• 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.

6S19 Circlotron (OTL)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Adjustable feedback (a two pole switch shunting R13/14) can be useful when using the amp with different speakers. I´ve noticed that some speakers sound better with lower damping factor.

Re. Iq I thought the output power would rise a bit due to bigger "current headroom" with lower Iq. The tubes will probably last longer at 60mA, listening tests might tell if it´s worth it:cool:

PS: For a while I was a bit confused about the relatively high gain, but with R13/14 in place it makes sense. Without them the output stage gain must be slightly lower than unity.
 
You´re absolutely right about more power with lower Iq. But it is so marginal and not worth mentioning.

Adjustable damping is absolutely a good thing!

Probably the amp will sound better the higher standing current you run through it, at the expense of shorter tube-life. So go for high Iq and better sound as long as they last;).

Also note, I was to lazy to draw gridstoppers for the output tubes and on the MOSFET. Maybe one should also add a small resistor on each anode for current monitoring. Some kind of bias adjustment is also needed. I changed it by varying the neg.-supply.
 
Well, if I was in need of high output power I wouldn´t build an OTL with eight tiny output tubes anyway!:D
I could wish for a little more voltage from the floating power supplies (120VDC would be nice) but I´m stuck with the transformers I have and there is no room to add turns to the secondary windings.
As usual I´m trying to get the best out of whatever parts I can find in my junkboxes or buy cheap used or as surplus.
If this was a no compromise, cost-no-object project I´d use at least twice as many output tubes, custom wound transformers with multiple taps for finetuning the voltages and linear 12VDC power supplies for the heaters.

What I do have is a fairly large stash oh output tubes, so running them a bit hot is definitely not out of the question!

Regarding grid stoppers I´m using 33k on each output tube.
Better safe than sorry! IIRC Bernt J uses 110k in his PL519 OTL.
I´ve also put a 2,2R 3W resistor on each cathode for current sharing and monitoring.

Except for the semi-ruined paint job this project has gone well so far. I´m done wiring up the output tube heaters and all the grid stoppers, cathode resistors and plate connections.
Must order some caps and resistors before I can finish the signal circuitry, in the meantime I got a bunch of power supplies to build!!:xeye:
 
Why 33k, for what reason? The best way to ruin dynamics is by using series resistors. Make them as small as possible with stability in mind.

It is totally unnecessary with higher voltages as current is the weak point. Below 100V will be great! But as you say doubling up the tubes would be the way to go, more current...........

I don´t envy you buliding this multi-tube amp. Looking forward to seeing (and hearing???) it when I pick up the horns.
 
As I said, better safe than sorry. A friend ran some tests on an OTL circuit with 6 x PL36 some time ago and he needed gridstoppers in that range to stops the tubes from oscillating.
Tubes in parallel connection seem to require much higher values than single tubes.

More voltage would allow the tubes to deliver more current into a specific load, right? Tubelab mentioned something in the 400mA range for a single 6C19 with 75V plate-cathode, another member claimed 1,5A at 100-something volts in another thread long ago.
I also remember Bernt (at hififorum.nu) mentioning that one of his OTLs sounded much better after increasing the plate voltage from 150 to 170V (not sure about the numbers though).

However, I´m going to use the transformers I already have and getting new ones would also force me to buy new PUS caps.

I´ve built OTL amps before (one finished SE OTL and two unfinished PP monoblocks) and I can assure everyone that wiring all those tube tube sockets is extremely tedious.
The SE OTL used DC powered series connected P-style heaters which was a breeze to arrange compared to this.
Each bank (with four) output tubes draws 6,3V 4A AC supplied by one 50VA 2x6V toroid transformer per channel.
That means a lot of tightly twisted wire, I´m glad I´m finished with that part.

These amps will not be finished before I move (and hence before the horns must be picked up...) but I´m not exactly lacking power amps to show you;)
 
Don´t think you should compare Berndts heavily NFBd and(almost?)Class-B amps with your Circlotron.

One should check the tube curves(which I haven´t done yet:clown: ) and choose the working poisnt where they are most linear. We must discuss this later!

Just for fun I will do a sim on the 33ks and see if it affects anything.
 
I will start a thread about this project at hififorum.nu too so that we can continue this discussion in our own language:D
(By the way, is it ok if I use the spice schematic there too?)

I´m not sure I want to know how much the 33k grodstoppers affects the HF...:xeye:
 
diyAudio Senior Member
Joined 2002
Hi,

An OTL amp is often best protected by 100K (value is not critial) or thereabout 1/2W gridstoppers.
These high values not only inhibit oscillations at RF frequencies but also isolate the grid from too high inrush currrents where the grid may tend to act as a fuse leaving you with a mighty diode iso of triode after the mishap.

Cheers, :D
 
You can run grid stop resistors that are as little as 600 ohms with no worries. One watt is a good value, as a shorted tube can fry a 1/2W device.

Cathode resistors are another concern. The higher the mu the lower the value can be, but 5W devices are recommended. An arc-over can take out a 3W device easy enough.

6AS7Gs are probably the most economical tube you can use. They hold up well, have good linearity and can handle an surprising amount of grid current, although the schematics shown thus far won't support the grid current operation.

Even in a lower-powered amp its a good idea to keep the B+ up. You can't just go for more current- the tube will saturate. 140V is ideal when dealing with the tubes mentioned so far. The right voltages are often hard to find in off-the-shelf transformers, so bucking the primary with a smaller transformer is one way to get to where you want. Of course the B+ transformer should have enough current reserve to operate easily without saturation at the higher voltage.
 
All 16 33k 0,6W gridstoppers are already soldered in position and I´m not planning to replace them unless I really have to.
33k should be high enough to protect the tubes from both oscillations and grid meltdowns and still low enough not to cause an early HF rollof.

A few more volts on the plates would be nice, buying new transformers and PSU caps would not be that nice...:eek:
 
More voltage will not make things better. It is still lack of current and high Zout that is the problem for an OTL.

Example: 1pr 6AS7 at max 13W with 140V/90mA gives more ClassB power but less than half ClassA power and double Zout. If you multiple the tubes the realtionship will remain.

Quantity or quality?
 
Just kidding, quality wins over quantity every time.
The expected 3-4W is perfectly normal in my world, everything above 8W is considered "high power":D

Still waiting for an interesting thread at hififorum, not so many nowadays!

I agree. I started a thread about my experimental OBs today, but it´s really not that interesting. (Though it might become very interesting if I get those four 15" drivers that Guru mentions in the thread...)
I´m working tonight, if nothing unexpected happens I might have time to start a new thread about circlotron over there.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.