• 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.

20W cathode follower amplifier

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
ampli 20w cathode follwer

Hi
I'd like to do a vacum valve ampli about 15 to 20w.
I like single end for good sound and cathode follower for performance on DHT and Zin out.
I think it 'll be easyer to do using valves like 6c33 or el 519
I knowsthat it 's possible to use boostraping to help driver to do his hard job : 200 to 250 vott cc
What do you think about this projet.
Thanks
Yan
 
Bruce Rozenblitt has a SE cathode follower design in his "Audio Reality" book. This design uses 3 X KT88 in parallel to generate about 20 watts. I have not built one, so I have no experience with the design.

I have been experimenting with cathode follower output stages. I fine that you need a tube that has a relatively high GM and can handle high peak currents. I do not have any EL519 or 6C33 to test, so I can't comment on those, but I have tested a bunch of different tubes, and come to similar conclusions. I have not made the triode - pentode decision yet, but I have decided that the triode would be a 6336A, or a pair of 5998's. Both are similar to the 6C33. My pentode choice would be a 6LW6, not to far from the EL519.

For the driver circuit, I am using a two stage 6V6 SE amp. I replaced the output transformer with a choke, and used capacitor coupling into the output tube grid (a parafeed connection). My choke is too small (kills the bass), so I plan to switch to a CCS load when I have more time to experiment.

The optimum load impedance looks like it is in the 500 ohm to 1K ohm range, but it depends a lot on factors that I haven't nailed down yet, like the choice of tube, B+ voltage, and tube current.
 
hi
Without schema : it's difficulte to discribe .
1 er stage : a usual commun cathode or SRPP with 12at7 or 6dj8 and + - 300v
2 eme stage a over commun cathode in direct link with 12b4 and + - 800 volt
3 eme in C F 2 EL519 tiode connected and OPT about - + 400 Ohm

There is a over way to drive the output stage with boostrap : a self induction is in beetwin +U and plate's charge resistor of driver.

In this way you can choice all power valves and using high charge impedance and use KT 88 KT90

I'd like to know if this way gives good sound

Thanks

Yan24
 
My first tube build was a 6c33c cathode follower. I tried bootstrapping, but had a problem with turn on thump from when the drivers stage cap charged. Though I'm sure it would work if done under different conditions. Bootstrapping is a form of positive feedback, which will raise the output impedance and add distortion, which is what you are trying to get away from when doing the cathode follower approch. However, if the driver stage is using an active load(CCS) it seems, you may not encounter the negative effects of bootstrapping. Since the CCS would in a sense isolate the outputs audio from the driver tube. Eventhough I now perfer passive loads to active and would suggest just going with a larger voltage transformer for the driver stage.

The hardest part about building a 6c33c SET(at least in the US) is sourcing transformers. Both output and power trannies. The outcome was rewarding though, the amp sounds very clean and detailed. Eventhough there are a couple things I'd do different if I were to do it again. It benches at .1xx% THD up till it starts soft clipping. Oh, forgot to mention the tight bass and that it can handle 3way speakers.

If you decide to go that route I can help you out.
 
yan24 said:
hi
Without schema : it's difficulte to discribe .
1 er stage : a usual commun cathode or SRPP with 12at7 or 6dj8 and + - 300v
2 eme stage a over commun cathode in direct link with 12b4 and + - 800 volt
3 eme in C F 2 EL519 tiode connected and OPT about - + 400 Ohm

There is a over way to drive the output stage with boostrap : a self induction is in beetwin +U and plate's charge resistor of driver.

In this way you can choice all power valves and using high charge impedance and use KT 88 KT90

I'd like to know if this way gives good sound

Thanks

Yan24

I have never built an amplifier of this type, but judging what Tubelab has to say on this thread, I would think that 13E1 would be a good valve to use. A regulator beam tetrode of 90W Pa, its gm max. is 35.

Bonne chance

7N7
 
Heres just a stab in the dark, but here goes.
When looking at tubes like the 6080,6C33 and such like I would guess that a common cathode with a normal output transformer will produce almost all of the benefits of the cathode follower, but with extra power and easier design. I say this from some personal experience with low output impedance valves and the resultant low ratio transformers.
The basis for my assertion been that the low output impedance gives maximum bite on the transformer inductance, and the low ratio minimises stray nasties which tend to bleed off the high frequency response.
My conclusion, for what its worth, is why bother with a cathode follower (with its degenerative feedback) when a common cathode will probably sound as good.

You may disagree.

Shoog
 
I do agree. There are several disadvantages to doing a cathode follower output stage, including the large drive voltage requirements. I don't think that I would build one for normal amplifier service (except maybe an SE OTL).

I need a BIG cathode follower for one piece in a much larger design. It is a big experiment far into uncharted teritory. Success or failure it will all be written up in November.

The 13E1 looks like an ideal tube, but they are extremely rare in the US. I have run dozens of tubes in cathode follower service under my unique conditions. The best were large sweep tubes, and the higher Gm regulator triodes, 6336A, 5998A, and 7236. The usual 6AS7 - 6080 is OK but not as good as the previously mentioned tubes. For smaller designs some of the dual dissimilar triodes made for vertical amplifier service (6EM7, 6DN7, 6DE7, 6DR7) work OK.

