[Redge78]

  In my amp after power is on the ventilators doesn't work. The amp is heating up for approximately 40 min (winter/spring conditions) and when temperature reaches 50 degrees the ventilators turns on cooling down the temperature to 40 degrees.
When I was planing the modification one of the priorities was to have good ventilation. Just to remind some photos of the casing done by me.

Mogos, I really like the aesthetic of your amp.
The only thing is that bothers me are the little holes you've got on the bottom box, they must **** up your ventilation path. The air is supposed to exit from the holes around the tubes ... Pretty sure we already talked about that earlier on, when Maxx sealed his bottom plate (and lost 10° (Celsius ?) ) in the process.
But maybe you already took care of that, told me and I forgot ... would not be that surprising !
 
Anyway, when my large mods are completed, I may come back to you and copy you shamelessly. 
That is if my plans of a fluorescent acrylic box is too complex/expensive to do.
 
@Mogos
Anothe subject : I have read that you have a paul Pang V2 USB card. I was planning to buy one myself. Can you tell me what you think of it ? That would be much appreciated/helpfull.
 
[EDIT] The potential problem you may have is that some areas of the amp may be significantly hotter than where you're taking your temp. I would (warmly

 ) recommend to always have at least a bit of forced air through the amp. It's noiseless and good preventive maintenance. 

 

[coinmaster]

Okay I solved the plight I was having with the Grid drive mod implementation. It seems my memory of transistor biasing has gone to trash.

 
Same as before but you connect R3 to the opamp output instead of ground. 
The 20v battery indicates the opamp and the two 470k resistors are the grid leak/buffer resistors in the amp between the opamp and grid.
 
So essentially like I said before, switch out the capacitor leg and the source leg and connect the capacitor leg to the gate of the mosfet.
Connect the drain to B+ of whatever stage and connect R3 between gate and the opamp output (pin6).
 
I could have sworn R4 and R2 are too high of a value for it to work but spice is telling me the otherwise. 
 
Of course it might actually be better to connect it to C2 as an AC path to ground for the gate current like this
 

But you're going to have to ask MrCurwen for the fine details like that, I only know the overall concept.
He is giving me the silent treatment so you're going to have to ask him directly.
 
 
 
As for the force balance mod you would have to splice apart or rip out the copper trace between the cathodes in the input stage for it to work but it's not hard to reconnect them whenever.

 

[Maxx134]

I have paper in oil metal case caps in tube amps.

I use vitaminQ caps as my second choice cap after the Jupiter copper.

Yes Celsius. I forgot that it is an international thread. And celcius was so obvious for me. I live in Poland Europe  .

The amp doesn't have to suffer in heat with the "warm-up" time you like to give your tubes. ..

Personally I want the whole amp to be as cool as possible inside and have my fan on even before I turn on the amp.

The tube are far away on top and are not affected by the fan.

The PSU generate a lot of heat, as does the sockets.
So we need the inside as cool as possible.
I would never wait for insides to get hot..

My fan is blowing directly onto my caps.
The air is forced out top because I taped holes on bottom I had...
Only way to remove heat buildup because of lack of easy air flow.

 

[Mogos]

Maxx your assumptions are correct. In case of my solution the temperature reches level wich I have observed when I had the stock solution (smaller fans turn on from the begining). The 50 Celcius

 degrees is a temperature measured in hottest points. I have two sensors. So the conditions around the precious caps are better then in the stock solution. Additionaly the ventilators does not stop after the power is turned off. They are turned off automaticaly when the temperature reaches 35 Celcius degrees. With this solution I see two advantages. The amp as a whole heats up faster (and as john57 pointed out the tube amp needs to heat up to spread the wings). Second maybe not so relevant is that if I am listening to the music for not so long then the ventilators don't start at all (winter time).

 

[Mogos]

  Mogos, I really like the aesthetic of your amp.
The only thing is that bothers me are the little holes you've got on the bottom box, they must **** up your ventilation path. The air is supposed to exit from the holes around the tubes ... Pretty sure we already talked about that earlier on, when Maxx sealed his bottom plate (and lost 10° (Celsius ?) ) in the process.
But maybe you already took care of that, told me and I forgot ... would not be that surprising !
 
Anyway, when my large mods are completed, I may come back to you and copy you shamelessly. 
That is if my plans of a fluorescent acrylic box is too complex/expensive to do.
 
@Mogos
Anothe subject : I have read that you have a paul Pang V2 USB card. I was planning to buy one myself. Can you tell me what you think of it ? That would be much appreciated/helpfull.
 
[EDIT] The potential problem you may have is that some areas of the amp may be significantly hotter than where you're taking your temp. I would (warmly

 ) recommend to always have at least a bit of forced air through the amp. It's noiseless and good preventive maintenance. 

