Antek's new trannys?

[Blair]

Hey everyone,

I posted on the other forum about 100w schematics and a guy that is giving me hell about getting my trannys I bought off ebay. I'm not giving up yet, but I'm not sure there will be much I can do besides show up at his door. We will see what happens though.

Anyway, all of this planted the desire to build a pair of 100 watt monos. I see Tom's point about building a few 30 watt amps and making them sound like 100 watt amps by improvising with the speakers, but I'm not a high efficiency speaker guy. They drive me nuts and then some....Sorry for all the hard core HE speaker/horn guys.

Back to my thoughts and musings because this may not leave my head until a few answers happen. I was referred to the Ikezilla project and that looked very interesting, but I was looking at a different, more cost effective solution. This led me back to Antek again for a quick peak at their trannys. I can buy two 4T400 800vct trannys to get a solid 530v B+, and now they have some nice looking and spec'd OPTs! The AT1008 are rated at 100w and have great looking freq. response. The only issue I can see if they only have an 8 ohm tap and they lack the 40% screen grid. They have 2,022 ohm input impedance though, so I thought they woudl be optimal for a PPP KT88/6550/EL34 amp, but without the 40% screen grid does it not limit you to well under the 100w output? Any suggestions?/Thoughts about these trannys?

http://www.antekinc.com/AT-1008.pdf

http://www.antekinc.com/AN-4T400.pdf

Blair

[EWBrown]

The utralinear "screen grid tap" mode generally results in somewhat lower power output than does the classic "pentode" mode, and "triode "mode offers the least power output.

A couple of KT88s in pentode mode should get you pretty close to 100 watts, or if you want to play it a bit more conservatively, use paralleled PP EL34s, 6550Cs or KT88s, which will require more filamnt and B+ current to feed the two extra tubes.

The 2K primary impedance would be a better match for paralleled push pull operation, as the normal prinary impedance for a single pair of any of these tubes is between 3400 and 4300 ohms A-A. half the primary impedance and twice the total plate curent will deliver approximately twice the output power than a single pair would with a 4.3K A-A OPT.

My own preferences, I would go with 450 - 500VDC B+ at around 300 mA current for each channel, as a single power trannie for both channels would have to be a "monster" :o Yellow_Light_Colorz_PDT_07

The other approach for higher power amplifiers is high B+ voltages (800VDC and greater) with tubes like 811s, 572s, etc. Which also require some very careful design and layout techniques to accomodate these higher (and more lethal) voltages.

Antek might be able to custom wind some ultralinear OPTs, probably at additional cost :o

If there is enough demand, then they might add the \m to their list of "offerings"


/ed B

[Blair]

Thanks Ed,

I was looking for something around 530v B+. Unless I pad it a bit, that is what I get from the tranny. Antek has a new version of the 800vct with a 70v bias tap now.

I was looking at a few schematics, but they all use the UL/Screen grid tap to get the 100-120w. I would not ask Antek to build any custom trannys because it took them over two years to come out with these! :o

With this OPT layout do you have any suggestions running a B+ just over 500v to get roughly 100w? How about 6 X EL34 or 6550/Kt88? Would the 2K impedance be OK there?

Blair

[Blair]

A better question is do you have an example of a Pentode amplifier that produces 100w with any tube that would work with a 2K primary OPT?

Or rather how it is wired? I know how to wire an amp in triode, but then I would need 6 KT88s in parallel to get 90w or so in triode right? I'd rather either use a cheaper tube or less tubes. It would look cool thogh to have a six pack of EL34s or 6L6GCs on each side.

Blair

[kheper]

A better question is do you have an example of a Pentode amplifier that produces 100w with any tube that would work with a 2K primary OPT?

Or rather how it is wired? I know how to wire an amp in triode, but then I would need 6 KT88s in parallel to get 90w or so in triode right? I'd rather either use a cheaper tube or less tubes. It would look cool thogh to have a six pack of EL34s or 6L6GCs on each side.

Blair

The Antec 35 PP toroids are on ebay now -

http://cgi.ebay.com/ws/eBayISAPI.dll?Vi ... 0124242838


Here is a PPP pentode amp -

http://www.montagar.com/~PATJ/mi12188a.htm

For pentode mode, you need to provide a DC voltage directly to the screens. It would be best to use a separate and regulated power supply, rather than tapping off the plate supply. Running in pentode mode will yield more output power than UL, but the trade off is more distortion as well. For a 100W amp, the distortion should be tolerable, because you won't be driving the amp too hard.

[Blair]

I see,

So, in turn, would it be beneficial from a distortion standpoint to run PP Triode with three sets of 6550s in parallel?

