• 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.

Compact All-In-One Chassis Design with good ventilation. And low noise, Not a Schemat

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
This is a rather quick sketch of an all-steel design with single-90mm fan forced ventilation and vents upon the underside and back of the chassis.

Paint is my friend :p

Anyway thought I would spread the idea...Hopefully I can use it or someone else can use the layout.

I have thought of rebirthing the old idea with valve amp chassis design where the valves are placed horizontally for both better heat dissipation and compactness as I have placed the KT90's on the bottom directly underneath the 6SN7's and faced them 'towards' each other..

The 6 large green boxes at the front of preamp are PIO.. and all transformers are Torodial Plitron...

All those wittle yellow pixellated resistors are to accomodate components...both resistors on blocks and capacitors.

This is not usually my idea of a schematic :p It's a block diagram (stretching it)...

With a heap of steel and a couple of KT90's this chassis could very well work...

I was hoping to see if this design would work against or for noise in both preamp and amp sections.

The sacrifice for a compact design is that the power supply cannot have a large base, therefore electrolytes must be used...

Cheers.
 

Attachments

  • my amp design-3.jpg
    my amp design-3.jpg
    48.7 KB · Views: 519
Fifth revision of the chassis sketch..

I have added a total of 6 electrolytic (got it right now-was tired :p)
caps in series for double volt ratings and a pair of chokes.

Furthered the distance from each Transformer and as the input
stage is furthest away from the power supply while being in the
same chassis I am quite happy to go ahead with a more specific &
accurate (by a mile) paper sketch to stamp out a steel base
for this revision :p

I have changed my mind about the all-steel chassis instead having
an aluminium case ontop of a steel one-piece base with all holes
drilled, so I don't really need to worry about magnetic interaction
(too much).

14th revision of the power supply :p

It's quite a simple power supply when you think about it.

C1 is two 100uF Jensens in series, so is C2..

C3 however is a single pair of 220uF Jensens in series to make
110uF.

All caps are 'rated' at 1200v... Should be enough :p

Here is the PSUD simulation.

Cheers. I think this will be my last post upon this matter.
 

Attachments

  • psud-rev14.png
    psud-rev14.png
    49.1 KB · Views: 330
Hint:

Design the electronics of the amplifier before you start worrying about the layout.

I suggest that you re-think mounting the tubes in the horizontal plane.
While this improves the cooling (on one side of the tube only), it puts great stress on the internal structure.
There is a reason why nearly all amplifier designs have the tubes mounted in the vertical plane with their bases down. Mounted that way, the heating of the glass envelope and the internal elements is symmetrical.
 
Which is exactly why I chose KT90's instead of KT88's....

The KT90's are a very long and cool operating tube from my understandings when placed at KT88 limits...

But even so, I can get plenty of power 50watts, out of a pair..

So that's 100-130 watts dissipation between the two... big deal.

Please tell me otherwise if i'm wrong :)

I could always place the 6SN7's at a diagonal position so they're also further from the surface of the KT90's...

I understand what you say, the internal structure of the KT90's would become heat stressed as all the heat moves to one side of the tube, the one facing up... which could turn a plate or grid red.

I'll give the 6SN7's placement another think...

Thanks for the heads up, that's what I'm here for.

Theres these two doctors doing it in a bar, anyway after a few minutes, the lady is screaming and all, she says "Oh look honey, fireworks!", and the guy says "Yeah aren't you glad, we got front row seats!" - Layberinthius
 
diyAudio Senior Member
Joined 2002
Hi,

The KT90's are a very long and cool operating tube from my understandings when placed at KT88 limits...

At the end of the day the dissipated heat will still be about the same, possibly even more.

The length of the KT90s is certainly not an advantage either when mounted horizontally putting more stress on the bases too.

I'm not saying you can't operate these tubes safely this way, just that the reasoning behind it seems flawed to me.

If you want to run them cooler, why not build in a fan at the bottom blowing air against the bulbs?
Any fan from a broken computer PSU would do nicely and if that's too noisey, add a rheostat to set the rotational speed.

Just a thought...

Cheers,;)
 
Yeah it looks like it I may have to have two fans...

SO another 90mm, that brings it up to a total of 30dB or thereabouts.