It may look like the airstream may be disturbded by the holes. But the effect is different. The heated air always run upstream and the tubes are the motoric force for it. The little holes act for suction of the air not dissmisal of it especcialy when the ventilators are off. I don't have any measuring devices but I have used very thin paper to check which way the air may travel trough the wholes and it looks like the chimney effect is working 

 with fans off. (of course it is very weak effect but at least telling me that the air is not pushed out). The air stream forced by the vetilators bounces against the PCB and for sure there may be tendency to run side ways and run away trough the wholes. But the same measuring with the thin paper dose not show it. Any way when the ventilators are on the temperature is stable and reaches the 40 Celcius degrees. As I have alredy written I have placed the sensors in the hottest points so I have good control of the temperature inside the amp 

. And this is what it shall be.
 
I can recomend the use of the Paul Pang V2 USB card. Maybe sonically it is not day and night but you are gaining few advantages with it for the further components. I will not go for the highest model and invest this money for adding eg. Wyred 4 Sound reclocker just in front of a DAC or USB-SPDIF converter. With Paul's card you have digital audiooutput separated from the computer USB bus. You can power it with dedicated PSU and this current may be used by the DAC or other reciver if it is needed DC for comunication (it dose not give you isolation). And you are sure that the signal going out of the computer is reclocked with audiofile grade clock with stable powering and not just "something"

 

[Mogos]

Olny one thing is in minus in the whole casing design considering it from the viewpoint of cooling. The cage over the tubes is not recomended. But looks sooooo gooood for me that I am willing to accept the consequencec of using it. Tubes have not an easy life conditions and the whole amp could run even cooler.
I think that for hot days I will take it off.

 

[Maxx134]

My room is always very warm 29 celsius , 84 Fahrenheit,
so I do not have advantage of effective cooling,
But I like your automatic and temp sensor solution.

Do you have specifics on these components?

I rather not look myself, if you already done it..

 

[SonicTrance]

  In my amp after power is on the ventilators doesn't work. The amp is heating up for approximately 40 min (winter/spring conditions) and when temperature reaches 50 degrees the ventilators turns on cooling down the temperature to 40 degrees.
When I was planing the modification one of the priorities was to have good ventilation. Just to remind some photos of the casing done by me.

You really did a beautiful job with your amp Mogos!
If I ever decide to extend my amp I would (try to) hire you to make the extension! 

 
 

My room is always very warm 29 celsius , 84 Fahrenheit,
so I do not have advantage of effective cooling,
But I like your automatic and temp sensor solution.

Do you have specifics on these components?

I rather not look myself, if you already done it..

29C, really? I would die!! We try to keep a steady 20C at home.

 

[baronbeehive]

******



Sorry if I came off rude.
But the Ideas presented to me can only apply to coinmaster unfinished amp which can change in design however he wants.

It would have been nice to have an actual mod to implement instead of nice ideas.

But "we" (me and my dog) are coming up with actual solutions. .

. .

Agreed! And I thought it would be best to stick to realistic solutions rather than theoretical at least at this stage, hence the need for more info on what the proposed mods actually are. I'm not at the same technical level as the others anyway so wouldn't be able to implement the more exotic ideas.

 

[baronbeehive]

Quote: 

................ As I have alredy written I have placed the sensors in the hottest points so I have good control of the temperature inside the amp 

. And this is what it shall be.
 
...............

 
Sorry Mogos, which is the hottest part?

 

[Mogos]

 

Measuring points of temperature.
 

 

[Mogos]

Maxx the temperature controler I use is ESCO company DC-20. Powered with 12V (there are options 230v and 24V). And has two measuring probes. 
 
Below Google translation:
 
2 CONTROLLERS in one housing
2 measuring inputs
2 control outputs
View T1 and T2 alternately
Mathematical functions (measurement of average, differential)
REGULATORS can be associated with each other (eg. WYJ1 = on when WYJ2 = disabled)
Digital input for manual or cat. failure
Built-in beeper (buzzer)
One controller for adjusting the temperature of heating systems:
- Electric heaters, stoves,
- Heating cables, deicing systems
- The incubators, terrariums, aquariums
- Boilers, heaters, etc.
- Pump C.O. and hot tap water
- solar panels
- Fireplaces with water jacket
- Servo valve 2, 3, 4 way valve
and cooling:
- Refrigerators, freezers, refrigerated counters
- Fans, coolers
- Units for automobiles (version for 12V or 24V), etc.
DC-20 stands. Abbreviation "double controller" is a de facto two temperature controllers placed in one housing with the possibility of any configuration of input-output and linking both regulators. The device has 2 inputs and 2 control outputs, which can operate in either heating or cooling, and the measurement is measured from the sensor T1 or T2, or both sensors simultaneously.
With mathematical functions measurement with two sensors can be calculated as the difference [T1-T2] (used in systems with solar collector) or average [T1 + T2] / 2.
The controller can perform both simple single or double heating system or cooling and more extensive application.
The controller has a digital readout of the temperature and calibration function display. The value of the sensor T1 and T2 is displayed alternately every 3 seconds. You can be set to control display only one reading.
Built-in beeper can be set to sound when switching outputs and alarm states.
The digital input can be used as an alarm input or manual output control. It can signal a failure of the activation of an external pressure switch or temperature limiter STB connected to a digital input. In case of failure disconnect the whole system with the ability to sustain the fault state until manual reset button.
The controller is equipped with 1 relay output high power 20A (4500W maximum resistive load) and Output 2 with max. Load 8A. Mounting in a panel or enclosure hermetic, wall-mounted.
 