I tested the 6HS5 and its brothers including the 6HV5. They need too much B+ for my particular needs, but I think these guys are good for the ultimate "spud" amp. All you need is the tube, the OPT and a power supply, further details later.
 
These tubes came at the very end of the vacuum tube TV era. They were an attempt to eliminate the 6BK4 shunt regulator which worked directly on the 25KV supply. Think about what a shunt regulator across 25KV goes through. They had a short life, and generated some pesky X rays, so after 15 or so years they were designed out. Then tubes were designed out of TV sets.

GE (and others) moved the shunt regulator onto the primary side of the flyback transformer, so in only has to deal with 2500 volts. These tubes were made for only a few years, and I don't know if the idea was ever adopted outside the US. Most of the US tube sellers have some in stock for fair prices. There are a few different type numbers for these, and they come in two dissipation sizes. The 6HS5 is rated at 30 watts, and the 6HV5A is rated at 35 watts. I have seen 30 watt 6HV5's and 35 watt 6HS5's with and without the "A" suffix.

There was a thread devoted to these tubes about a year ago. I never got to making any sound with them, but that is about to change.

http://www.diyaudio.com/forums/showthread.php?s=&threadid=78041&perpage=25&highlight=&pagenumber=1

The 6HV5 data sheet has curves, but they don't cover the "nornal" range of operation.

http://www.mif.pg.gda.pl/homepages/frank/sheets/123/6/6HV5A.pdf
 
tubelab.com said:
I never got to making any sound with them, but that is about to change.[/url]

Rats, that's the signal to stock up. I have made sound with them, CF driving a transmitter tube. The cathode load was a 150H choke for max impedance in series with a resistor to set the operating point. It worked very well.
 
6hv5a

I have just had a look at the GE data [URL]http://frank.pocnet.net/sheets/123/6/6HV5A.pdf]http://frank.pocnet.net/sheets/123/6/6HV5A.pdf[/URL] (why doesn't this work?) on Frank's site for this very interesting valve.

The data sheets are dated 10/69, so here is an good indication of the way development as going right at the end something that has always interested me. Apart from the very high gm of 65 mA/V (only equalled as far as I know by WE 416) extraordinary for a "consumer" product unlike 416, I found the description of the diffusion coated cathode fascinating.

As for using it in am amplifier, well the high anode resistance really rules it out for a conventional output stage, although for SE I suppose one might get away with a 10K transformer and as pointed out on the other thread referred to here, Miller capacitance would be a problem.

However, for a cathode follower output stage I would imagine it would be superb. I would love to build a p-p amplifier with output transformer using this type. Agreed that the driver stage would be challenging, but not impossible. Of course one would have to be careful about heater cathode voltages, but 6HV5A does offer a decent range.

7N7
 
What do you think about this disign but i think it'll be difficult to find a good intersage transfo

I am sure that this design would work, and sound good with good transformers. As you point out, a good IT transformer is hard to find. The specified Lundahl should be available. I will keep experimenting until I find a driver circuit that I like that can deliver 400 to 500 volts P-P. I will post it when I find it. So far the choke loaded 6V6 works good, but doesn't suit my needs.

I never got to making any sound with them, but that is about to change.

OK, last night I made sound with a 6SH5. My ultimate "spud" amp (a single tube amp) needed one resistor to work right. I set up a 6HS5 in common cathode using clip leads. No bias, driven directly from the headphone jack of a portable CD player, 5K ohm Edcor OPT, variable B+ supply. I had to add a 1K ohm resistor from grid to cathode (parallel with the CDP output) to provide a path for grid current. I had to switch to my 550 volt power supply and max it out to get enough plate current (30 mA at 550 volts). Power was about 2 watts limited by lack of drive.

I don't think that the sound would cause anyone to run right out and buy up all the existing tubes, but it was better than I expected. The bass was kind of "loose" due to the highish Rp. Highs were crystal clear, so Mr. Miller was stopped dead by the low source impedance. Would I build one of these? No, a chip amp would blow it away, and it needs a lot of voltage.

I may try these in cathode follower mode, but they would need a lot of B+, which would cause over dissipation, or they would need to run with a few volts of positive grid bias, where they may not be linear. Only time will tell.
 
Today, I had some time to experiment with a cathode follower amp. I built a very simple driver that could deliver 490 volts peak to peak from a 550 volt supply. Then I started thinking (very dangerous). To make any serious power from a cathode follower, I'm going to need a lot of B+ voltage. The most my bench supply can give is 550 volts. I need an OPT, but the lowest impedance transformer that I have is 3K ohms, and it is rated for 110 mA max. If I want to be able to test at full current (and maybe more) I am going to need a tube that can dissipate about 50 watts. With the thinking part done, I went for the biggest tube in my junk box, the 6LW6. These will laugh off 50 watts.

So, I wire one up on my breadboard, and crank it up.
 

Attachments

  • amp_1_b.jpg
    amp_1_b.jpg
    94.1 KB · Views: 4,551
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.