The fans that I have, have a rotational speed (aka RPM) of 2400...

So I don't think I'll see much noise as it is, it is a completely silent fan at 12v already...

Of course once I put a dust/hairball grill on infront the noise is going to go, up up up and away!

Depends also upon the distance between the fan and the grill, Maybe I can make up a 'tunnel' bracket so I can space it a tiny bit between the opening of the chassis, to further reduce noise.

Thanks guys :D

:p
 
Instead of pulling air from inside the case to vent outside why not do the opposite, pull air from outside and through the case, slight pressurization of the case?
With the help of some baffles to direct airflow particularly over the tubes you may be surprised. You can use cardboard for the baffles (I like Fedex envelopes for this, the paper used is stiff enough and you can't beat the price ;)
Use a filter over the fan intake, in that way when you pressurize the case it does not get dirty inside. You will also keep the noise level lower.
Just a thought.
 
Hi there....for Hi Fi listening those fans be quieter than the amp S/N ratio otherwise there's no point in going for a super spec. On some apps I use Micronel very low profile, and as others mentioned run fans on LV. One doesn't need to create a <gale> through the chassis to cool, you'll be suprised what a small breeze can get rid of and while listening one can also turn the radiator off in the room.

C1 on that PSU simulation better be a pretty good quality cap; Those 4007's for the PSU...... I'm not sure of the reverse leakage rating for the standard IN4007 are at high voltage and temp.
The ratings on simulation may be sound, but in the real world without supression caps, and polluted AC supply, I wrap this up for a well rated square bridge type KBU series or sim. Discrete arrangement semi's are generally more noisier than a 4-in-one encapsulated die.

Series elect caps are a headache....sometimes unavoidable....remember... use equalisation resistors across them.

rich
 
Hi there....for Hi Fi listening those fans be quieter than the amp S/N ratio otherwise there's no point in going for a super spec.

They will be, and so will be my entire room, by the time I'm done
all walls will be covered with Sonex noise reflection foam..

It still wont do anything for Bass, but I don't listen to many bass
sources anyway :p

One doesn't need to create a <gale> through the chassis to cool, you'll be suprised what a small breeze can get rid of and while listening one can also turn the radiator off in the room.

Yup, As I said 2400 rpm at stock I should be able to vary it down to 6-7volt with a pot and make a smooth Sailor :)

C1 on that PSU simulation better be a pretty good quality cap; Those 4007's for the PSU...... I'm not sure of the reverse leakage rating for the standard IN4007 are at high voltage and temp.

I was actually planing on using BY 127's and a pair of heatsinked 400v bridges for filament.

Is there any way I can prevent all the poop coming from the grid
other than placing a UPS in-line?

Series elect caps are a headache....sometimes unavoidable....remember... use equalisation resistors across them.

I'll look up how to do it but it's good to hear from more reliable designs...
 
Some advice

Layberinthius said:


They will be, and so will be my entire room, by the time I'm done
all walls will be covered with Sonex noise reflection foam..

It still wont do anything for Bass, but I don't listen to many bass
sources anyway :p



Yup, As I said 2400 rpm at stock I should be able to vary it down to 6-7volt with a pot and make a smooth Sailor :)



I was actually planing on using BY 127's and a pair of heatsinked 400v bridges for filament.

Is there any way I can prevent all the poop coming from the grid
other than placing a UPS in-line?



I'll look up how to do it but it's good to hear from more reliable designs...

Ditch the fans , they're more trouble than they're worth , unless you're going to take the p*ss out of the output valves . Build the chassis as big as possible . If you build an amp thats too small , it may limit future modifications and make layout a nightmare . Use perforated sheet for a base and keep the design simple . Weigh all of the iron on bathroom scales and decide on the top plate thickness (I'm now on my 5th attempt at a monster class A amp , each time the chassis was too puny for the 35kgs of iron!) If it's a first attempt it would be a good idea to keep the HT choke well away from the input valves , space the output valves evenly across the width of the chassis and keep a few inches between them and input valves . Hope this helps . Why use an uninterrupted power supply (UPS) ? Better off using a mains filter and snubbing the choke and rectifier diodes .

316a
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.