Zdjęcia

Do góry​

 

Entrance:
2 NTC temperature sensors 5kOhm at 25 ° C
digital input (normally open or closed)
Measurement range:
-50 ... + 150 ° C
Accuracy:
+/- 0.5 ° C
Sampling period:
330 ms
Display resolution:
0.1 ° C over the range of
Resolution settings:
0.1 ° C over the range of
display:
LED, 4 digits with a height of 11mm with graphical icons
Control method:
ON-OFF with hysteresis
Accuracy of adjustment: adjusted in the range 0.1 ... 20.0 ° C
Control outputs:
1 relay 20A 4,5 kW, durability of 100 thousand. cycles
1 relay 8A 1.5kW, durability of 100 thousand. cycles
Additional features:
- Mathematical functions (differential measurement, average)
- Channel selection temperature read
- The audible alarm signaling
- The possibility of connection of both outputs (application conditions)
Montage:
in the hole dimensions: 71 x 29mm
Degree and class of protection:
IP65 / II
Power supply:
230 VAC +/- 15% or 12VAC / DC or 24AC / DC
Power consumption:
max 3 VA
Working conditions:
-5 ... 60 ° C; 0 ... 85% RH (non-condensing)
Storage conditions:
-40 ... 85 ° C; 0 ... 85% RH (non-condensing)

 

 
I think you can find somthing similar localy.
 
Below how it was fitted in the amp casing.
 

 

 

[Maxx134]

My room is always very warm 29 celsius , 84 Fahrenheit,

so I do not have advantage of effective cooling,

But I like your automatic and temp sensor solution.


Do you have specifics on these components?


I rather not look myself, if you already done it..

29C, really? I would die!! We try to keep a steady 20C at home.

Well  actually right now it says 28.1c





That's at night with amp off so coldest it gets..

 

[coinmaster]

It turns out you were right about the "common cathode connection" While technically you can do it, the current source will be fighting the antiphase signals of either tube as well as the different tolerances of the tubes and you will not end up with constant current at the cathodes.
 
On the other hand it doesn't matter because the current sources will make sure they are both running at the same current anyway and maintain balance. Even with different tolerances the balance will be the same since both current sources will be individually ensuring that the tubes operate the same.

As for the force balance mod you would have to splice apart or rip out the copper trace between the cathodes in the input stage for it to work but it's not hard to reconnect them whenever.

So I was thinking, if you cant keep the "common cathode connection" with the current source because it would be fighting anti-phase currents then what does the situation look like when a single 1.5k resistor is fighting the antiphase currents?
 
This is our stock input stage with the 1.5k tail.

 
 
This is what the voltage swing looks like with this

Almost a sawtooth wave.
The current swing through each tube seems fine
 
Now let's see what happens when we separate the "common cathode connection" independently bias them.
 

 
Let's take another look at the voltage when the two were connected

and now when they are separated

Much better until you realize the voltage swing of the top one is a couple milivolts and the bottom one is hundreds of milivolts.
 
This results in lower current and voltage swing at the cathode and anode because the voltage swing on the separated version is following the input to an extent, meaning there's less cathode-grid voltage differential, meaning the tube is not turning on/off as hard.
 
The reason the "common cathode connection" is maintaining a stable voltage must be due to the constant antiphase currents canceling out changes in voltage.
 
So it seems that the balanced design has a unintended bonus of creating a stable voltage reference.
 
Although you get more gain by having a stable cathode voltage, using a single resistor for each tube generates its own local negative feedback so it could still possibly be better.
 
Another option is a current source for each cathode but that is another can of worms.
 
So if the cathodes have a stable voltage reference via the resistor then another option for that route is a LED biasing, yes as in 

 
 

They are constant voltage devices and are often praised for superior sound to resistors, they are often used in input stages.
 
You'd want to use an infared LED with a forward voltage drop of 1.5v and strap a 4.5k resistor (depending on max current of the LED) between cathode and B+ so it has enough current to maintain linear voltage regulation.
Will this resistor affect tube operation? No, the excess current will flow from ground to B+ and voltage will be maintained.
 
The last option is also highly praised, perhaps more so then the LED option. It is often considered very musical.
It is battery biasing. 
 
Yes as in this
 

Batteries are also constant voltage sources. If you connect the positive terminal to the cathodes and the negative terminal to ground it will maintain a constant voltage and many people like it for its musical sound.
Different battery types sound different, lithium/alkaline/life/etc,etc, all have their own sound. 
There was a contest once in vegas and the winner was some sort of phosphurous airliner backup deep cycle battery or something, in any case you can "roll" them like tubes.
Luckily 1.5v batteries and battery holders are everywhere.
 
Another voltage source is a solar cell, as far as I know it's never been tried but I have a stack of cells I got on the cheap for a rainy day.

 

[Maxx134]

You mean for a sunny day..