Hi,
I’m wondering what “rules” to follow in order to determine the minimal required mechanical distance between a set of KT-150 tubes and between KT-150’s and 6SN7 drivers. All running Class-A.
Regards, Gerrit
I’m wondering what “rules” to follow in order to determine the minimal required mechanical distance between a set of KT-150 tubes and between KT-150’s and 6SN7 drivers. All running Class-A.
Regards, Gerrit
Between tubes the coupling is mainly electrostatic (For example between output tubes plates and drivers plates), and then the inverse square rule applies. Halving distance will quadruple the coupling capacitance. So, as much distance as you can is the best. However, this contradicts the long distances will augment chassis capacitances. So it is a large previous study of the problem.
No such rule unless RF yet for heat dissipation distance between edges (not centers) = diameter of tube's glass bottle is enough.
Hi,
I checked various commercial amplifiers and noticed the distance between KT150 tubes is only 1.5 - 2 cm. And yes, these tubes can create a lot of heat. It’s only the heat I’m worried about, tubes heating eachother.
Regards, Gerrit
I checked various commercial amplifiers and noticed the distance between KT150 tubes is only 1.5 - 2 cm. And yes, these tubes can create a lot of heat. It’s only the heat I’m worried about, tubes heating eachother.
Regards, Gerrit
M.Jones recommends in his book (D=output tube envelope diameter) D between tube and output transformer and D*1,5 between front end tubes and the same between output tubes.
Last edited:
"The Tung-Sol KT150 glass envelope is a special egg shaped balloon that was developed to improve thermal dissipation." Cooling should profit from a perforated chassis for airflow, allowing close mounting.
Convective cooling of the envelopes is one thing but I would be concerned about adjacent anodes heating each other by radiation.
I will use special sockets with perforation for all tubes, so I can cool all sand regulators under the hood as well. I asked the manufacturer about KT150 mounting 2 months ago, but they never responded.
Just purchased 9 octal socket mounting plates with ventilation holes; 9 of these plates in a row means I need at least 57 cm chassis width. I guess this will give enough spacing and cooling.
Regards, Gerrit
Regards, Gerrit
If I understand you correctly, you are planning to mount all your socket plates butted against each other, and thus drop spacing to smaller than that suggested by Morgan Jones. I would strongly recommend against this. Making this compromise just to avoid some additional cutting or drilling is a bad idea.
The tube socket plates measure approx. 6.2 x 6.2 cm, I guess this will work out fine. Yes, 9 in a row. I can’t wait to start building!
The KT88 GEC datasheet recommends 4 inches apart centre to centre. See page three for details. https://frank.pocnet.net/sheets/084/k/KT88_GEC.pdf
If any of those were Bobby Carver's designs he runs the output tubes at a ridiculously low bias current. Definitely not class A.
Yes you may be right but given that you have asked for advice, and haven't started your build yet, why not space the valves as far apart as possible? It will be you that has to pay for new replacement valves, not the manufacturer or fellow DIY'er.
Because it'll look like a 5yo did it with one tube in each corner. Use the plates and call it a day. There are many other factors that cause tubes to fail. Excessive heat is pretty far down the list if you use Morgan Jones and common sense as a guide. If you've ever seen the audio amp in an old film projector (Kodak "Brownie" or Bell & Howell Film-o-sound comes to mind), the tubes are almost touching and they are housed inside a tight enclosure with meager air circulation. I wouldn't pack them that tight in a DIY audio amp, but just to give you a frame of reference. Those tubes lasted years. As long as there is ample air circulation you should be fine with those plates with a little bit of room between. My $.02.
A fan to blow some air up past the valves helps. My 120W monoblock chassis get injury hot with six EL34's after 1 hour. Valves are two parallel rows 11cm apart, each valve 7cm apart.
Yes the fan makes noise, about 60dB, but this isn't heard when music is played. I tried without fans, but both chassis and mains tfmr ( chassis gets hot, mains tfmr though 2.5cm below chassis gets too hot) so like everything in engineering it's a compromise.
I'd seriously advise you to give a lot of thought to thermal design in your build, the best way to test is to build a chassis from scrap sheet or even plywood ( with part of the chassis covered in metal to see how reflected heat effects it), then take some temp measurements
Have fun with your build, Andy.
Yes the fan makes noise, about 60dB, but this isn't heard when music is played. I tried without fans, but both chassis and mains tfmr ( chassis gets hot, mains tfmr though 2.5cm below chassis gets too hot) so like everything in engineering it's a compromise.
I'd seriously advise you to give a lot of thought to thermal design in your build, the best way to test is to build a chassis from scrap sheet or even plywood ( with part of the chassis covered in metal to see how reflected heat effects it), then take some temp measurements
Have fun with your build, Andy.
I've used some SilenX low noise fans to fix some noisy Hoefer HV supplies, and you can hardly hear them without any audio.
https://www.ebay.com/sch/i.html?_from=R40&_trksid=m570.l1313&_nkw=silenx+fan&_sacat=0
https://www.ebay.com/sch/i.html?_from=R40&_trksid=m570.l1313&_nkw=silenx+fan&_sacat=0
That looks like a typical chinese made amp.
"Hi end, commercial & Chinese", = oxymoron.
I don't understand why people don't look at historical amps.
Eico, Scott, Dynaco, Leak, Quad give you a perfect idea of proper practice, or where they decide to cramp it.
Scott insisted on jamming it all inside a wooden cabinet, which gets way too hot.
That with the 4 x 7591stuffed in there still managed to get some brilliant OPT but sacrificed to an undersized PSU, and way too much heat.
If you want the best of the best example using multiple KT88, take a look at a HIWATT guitar amp, esp the 200W and 400W versions (real=300W).
Not only were they uber-powerful, but they run the stuff on the limit, and their fantastic wiring quality was a lesson to practically every serious amp builder in the world, and has never been bettered.
I don't understand this fixation from looking at "hi end" modern amps, they are mostly a pile of crap, and rarely perform even close to their claimed figures.
6vheater,
The sample image is from PrimaLuna... Not typical Chinese I guess although I have no personal experience with this brand. I’m quite confident they know how to build a good amplifier.
I’ve actually build tube guitar amplifiers before, so I know what I’m talking about. I build RF amplifiers with TB4/1250’s and even one with a TBW6/6000 water cooled tube, with a 5000 volt @ 1 amp PSU. All homebrew equipment...
Regards, Gerrit
The sample image is from PrimaLuna... Not typical Chinese I guess although I have no personal experience with this brand. I’m quite confident they know how to build a good amplifier.
I’ve actually build tube guitar amplifiers before, so I know what I’m talking about. I build RF amplifiers with TB4/1250’s and even one with a TBW6/6000 water cooled tube, with a 5000 volt @ 1 amp PSU. All homebrew equipment...
Regards, Gerrit