• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

TUBENEWB - Quad KT88 Monoblocks

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hey everyone...
went for a listening sesion this past long weekend and fell in love with the sound a friends twin KT88 setup ( not sure exactly what hes got but its a homebrew of sorts )
which got me thinking at possibly starting another amp build :p
Something that has a lil bit more "punch" than his as it sounds a bit "weak" down bottom. Thinking of a Quad KT88 in a PP configuration.

I know good iron is expensive, but what even worse is trying to build a tube amp when you know nothing about tubes !!!

Found output trannies at hammond, but dunno what power requirements would be for such a monstrosity...
 

taj

diyAudio Member
Joined 2005
I know good iron is expensive, but what even worse is trying to build a tube amp when you know nothing about tubes !!!

You'd better stick to a project that is well tested and well supported. At least until you learn about tubes.

Yeah, Quad KT88's (parallel PP) is a little extreme isn't it? 2 usually nets you about 75 watts. If you need more, maybe using tubes isn't in your destiny.

If your pal's rig didn't have enough bottom end grunt for you, it may just have been a design limitation of the amp he built. There are plenty of [cheap and easy] methods of reducing the bottom end. Cheap output transformers is a good start. Too small coupling caps, etc. etc.

..Todd
 
Nothing wrong with overkill, at least in my point of view...
The way I always look at it is to have plenty of headroom :D

Thanks for the Edcor links but already have a local supplier that can get me both Hammond Outputs for well under $250. The ones im worried about is to find the specifics of what kind of current is needed for 4 x KT88's as i always like a "stout" power supply as its the foundation of every amp.

What I'm trying to figure out is where i can find a stable circuit that i can start with and slowly gather parts. In the meantime i have ordered the sockets and 2 sets ( quad ) of matched Gold Lion KT88's through a "friend" a that lives a couple of blocks away ;)

Any ideas or pointers ???
 
there are plenty of competent designs for the KTs. Similarly, lots of horror stories!

If you are going for a quad, power supply is going to be... well, you've already identified it as critical.

If you are spending like a drunken sailor, there are the usual suspects in terms of voltage amp/splitter/driver set-ups. If you want to get a bit more funky, a few here will help you in the "how to drive a high power output" stakes.

Check out chrish's now infamous 6L6 thread which has almost become a primer in pentode circuit design.
 
Last edited:
Triode Dick has a PPP KT88 amp. Look under projects and KT88 (surprise!) and on the second or third page he has the PPP version. I have not built it so have no idea what its like.

I did build a stereo KT88 PPP amp ~ 20 years ago, UL using 100W Massicore OPT's (not really the correct OPT's for PPP but what the hey and 800V on the anodes with the then new "Golden Dragon" tubes from Shuguang (sp). Lots of pops and clicks and pretty colours until I lowered the voltage / current. I used them to drive Infinity Kappa 8's. The lights in the apartment would dim with the bass transients. Great fun at the time but as others have questioned, do you really need KT88 PPP or is it something you want to do.... just because?
 
The way i want to set it up is to have the heater transformer separate from the HV circuit.
This will also allow me to run a regulated heater supply along with a preheat function to the tubes for 15-20 seconds prior to blasting them with HV.

So with a bit of quick math...
1.6A per tube will need a transformer capable of 6.4A without any other tubes in a circuit !!
I would say a 10A should be plenty of juice for 8 tubes and a few more left over for "driver" heaters.
 
Old PPP KT88 design of mine from a few years back but may give you some ideas.
I ran these monoblocks with both Hammond 1650T and some expensive Plitron PAT4006 Torroidal Output transformers and actually liked the sound of the Hammonds better in some regards.
See here:
Evatco Hammond Project Page
4th project down.
Schematics for amp and power supply and a "designers comments" document I put together which included some power output and frequency response measurements.
Cheers,
Ian
 
Old PPP KT88 design of mine from a few years back but may give you some ideas.
I ran these monoblocks with both Hammond 1650T and some expensive Plitron PAT4006 Torroidal Output transformers and actually liked the sound of the Hammonds better in some regards.
See here:
Evatco Hammond Project Page
4th project down.
Schematics for amp and power supply and a "designers comments" document I put together which included some power output and frequency response measurements.
Cheers,
Ian

This DEFINATELY meets the built and tested part :D
Thanks !!

What did you use for the power transformer ???
 
Sorry to say -
I worked out what I needed and had a local transformer manufacturer wind them for me.
For +450V HV you will want 340V AC @ 500mA, I specified 370V @ 700mA and ended up with HV of a bit over 500V which meant that the amp was capable of more power than the 120W rating of the output tranny. I had multiply heater windings since with that design I need to float the driver tube heaters on top of +200V and the top tubes of the cascode diff amp was floated on top of +100V.
When you finalise a design - choose something from the Hammond (or other) range to give you the HV voltage and current required and then add an extra small tranny for the extra heater supplies and/or the bias supply.

Of course I would do that amp a little differently today which would change the power supply requirements.
I would:
replace the 6SN7 driver tube with 4 off MOSFET Source Followers, 2 each side, each direct coupled to a KT88 grid - that requires a -ve supply of about 3 times the bias voltage. MOSFET Source followers would be current source loaded.
Individual tube bias would then be applied to the mosfet source follower gates and these gates would also be capacitor coupled off the cascode diff amp anodes (2 each side). That deletes the requirement for the driver heater supply.


Cheers,
Ian
 
Last edited:
hmmmm...
What if i want to push a bit more juice into the KT88's ??
More like 550-600V ??
As far as the "hybrid" front end...
I do have 5 pairs of THESE from a previous project :p

My "other" build uses 8 pairs of these boys on a single channel pushed by a dedicated 1KVA toroid and 90000uF of capacitance with 55V rails ;)
 
Last edited:
You would need to choose a different Output transformer.
1650T will saturate below 30Hz at 120 Watts

550 - 600V on the KT88 will give you around 150 -160W and the 1650T (at that power) will saturate below about 45 -50Hz.

If you go to the next tranny up in the Hammond range, the 1650W, it is rated at 280W BUT it has a lower primary impedance, that means it doesn't want more voltage swing, it wants more current swing and you get that from more parallel pairs of output tubes - not running the ones you have a higher voltage.

As for "issues" - who here has'nt?

Cheers,
Ian
 
Called EDCOR today for a quick talk with the sales staff there...

Got quoted $109 for a custom Power Transformer ( 360-0-360 @ 650mA HV and 50V @ 100mA BIAS windings ) as i want the filaments to turn and warm up prior to the application of HV to the tubes.
There will be a separate rectified and regulated heater supply as that will feature a 5-10 second warm up prior to powering up the HV transformer.

Will be posting a quick mock up of the Power Supply latter on this evening.
Any design I will decide to go with, will need a decent power supply to start off with !
 
Hey guys...
I got really busy the other night and didnt have a chance to get near the laptop.

We had 2 planes come in needing a full inspection on the APU's and service as well.
Anyways...

Here's the schematics...
Filament Supply and timer circuit:
An externally hosted image should be here but it was not working when we last tested it.


HV Supply:
An externally hosted image should be here but it was not working when we last tested it.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.