The Tranny has more than enough juice for this even if I ran a pure choke input. The Antek has 500ma per 400v winding and dual 5A 6.3v windings.

That would get 90w triode right?

Blair

[kheper]

I see,

So, in turn, would it be beneficial from a distortion standpoint to run PP Triode with three sets of 6550s in parallel?

The Tranny has more than enough juice for this even if I ran a pure choke input. The Antek has 500ma per 400v winding and dual 5A 6.3v windings.

That would get 90w triode right?

Blair

Triode mode will be less distorty and make a better sounding amp, than pentode mode.

90W theoretically.

[EWBrown]

Three pairs of 6550s should readily generate 100 watts in "triode" mode.
An OPT somewhere around 1600 to 2200 ohms A-A would be a good match, and ANTEK's 2K unit is right in the middle.

For 500VDC or above , I'd go with 6550s or KT88s, as not all EL34s are created equal, and many of the "lesser" ones will detonate, or at least, suffer premature failure at this higher B+ and plate voltage.

With the current crop of modern production tubes, a KT88 can always replace a 6550, but a 6550 can't always replace a KT88.

VTL has such an amplifier, the MB450s. They use 4 pairs of Svetlana 6550Cs, in pentode mode it kicks out 450 watts, and in triode mode it generates a more genteel 250 watts. There is no UL mode. IIRC, the B+ is somewhere around 550VDC. These MB450 monoblocks are heavy beasties (no I don't own any, I've just seen and heard them).

The total B+ current consumption is about the same in either mode. Class A triode operation is less efficient, but then the sound quality is the best.
Pure Class A has its advantages, at the slight disadvantage of nbeing relatvely low efficiency in terms of power consumption vs power output.
THe better sound quality is because of the lask of "crossover distortion" found in Class AB or Class B amplifiers, which is an additional level of distortion beond 2HD and 3HD..

THe driver tubes are a parallel connected 12AT7 for VA, and a GE 6350 for the LTPI. The 6350 is a computer tube, basically a 12BH7A on steroids Yellow_Light_Colorz_PDT_04

/ed B

[Shannon Parks]

Thermion's Ikezillas will probably come off the back burner sometime in 2009. He's acquired some full range electrostatics (super inefficient beasties) and he *needs* 100W. Tubed power of course! Separately, nn my quest to finally revise the Ikes, I've been looking at the Anteks over the last few weeks. I'll let you know if I purchase them.

[Blair]

Hey Shannon,

Just a few questions.

1) Are you working on the Ikezilla in fullboard form...i.e.. will there need to be all of the lead wires like in the forum post?

2) Which Antek trannys are you looking at? Are your going to be pentode amps with a triode switch? How audible is the difference in distortion between pentode and UL? I woudl prefer not using 6 KT88s per amp! :o

When the tube cost outweighs the tranny cost, it is a bit rediculous to me.

I'm not making any promises, but John at Antek also hinted about making some SE trannys soon Yellow_Light_Colorz_PDT_11

He said he can make them with the air gap and all, so they should be nice.

In the mean time, could someone come down here and break my hands before I spend too much on amps!!!!

Blair

[Ty_Bower]

How do you put a gap in a toroid?

[WA4SWJ]

Saw it.

[Blair]

Saw it.

A little slow of the uptake this morning. I was trying to figure out what you were talking about. Yellow_Light_Colorz_PDT_02

As far as gapping a toroid, I do not know how he woudl do it. I just know he said he was contemplating building SE toroids and I he asked me if I thought it would be a good idea. I said it would definitely be a good addition. I also suggested a project page to appeal more to the DIYer with reviews, etc....

I think I will pick up a pair of the 100w OPTs for now though. It will definitely get me to the "have to finish" point versus the thought process to get there.

Blair

[EWBrown]

I dunno just how well a sawn toriod would work (or even stay together)

The usual SE OPTs are the EI's stacked with an airgap or the C cores witn a thin gap between the two Cs. A cut toriod would probably unwind, or at least get really screwed up (I won't use the more commonly used expression here).... Yellow_Light_Colorz_PDT_09 Yellow_Light_Colorz_PDT_04 :o

The smaller 20W and 35W ANTEK OPTs look pretty intereting, too.

Now if they come out with a line of similarly poted / shielded power trannies, they could give Trafomatic a run for their money.

/ed B

[Blair]

Now if they come out with a line of similarly poted / shielded power trannies, they could give Trafomatic a run for their money.

/ed B

I agree, and I looked at the sizing to see about having them potted just like the OPTs, but if worse comes to worse, I'll pot them myself.

It looks like the PT is a bit taller, but I'll ask anyway.

Blair