correct in your mind, but it's just a VERY slight differential (see what I did there) in terms. However, the terms and the notations around them are important... particularly if you want to convey what you want to a builder. Like I said... it's fun to play around with the numbers. I set all my stuff up to run with a nominal 4ohm load.... b/c I can. I don't need the voltage swing, but I tend to run more devices in parallel for ooooooomph in the bass where my speakers dip a bit lower than 4R.

Appreciate you reading my ramblings re: why I was confused by your original post, and how I think through it. Again, not saying you're incorrect, just explaining my confusion.

If you want 100W into a 6ohm load... I'll let you finish the math.

Enjoy!!!!!

:cheers:

Edited to add - If I were being pedantic and even more correct ... but possibly more descriptive ... in the notations in the graphs and above, I'd have said... The speaker reacts to the difference in "potential energy between the two connections, which is measured as a voltage".... You know that, but...
 

Attachments

  • Differential Example.jpg
    Differential Example.jpg
    102 KB · Views: 38
Last edited by a moderator:
  • Like
Reactions: 1 user
Official Court Jester
Joined 2003
Paid Member
one "normal" amp is having +/-24Vdc rails (nominally, but we'll stay here)

that means - available voltage envelope for output signal is 48Vpp; call it window, call it range, whatever

we always have some voltage losses ( due to drive mechanismus ( rail to rail, less, more), due to series losses in OS devices itself) so, let's say that F4 is having some 4V losses yadayada, leaving 44Vpp

44Vpp/2=22Vp

22Vp/1.41=15V5rms

15V5rms@8R ....... as P=Urms^2/Rload, means 30W


now, if we take two "normal" amp channels and connect them in bridge fashion, these two ( now) halves are working see-saw, meaning they're working antiphase

which means - speaker is now connected between their hot outputs, GND is not involved in speaker current path, and while one amp half is going up with voltage, second halgf is hoing down with voltage, result being that zpeaker is getting doubled voltage, as sum of two voltage halves

so, if one half-amp is able to give 44Vpp, second is same but antiphase, and sum is 88Vpp

88Vpp - 44Vp - 31Vrms

31Vrms@8R - 120W

in praxis , we have less than 30W, and less than 120W, reason being that rails are just nominally 24Vdc; 18Vac secondaries, CRC filtering........ rails are in vivo 22V5dc, under heavy modulation slightly going down

so, 25W@8R, 100W@8R

what's not to like?
 
  • Like
Reactions: 4 users
Member
Joined 2019
Paid Member
@tonyEE... That's what I get for posting before second cup of coffee... that was supposed to be continuation of our PM. So sorry.

Oh well. For all others that had to read my ramble... apologies. Was trying to keep this tidy for the OP. Too late to edit for better clarity or delete. Context from other PMs is missing.
 
  • Like
Reactions: 1 user
Member
Joined 2010
Paid Member
No problemo... On weekdays my wife brings me my first cup of drip coffee before I get out of bed. :)

On weekends, it's usually my task to bring it to her. :p

Yes, I was basically incorrect in my terminology. The reason I was using +/- was to show that both sides are out of phase. So, the effectively swing in voltage will be 28V.... but the power supply itself doesn't change.

Thanks to you all for the comments.

Now... how about figuring out how to use a B1K to drive a single F4 or balanced F4s?

By the way, "Zen Mod" is actually quite funny and helpful.
 
  • Like
Reactions: 1 user
Member
Joined 2010
Paid Member
yes, please. I just wanted "names" crossed out, that's all, the rest of the discussion is good and fine with me.

Now perhaps we can get back to the question about driving the F4 with the B1K.... Either I buy a transformer or I use an ACA.... hmmm? I guess I will need to make some cables.

Thanks.