ECC88 with 12V heaters

[ruffrecords]

Yes, please help to spread the word - people should stop using the Koren and Munro models since the grid currents are poorly modeled (if at all). When you distribute the Ayumi's models, just be sure to include the top three lines of the SPICE code, as it's his copyrighted material.

Yes, I am happy to do that. I have noticed significant differences between most models and actual circuit performance, especially with regard to distortion, so I am always interested in more accurate models. I want to try a few more of his models, especially the ECC88 one.

I have just completed a small PCB that can take either an 6992 or a 5965 in this circuit. I'll post the test results when I have them.

Cheers

Ian

 

[ruffrecords]

I finally got round to building the PCb and testing it with a number of 5965 tubes and also with a couple of 6922 tubes (with different component values). The feedback resistor was set to 430K and the input resistor to 100K which gave a gain with the 5965 of close to 10dB. I tested five 5965 tubes and the average distortion figures into a 2400 load at +20dBu were:

2H -47dB (0.45%)
3H -61dB (0.09%)
4H -85dB

These are average figures. One tube measured -44dB for 2H and another -51dB. Most were close to -47dB. The 3H range was from -59dB to -63dB.

This is rather worse than the simulation which gave 0.6% into the same load at +25dBV (+27dBu). I expect at +25dBV the 5965 2H distortion will exceed 1%.

I then tried the same tests with a 6922. Only the cathode resistor was changed to set the plate current to a similar value. The distortion at +20dBu into 2400 ohms was:

2H -60dB (0.1%)
3H -65dB (0.056%)
4H -83dB

It seems clear that the 5965 is not a very good 12V heater substitute for a 6922. The 6922 has a higher open loop gain so the feedback resistor could be reduced to set the gain back to 10dB with a resultant further decrease in distortion.

Cheers

Ian

 

[kevinkr]

That's a rather substantial improvement, so is there some space on the card for at least a dropping resistor for the 6922 filament?

 

[Funker]

RR,

I'm going to ask a "stupid" question. Is this a situation where each channel has its own bottle? Should that be the case, simply run the heaters of your favorite 6922 in series.

Hi Eli,

eletrically it will work. But how about hum balancing in sensitve circuits like pre-amps? The only choice left is dc heating for valves in those application.
Which may not a problem if you planning to build an amp.


73
Wolfgang

 

[ruffrecords]

That's a rather substantial improvement, so is there some space on the card for at least a dropping resistor for the 6922 filament?

This PCB interfaces via a 32way DIN connector to my backplane PCBs for use in my tube mixer builds. The standard two module backplane PCB has a pair of mating halves for these 32 way connectors and provides both boards with 12VDC. My plan is to make a variant of the twin backplane board that connects the heaters in series so a pair of boards can be used from the same 12VDc supply.

You could just add a dropper resistor. it would need to drop 6.3 volts at 0.3 amps so it would need to be a 21 ohm resistor capable of dissipation 1.9 watts.

Cheers

Ian

 

[kevinkr]

Sounds like a good solution to me.. (Or a 5W 22 ohm power resistor)

 

[ruffrecords]

Hi Eli,

eletrically it will work. But how about hum balancing in sensitve circuits like pre-amps? The only choice left is dc heating for valves in those application.
Which may not a problem if you planning to build an amp.


73
Wolfgang

In this, dc heaters are being used.

Cheers

Ian