HK10 - Stereo KT88 / 6550 Push-Pull Amplifier

the thermionic watercooler

Postby EWBrown » Sat Dec 27, 2008 2:11 pm

SInce that amp's driver and initial LTPI stages run from 325V B+, it should be a good candidate for driving EL84s using the standard ST35 voltages. (365 to 380V B+ for the power tubes, and 325V for the VA and LTIP section)

I ran a sim on tubecad, with 12AU7s and with 6CG7s, and the 6CG7s come out a little better. The overall gain with 12AU7s seemed to be marginal for driving 6550s or KT88s to full power, but they have more than enough drive (even without the CF stage after the LTPI) for driving PPP EL84s.

Alas, it can exist only on paper for now, but next year...

/ed B
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby Brik » Mon Dec 29, 2008 10:15 am

DEC Littleton was a relatively sedate place compared with bustling "The Mill".
We could have used a 12lb turkey that DEC used to give out to each employee around this time every year. Yellow_Light_Colorz_PDT_12

deicide67 wrote:Does anyone know what the voltage swing on the poseidon boards is? I'm not too limited on space, but I also want to keep it simple.


At B+ of 440V (max for in the triode mode according a spec sheet), a pair of triode-connected KT-88s/6550s would require an input signal of around 70V-80V peak-to-peak to achieve full power.

In a fixed bias scheme you want to have a grid-to-ground resistance that is much lower than 470K that is used for a cathode bias scheme.
For example, in the HK-10 the bias voltage is applied to the KT-88's grid via a 100K resistor from the "fixed bias adjustment" Yellow_Light_Colorz_PDT_17 resistor network. Therefore, each KT-88 has a grid-to-ground resistance of ~110K or so. Both Citation II and Mark III use a similar scheme.

Looking from the driver's side, the tri-paralleled output stage will present 1/3 of the impedance of the non-paralleled output stage which makes it a more difficult load to handle. Which brings to Ed's post regarding the PPP KT-88 amplifier and its driver stage:

EWBrown wrote:I ran a sim on tubecad, with 12AU7s and with 6CG7s, and the 6CG7s come out a little better. The overall gain with 12AU7s seemed to be marginal for driving 6550s or KT88s to full power, but they have more than enough drive (even without the CF stage after the LTPI) for driving PPP EL84s.


Can the Poseidon board drive the Tri-PPP KT-88 stage to its full power?
I have to get back to you on that unless someone already knows the answer. Yellow_Light_Colorz_PDT_12

/b
User avatar
Brik
 
Posts: 204
Joined: Fri Apr 07, 2006 5:30 pm
Location: West of Boston

Postby Blair » Mon Dec 29, 2008 10:29 am

Well,

I did not look at the max B+ rating for the 6550! That kind of puts a nasty damper on things. I was planing a B+ of 510-530v.

Not knocking my own skills, but this project may very well be a little too much to tackle for me. I have the tubes and sockets coming, and was planning to order output trannys this week but this is getting to the point where I may need more than a steady hand and the gear to build it.

Blair
Blair
KT88
 
Posts: 1272
Joined: Thu Nov 09, 2006 10:22 am

Postby Blair » Mon Dec 29, 2008 1:28 pm

I take that back. Looking at the spec sheet for the Svetlana 6550c, it looks like 450vdc for the plate. someone may be able to answer this, but when I actually put the correct current draw for this amount of tubes in PSD II, it pulls my voltage down quite a bit. Is there something I am doing wrong? I'm using 150mA X 3 thus 450ma. I cannot see even 6 tubes pulling this toroid down that much. If it is accurate, then the B+ will be around 421v with combination CLCRC filter. That would be perfect.

Blair
Blair
KT88
 
Posts: 1272
Joined: Thu Nov 09, 2006 10:22 am

Postby Brik » Mon Dec 29, 2008 7:32 pm

Hi Blair,

A triple-PPP 6550 can output 90W into an 8-ohm load with B+ at 450V.
Here are the simulation results from TCJ PP Calculator:

http://home.comcast.net/~brikma/HK10/TriPPP_6550.pdf

/b
Last edited by Brik on Sat Mar 20, 2010 10:16 am, edited 1 time in total.
User avatar
Brik
 
Posts: 204
Joined: Fri Apr 07, 2006 5:30 pm
Location: West of Boston

Postby Blair » Tue Dec 30, 2008 8:33 am

Thank you! I would assume this would slightly increase into a 4 ohm load. Whereas I really want to build these amps, it becomes an issue of tradeoff. I already have an amp pushing 65wpc or so with UL PP KT88s. Is the extra 25 watts going to really make a difference for the power consumption? I think if I were to use a pair of 200w UL trannys it would. Then I could say I have a 180-200w UL / 100wpc Triode amplifier. It is a tough call to run 12 X 6550s at once for 100w without taking full advantage of the dissipation ability of the tubes.....It would be different if this were a smaller amp on the cheap, but even heavily budgeted, these amps will cost around $1K to build.

