Ikezilla Revisited: The listening test

a DIY, modified Mullard 5-20 monoblock design

Ikezilla Revisited: The listening test

Postby Thermion » Mon Jan 09, 2006 5:40 pm

If you have been following the Ikezilla threads, here is an update. After the initial test, Shannon added AC coupling between the GC and LTP stages and started tweaking the feedback. He also added a CCS on the LTP and changed the LTP plate resistors to 62K (as in the updated Ike manual). We tested this configuration with a 500V B+ and 70 ma bias and got 0.8% THD at 151 watts at 1k. We were also under 1% THD from 30 to 10K at 151 watts. Still room to tweak, but thats another story.

I know some of you are waiting for schematics, and I have been working on this. Should post something in a week or two.

Anyway, I snagged the prototype back from Shannon to do some additional testing. Today I had a day off from work and the wife and kids, and conducted some listening tests. WOW!!!!
:o

Mind you, these were mono listening tests, so I wasn't sure how it would go. After and hour or so of warm-up, I started cranking and this amp really delivered. I was playing a Sheffield Labs CD "Prime Cuts". which gives a variety of great recorded music. Female vocals, big band, horns, drum solo, and more. The 'Zilla passed every test with flying colors. I was amazed!!! One Apogee Centaur said, "its about time", and the other one said, "what about me."

The music had it all. Detail, natural sound, attack, low end, high end, amazing vocals. Shannon stopped by for some of this and may have his own take. Anyway, I was very pleased and will be ordering parts for the pair very soon.

JT
Thermion
 
Posts: 181
Joined: Mon Jul 07, 2003 6:04 pm
Location: Fisher, IL

Tease!

Postby dhuebert » Tue Jan 10, 2006 8:05 am

I can tell ya right now that if you don't post some schematics soon, I'm gonna get mad!!! I have a pair of eiclone boards and boxes waiting for some finals on iron, tubes and schematics.


Learning patience the hard way:

Don
User avatar
dhuebert
KT88
 
Posts: 820
Joined: Thu May 01, 2003 9:26 am
Location: Winnipeg Manitoba Canada

'Zilla Wafers

Postby Shannon Parks » Tue Jan 10, 2006 8:04 pm

Indeed, the 'Zilla smacked the Apogee Centaur around like it was a schoolyard bully shaking down a 1st grader for his milk money. Impressive.

If Thermion gives me a sketch of the schematic, I'll be happy to put it into a regular schematic with my software. There's just too many little things I don't know about (read: forgotten). Thermion has good notes though.

Shannon
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Tease No More!!!

Postby Thermion » Wed Jan 11, 2006 8:55 pm

dhuebert wrote:I can tell ya right now that if you don't post some schematics soon, I'm gonna get mad!!!
Don


Well Don, I wouldn't want you to get mad. I honestly didn't realize my pet project was holding you up. When I built the prototype, I didn't need to develop a schematic. It is just a stock Ike with a paralleled output stage.

Anyway, I scabbed the schematic together in three parts. The first is the front end (grounded cathode stage and long tail pair phase splitter using the CCS).
www.diytube.com/ike/gc_ltp.jpg
We are still tuning, so I left the values for C7, R8, C8, and R49 blank.

The second part is the output stage.
www.diytube.com/ike/output.jpg
When I build my pair of these, I am planning to use 47 ohm resitors in place of the 39's and 100's.

The third part is the bias network. This simple voltage divider network was borrowed from the article, "60W Triode/100W IL Control Amplifier," by Joseph Norwood Still. The article was published in the June 2003 Audio Express.
www.diytube.com/ike/bias.jpg
Remember, the positive side of the caps go to ground. With the Toroid of Maryland PT, you get about -70v.

All three schematics link up together on the numbered dots. I didn't include a schematic for the power supply because it is completely stock, except that there is a jumper instead of R27 and R28. If you have any questions, please let me know.


dhuebert wrote: I have a pair of eiclone boards and boxes waiting for some finals on iron, tubes and schematics.
Don


I'm open to any suggestions you have on iron choices. The Proto used a Toroid of Maryland PT and a Hammond 1650T output. Magnequest, Plitron, and Sowter have good OT choices, but they are quite pricey. I have not found any other PT choices with enough filament capacity. If you use a Hammond PT (e.g. 278CX), you will also have to use a separate filament transformer and bias supply.

JT
Thermion
 
Posts: 181
Joined: Mon Jul 07, 2003 6:04 pm
Location: Fisher, IL

Postby dhuebert » Thu Jan 12, 2006 8:15 am

Oh sure, now you're gonna make me think.

Ignoring the danger of asking a stupid question, I plunge ahead... I don't remember seeing resistors on the plates of the O/P tubes in any other schematics, or anywhere else for that matter, what is their purpose? Have you tested the circuit with/without them? What results?

I am planning to use 47 ohm resitors in place of the 39's and 100's.


A common value I have seen for the screen grid stoppers is 470 ohm, why so little here? Personally I have used 470 and 1K5.

I'm open to any suggestions you have on iron choices


I like Hammond because they're Canadian, I'll have to have a look.

BTW: Your schematics look very much like mine

http://www.diytube.com/forumpix/power_amp.jpg

This is no longer current, I now have the 8 ohm load on the 16 ohm tap. This increased the power from ~55W to ~70W. I don't have a distortion analyser apart from my ear so I can't tell ya that.

I honestly didn't realize my pet project was holding you up


Nah, you're not holding me up. I guess I'm just a very concrete thinker and without schematics its hard to know what you're talking about.

