845 PP

[artosalo]

I am asked to do a design of 845 PP-amplifier for a DIY hi-fi enthusiast. My plan is to keep things simple and straightforward and therefore use Williamson type front end and fixed bias output stage. I have already done some simulations and all seem to be OK. The only thing I wonder is the supply of the filaments of 845's. I have a bit doubts about the AC-heating. Could the hum be nulled to hifi-levels (-70...-80 dB) ? The other and even cheaper alternative is to use cheap AC/DC power supply (or two) with proper inrush current limiting NTC. Like these: AC 110V-220V TO DC 5V 12V 24V Switch Power Supply Driver Adapter LED Strip Light | eBay

Anybody having some experience/opinion about these ?

Tentative design:

 

[kodabmx]

Nice design!
You might need a small cap in parallel with the feedback resistor depending on how 10kHz square looks on the scope

In the Williamson amps I've built (basically copied directly from the Acrosound catalog), I make R11 470, and put 220 in each cathode leg of the driver tube for current sharing. Dunno if it's required, but it works

As far as DC SMPS, I have had great success with some supplies, and blown up others.
I would use 12V and drop what you don't need in a resistor to make 10V for the tubes (The amp is already wasting lots of power, what's a little more?) and use at least a 12V30A supply for it (unless you want to use a cooling fan...)

Try this: AC 110V-220V TO DC 5V 12V 24V 2A 10A 15A 20A 40A 60A Switch Power Supply Adapter | eBay

It's a piece of crap, but it should work for you. Also, it has a voltage adjustment that gets close to 10V, and as high as 13.5V or so.

 

[kodabmx]

This supply WILL work for you, but it only works at 220V unless you modify it for 110V (even though it says it works at 110V on the label, it doesn't). I got two through eBay and found this using google image search.
Купить Блок питания компактный (узкий), 250 W, 12V в интернет-магазине SWGShop

I would use one per channel so they don't get as hot.
EDIT: Seems I was right, reading the datasheet says to derate to 40% of rated power if running at 50c with no fan

 

[artosalo]

You might need a small cap in parallel with the feedback resistor depending on how 10kHz square looks on the scope

Sure. It is done when the prototype has been built.

As far as DC SMPS...

I also would have one/channel.

But I am interested to have opinions about the AC-filaments and how low the hum can be nulled.
I plan to use some 12 dB of GNFB, so this adds some (hum) attenuation.

 

[kodabmx]

Found this thread: 845 AC filaments, my experience

George from Tubelab: "I tried AC heating and found that the hum could be reduced to the 3 or 4 mV level on the 845. The 211 was worse at about 6 mV. The issue is not the hum, it is IMD. The 60 Hz (50 Hz) will intermodulate with the audio signal creating mixing products. Run a 1 KHz tone through the amp, and you will get the 1 K tone out....WITH some 1060 Hz and some 940 Hz (1050 and 950 in the UK). You will also get some hum at 60 Hz AND its harmonics, 120 Hz, 180 Hz..."

 

[Koonw]

In this article, it tells how to apply hum cancellation technique to driver and output stage in addition to hum balance. Btw, when you use DC method, you no longer have the hum balance pot, so is current balance, the tube current going to one side of the heater and how to deal with this? Or you keep the hum pot even when using DC?

 

[artosalo]

845 AC filaments, my experience

Thanks for the link. I tried to search forum but found nothing. George found IMD products that is obvious. I am wondering did he has any GNFB and how much that may improve the performance with AC-heating. Need to test this.

 

[cerrem]

Using DC SMPS is actually a good idea and it works well....I have designed and used SMPS for a number of years.... On the other hand the SMPS in your link are cheap low-end models that are not reliable an very noisy..

 

[Hearinspace]

IIRC , in his earlier DHT filament regulator posts, Rod Coleman talked a little about what would be needed in a circuit for push pull. I wonder if he ever got around to making boards for them.

Twenty plus years ago when there a lot more debate about the merits of ac vs. dc (and between voltage regs and current regs) there was anecdotal evidence to suggest that some preferred ac heating because of a certain low level of hum intermodulation, resulting in a "lively" sound. I guess these days most would scoff at the thought, but when it was pointed out to me back then it did seem to tally with my experience listening to lots of different circuits.

 

[kodabmx]

I suppose if I was to create an amplifier with directly heated tubes, I'd use AC. If I wanted an amp that had no hum etc I'd use indirectly heated tubes or solid state parts, right? Besides, the added complexity of DC'ing several tubes separately annoys me.

 

[multi]

845 circuit

Rather complicated Japanese Circuit for 845 uses AC. Can you hear the Modulation of the music by AC? I have made many amps using filament triodes using only DC with big caps 30 thousand UF and use 100+ db speakers.

 

[45]

Artosalo use DC heaters. With GNFB a simple SS bridge rectifier + capacitor + a small resistor (to limit the initial current) will already make a difference. Try it and see by yourself.

 

[artosalo]

I have got a lot of information from the given links. This is becoming so interesting that I will surely test both AC- and DC-heating. Especially I want to see if GNFB has any effect to hum + signal IMD. This is the issue that was not considered in any of the articles I have seen yet.

 

[45]

GNFB surely has an impact however with DC heaters is better. I know use DC heaters also with indirect types.
If properly sorted out you won't hear any hum even with the ear attached to the speaker and IMD will be minimised.