Thank you for your help!

Blair
Blair
KT88
 
Posts: 1272
Joined: Thu Nov 09, 2006 10:22 am

Postby skidave » Tue Dec 30, 2008 12:36 pm

Brik,

Nice job on the project! What type of potentiometers did you use for the bias circuits? Do you have a supplier and part number?

Thanks, Dave
User avatar
skidave
 
Posts: 141
Joined: Fri Jan 06, 2006 2:04 pm
Location: York, PA

Postby Brik » Tue Dec 30, 2008 1:59 pm

Dave,

Thank you for the kind words. :)

This trim pot has its own thread from "kheper":
http://www.diytube.com/phpBB2/viewtopic ... highlight=

It is from Mouser Electronics, part number 72-P16NP-10K:
http://www.mouser.com/Search/ProductDet ... LvcQ%3d%3d

The bill-of-materials for the HK10 is here:
http://home.att.net/~Brik/HK10/HK10_BOM.xls


Blair,

I would go for more power also.

Cheers,

/b
User avatar
Brik
 
Posts: 204
Joined: Fri Apr 07, 2006 5:30 pm
Location: West of Boston

Postby Brik » Mon Mar 28, 2011 7:29 pm

Update:

I replaced the CCS for the long-tail-pair inverter from LM317 to LR12.
The LM317 works well when it is used to control medium to high current (10mA to 1500mA) and when the voltage differential across the device is less than 40V.
In this LTP application, the CCS is expected to control a small current (4mA to 8mA) and a voltage differential across the device in the 70V~85V range. So, I was operating the LM317 at the lower extremities of the effective current range while dodging the voltage differential requirement by adding a 9K Ohm tail resistor. Clearly not an ideal application of this device.

The Supertex LR12 turned out to be a more appropriate choice for this duty:
Supertex LR12 General Description:
The Supertex LR12 is a high voltage, low output current,
adjustable linear regulator. It has a wide operating input
voltage range of 13.2 - 100V. The output voltage can be
adjusted from 1.20 - 88V, provided that the input voltage
is at least 12V greater than the output voltage. The output
voltage can be adjusted by means of two external resistors
R1 and R2 as shown in the typical application circuits...

Image

As you can see, in many low-to-medium current (< 100mA) situations, you can replace the LM317 with a LR12 provided that you can live with the voltage differential requirements (13.2V < Vd < 100V).
/b
User avatar
Brik
 
Posts: 204
Joined: Fri Apr 07, 2006 5:30 pm
Location: West of Boston

$ 78.00 Power Transformer

Postby Thor61 » Mon Sep 19, 2011 11:10 am

Hello guys, Looks like a really good project.... I have a Fender transformer from a cheap garage sale amp, see spec > http://www.classictone.net/40-18004.pdf also here are some pics > http://www.classictone.net/40-18004.html http://www.classictone.net/40-18004_side.jpg My question is will this work for this amp with little or no trouble ? I plan to use # 6550A tubes, and after going back through the thread I just noticed the Hammond 290FX. The Fender amp it came from had a quad of KT88's and 6 other smaller tubes....
Thor61
 
Posts: 50
Joined: Sat Feb 13, 2010 9:31 pm

Re: $ 78.00 Power Transformer

Postby Writer Frog » Tue Sep 20, 2011 10:46 am

Thor61 wrote:The Fender amp it came from had a quad of KT88's and 6 other smaller tubes....

As far as the B+ rail capacity goes, 320V-0-320V at 450mA should be good enough for the original design.
The heater capacity of 6.3V at 5.5A, however, comes up short to support a quad of 6550A,(4 x 1.6A = 6.4A).
You will need an additional 6.3V transformer to supplement the Fender transformer.
Also, I think one can get away with a smaller power supply choke, say 3H rather than 5H, without degrading the sound.