Don
User avatar
dhuebert
KT88
 
Posts: 820
Joined: Thu May 01, 2003 9:26 am
Location: Winnipeg Manitoba Canada

Resistors and more Resistors

Postby Thermion » Thu Jan 12, 2006 9:05 am

Don,

The grid stop resistors we used were the 1.4 k on each grid. The ones on the screens and plates are unique to paralleled outputs. In PP UL there are typically no resistors used between the screens and UL windings or plates and primary windings. With the paralleled outputs, you make the amp much more susceptible to instability. We tried running it without these resistors, and had trouble with RF oscillations. This was with the input shorted and a 4 ohm resistive load on the output. Adding the resistors solved the problem. Now the prototype is very stable.

BTW, with all of these resistors, the closer to the pin the better. Solder them directly to the tube socket.

JT
Thermion
 
Posts: 181
Joined: Mon Jul 07, 2003 6:04 pm
Location: Fisher, IL

Postby erichayes » Thu Jan 12, 2006 11:32 pm

Hi All,

If I may throw out some ancient terminology . . .

"Stopper" resistors are put in the control grid circuit to prevent, or "stop" parasitic oscillation by lowering the "Q" or efficiency of a potential tuned circuit between the plate and grid of the output tube. Values vary from around 470Ω to 10KΩ, and seem to depend more on the mood or religious beliefs of the designer than anything else. Stopper is a Brit term; I've forgotten what we call it on this side of the pond.

The series resistors in screen and plate circuits involving paralleled output tubes are referred to as ballast or equalizing resistors--the latter falling out of general usage due to confusion with frequency EQ. They allow the output tubes a little wiggle room as far as current consumption's concerned. They also act as fuses to prevent damage to the OT should one or more of the tubes go into 4th of July mode. Values are typically from 10Ω to around 240Ω, again depending mostly on the whim of the designer (I like 47Ω because yellow-violet looks nice).
Eric in the Jefferson State
erichayes
KT88
 
Posts: 987
Joined: Fri Jan 23, 2004 9:01 pm
Location: McKinleyville CA

Postby cfranz » Fri Jan 13, 2006 11:43 am

erichayes wrote:Values vary from around 470Ω to 10KΩ, and seem to depend more on the mood or religious beliefs of the designer than anything else.


Really? Is there consistant differences between say Hindu and Muslem. Do they very across the various branches of Christianity. Catholics, more strict then Protestants? Inquiring minds wanna know... Yellow_Light_Colorz_PDT_09

There could be a research grant in this! A couple of hundred grand from the McArthur Foundation would pay for a lot of PT and OT iron. Yellow_Light_Colorz_PDT_49
Carl
cfranz
 
Posts: 63
Joined: Fri Mar 11, 2005 7:29 am
Location: Outside Chicago

Postby TerrySmith » Mon Jan 16, 2006 5:34 am

Would the trannies from a Dynaco Mark 6 work with the Ikezilla?
T. Smith
User avatar
TerrySmith
KT88
 
Posts: 973
Joined: Sat Jan 31, 2004 12:51 pm
Location: Maryville TN

Mark VI's??

Postby Shannon Parks » Mon Jan 16, 2006 7:10 am

TerrySmith wrote:Would the trannies from a Dynaco Mark 6 work with the Ikezilla?


Hi Terry,

You have a pair of Dyna MK VI's to part out? :o

Shannon
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

MK VI Trannies

Postby Thermion » Mon Jan 16, 2006 9:07 am

TerrySmith wrote:Would the trannies from a Dynaco Mark 6 work with the Ikezilla?


Hi Terry,

Do you know the primary impedance on the Dynaco MK VI trannies? The MK VI used the rare 8417 output tubes, so I suspect the primary impedance would be higher then ideal for a quad of KT88's. You might be able to get away with it depending on the impedance of your speakers.

JT
Thermion
 
Posts: 181
Joined: Mon Jul 07, 2003 6:04 pm
Location: Fisher, IL

Postby erichayes » Mon Jan 16, 2006 3:37 pm

Hi All,

The Mk VI used an OT that was a modified (it had a 4Ω tap) A-450. The A-450 was spec'd at 2200Ω P-P, but the ones I have have measured impedances closer to 2700Ω. The OTs designed for my 100W proto's were 1950Ω P-P and measure around 2150Ω. The amps I harvested the A-450s from were homebrew Mk VIs using KT88s instead of 8417s, and they put out 125 watts at clipping and sounded quite nice, actually, for having a pentode/cathodyne front end. Since the Mk VI is really nothing more than a Mk III with 4 output tubes, and since the majority of VIs were eventually converted to KT88/6550 operation (due primarily to the 8417's penchant for going into explosive ordnance mode at the tip of a hat), the Eiclone should work quite well with the Dyna iron. Bear in mind, however, that there were only around 2000 Mk Vis built, which makes it a rarie (read: valuable). Unlless they're in really rotten condition, you might want to consider restoring them rather than salvaging them.
Eric in the Jefferson State
erichayes
KT88
 
Posts: 987
Joined: Fri Jan 23, 2004 9:01 pm
Location: McKinleyville CA

Postby TerrySmith » Mon Jan 16, 2006 8:47 pm

NO! I don't have any MK-6's. If I did, I would not part them out. I am bidding on a pair of output trannies on ebay.
T. Smith
User avatar
TerrySmith
KT88
 
Posts: 973
Joined: Sat Jan 31, 2004 12:51 pm
Location: Maryville TN


Return to eiclone

Who is online

Users browsing this forum: No registered users and 6 guests