Please keep us updated of your progress. O:)
/Matt
User avatar
Writer Frog
 
Posts: 95
Joined: Mon Mar 06, 2006 9:17 am

HK10 - Stereo KT88 / 6550 Push-Pull Amplifier

Postby Keepsake10 » Mon Aug 11, 2014 8:13 am

Hi Brik,
Is a long time since you presented this project - hopefully I'm not too late to get some answers....
I was looking for a schematic for KT family tubes and I found this one, nevertheless some of the components I have available are slightly different - therefore I am a little concerned. Tube choice: 5814 (same as ECC82 but MIL specs) for the first stage. The LTP get an European ECC813 (direct equivalent of 6463 - lower Ri 3,8K instead of 7,7K; higher S 5,2mA/V instead of 2,6mA/V and Ia 14,5mA instead of 10mA)... all in one a promising tube. CCS for the tail is a cascade mosfet from K&KAudio. For the power tubes I have from a previous project four KT120... Transformers are all from Lundahl - LL1650 for power supply and LL1623PP as OTP.
I would go for monoblock from electrical point of view but twice on a single chassis. The Lundahl 1650 have plenty of power (350V@630mA and 4* 6,6A@3,1A) but is missing a bias winding... For that reason I had to go for a separate bias supply - and I choose the automatic bias supply from Tentlabs.
I put everything together on a test bench ... and the results where not as expected.
The math is done, but when I start: the anode R for the first stage and as well for LTP get really hot... The anode voltage of the first stage is higher as it should and do not bias correctly the first G from LTP.. I was recalculating the R's for A and K again and again... always worked but the sound was missing details - this is unusual for Lundahls... I suppose the different CCS and tube for the LTP might be the reason, but here is where my knowledge is limited...
Any thoughts are welcome.
Thank you and best regards,
Dan
Keepsake10
 
Posts: 6
Joined: Mon Aug 11, 2014 2:49 am

Re: HK10 - Stereo KT88 / 6550 Push-Pull Amplifier

Postby Brik » Mon Aug 11, 2014 5:33 pm

Hi Keepsake10,

I am sorry to hear that things are not working out well with your project.

I am trying to get some information on the ECC813 to see what might be going wrong.
Since I am on the road with only my tiny netbook, I do not have the usual access to my handbooks,
simulation tools, and electronics database access rights, so I will have to do some data mining to get
information needed to be able to answer your question properly.

It does appear that the LTP may not be within an optimal operating envelope.

In the mean time, can you post your schematic with revised passive component values
as well as voltage measurements at anodes, cathodes, and grids of the first and the LTP
stages?

Thank you in advance.
/b
User avatar
Brik
 
Posts: 204
Joined: Fri Apr 07, 2006 5:30 pm
Location: West of Boston

Re: HK10 - Stereo KT88 / 6550 Push-Pull Amplifier

Postby Keepsake10 » Tue Aug 12, 2014 1:42 am

Hi Brik,
Initially the schematic was exactly the same.. only for the power supply I made a PCB including a delay with NE555 + relay and after bridge rectifier a Common Mode Line Choke. First two stages are DC heated. As I was not happy with the results, I started recalculating and making new many tests. For instance I increase / decrease the anode currents, reconsidering Ra with the hope I can match the LTP with the first stage... the best I got was a bias of -6V on the grid of the LTP ...
I will post the schematic including the changes I made and actual measurements. One thing I didn't made was to use the NFB...
As regards to ECC813 - this is a direct equivalent of 6463... both of them kind of 12AU7 on steroids with lower Ri.
Thank you and best regards,
Dan
Keepsake10
 
Posts: 6
Joined: Mon Aug 11, 2014 2:49 am

HK10 - Stereo Push-Pull Amplifier modified KT120

Postby Keepsake10 » Tue Aug 12, 2014 12:59 pm

Hi Brik,
in the meantime I came to some positive results... nevertheless I have the feeling I coud get some more "microdetails" to the sound. The rest is all there.. including some hummm ';~) this might be the reason I believe the amplifier could be improved regarding microdetails.
Here is the actual schematic, including basic measurements... didn't had the time for a deeper research on the scope.. but will be done as well.
As you see I did not used yet NFB... and if not necessary I would avoid it...
If you have the time give a look and let me know what you think about.
The Auto-Bias Supply - http://www.tentlabs.com/Components/Tube ... age24.html
MEC in the end of the power supply came from the same supplier...
CCS is cascade mosfet from K&K...
KT120PP - power supply.JPG
KT120PP - power supply.JPG (103.66 KiB) Viewed 29801 times

KT120PP - amplifier.JPG
KT120PP - amplifier.JPG (111.29 KiB) Viewed 29801 times
Keepsake10
 
Posts: 6
Joined: Mon Aug 11, 2014 2:49 am

PreviousNext

Return to diy hifi

Who is online

Users browsing this forum: No registered users and 8 guests

cron