• 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.

FU50 + 6N8P + 5Z3P Amplifier

p { margin-bottom: 0.25cm; line-height: 120%; }

FU50 + 6N8P + 5Z3P Amplifier


I built my first 300B SE amplifier nearly thirty years ago. Back then SE OPTs were very difficult to find so I contacted Sowters and asked them to build me two. At first they refused, saying that SE OPTs were of antique interest only and too technically compromised for a modern hifi application. After an exchange of several letters they finally agreed to consult with their founder and technical advisor Dr G A V Sowter (then retired and in his nineties) for a design which he duly produced. From today’s perspective the OPTs and the 300B tubes I sourced were very cheap, although at the time I thought them very expensive. I bought a quantity of WE300B tubes for the steep price of $15 each. These were from the UK factory and branded STC4300A.


Some years ago I thought I would build another 300B amplifier and began searching the net for the iron and tubes. Severe price sticker shock followed. There is a very strong interest in diy hifi in China and following Chinese sources for OPTs I found the FU 50 and an amplifier kit. The price of the FU50 kit with all its iron ($130) made it much, much cheaper than buying parts separately from UK and US sources. So I took a punt and ordered the kit. It came as parts and a circuit schematic. I wired it up following the classic tube amp wiring rules. Initially there was too much hum. I had used a star earth system but for convenience I had brought the two sides together using the holding screws on each side of the rectifier tube base and about an inch apart. Bringing the two sides together on the same screw solved most of the hum problem. There was some residual hum which I thought came from the transformers. Tightening the bolts made little difference but, over about 6 weeks of operation, this hum too disappeared.


I have used this amp for some years now and I am very happy with it. I rate the FU50 as better than the 300B.


So, how does the amp sound? Like all the best amps I have ever listened to, built and/or owned, it recedes into the background. It has no character in the same way that the Linley Hood 10w and the Sharp DeltaSigma amp (SM-SX100) have no defined character. One simply listens to the music – the amp does not “impress” the listener at all. If there is any character then it is simply a slight overall grey tonality, when compared to the real thing live. All the other amps I have listened to have had a bit too much "character".


I use the FU50 amp with a Sharp DX-SX1 sacd player and a Kenwood KT990D tuner. For speakers I use a pair of probably 50 year old Australian or New Zealand made Rola 12” AlNiCo speakers in a cabinet made by AWA for broadcasting monitor purposes (AWA Studio Monitor Mk3). For the top end there are some Fostex (Foster) horns and a capacitor. This gets me close to the sound of the Altec Lansing model 19.


This FU 50 amp is still available and now can be purchased as either a diy kit or built up.


Vailala
 
I know little of Russian tubes because I never use them. Please educate me as to why it is a waste to use it as SE A1. Please be specific and precise. I only know that it is a power pentode.

It is a power pentode designed by Telefunken for mobile amplifiers, including RF amplifiers. Filament power is low, G2 is quite dense, G3 is too (for modulation), but G1 is quite far from cathode, that's why it wants high B+, low G2 voltage, G3 is better to bias by about +15V, and use it in A2 mode.

If you find LS-50 datasheet with curves you will definitely understand what I mean.
 
FU50 as triode

p { margin-bottom: 0.25cm; line-height: 120%; }a:link { } Pentode curves are irrelevant.



The advertisements for this amplifier used to make it clear that the FU50 (GU50, ГУ-50, Telefunken LS50) is being operated in triode mode. I see now that they no longer always do so.


The FU50 has a following among diy-ers in China as a triode-strapped output power valve in both SE and PP modes. It is often referred to as the “little 300B” because of its relatively short stature when side by side with a 300B. It is popular because when triode strapped the curves resemble that of the 300B and because it is widely available at low cost as NOS ex-military ($2 -$7 per tube). Much less than the baroque prices now commanded by new production 300Bs. An additional attraction is that the cathode is separate from the heater.


It is my understanding that the triode strapped FU50 (GU50, ГУ-50, Telefunken LS50) as SE and PP also has a following amongst Russian and German audiophiles, especially for use with LOMO Kinap, Tesla and BEAG speakers.


If you are interested in following recent Chinese diy FU50 circuit iterations you may want to look at this for a 6C16 (5842, PTT216, WE417A) + FU50 choke coupled instance.


6C16+FU50ÒôÉ«¶ÀÌصĸ߱£Õ浨»ú¹¦·Åµç·--6C16+FU50 POWER AMPLIFIER


Vailala
 
I see, you can't read... You just repost ads... ;-)

For those who can read, here are curves for comparison. Do you see anything in common? I don't.

300bcrv.gif
 

Attachments

  • gu-50_penintriod.gif
    gu-50_penintriod.gif
    20.3 KB · Views: 545
I see, you can't read... You just repost ads... ;-)

For those who can read, here are curves for comparison. Do you see anything in common? I don't.

300bcrv.gif

Ha!, Wavebourn 100% right

As hard a 300B is to drive, the FU50 is unusable in a tube design.

As to why fu50 is built... same thing as usual : people like the look so it sells.

I wouldn't even use that :snowman: in push-pull if it was the last tube I had. I would leave tubes in their box and go transistors.
 
Ha!, Wavebourn 100% right

As hard a 300B is to drive, the FU50 is unusable in a tube design.

As to why fu50 is built... same thing as usual : people like the look so it sells.

I wouldn't even use that :snowman: in push-pull if it was the last tube I had. I would leave tubes in their box and go transistors.

Absolutely not.

FU-50, GU-50, FL-152, SRS-552, and their original version LS-50 are excellent tubes for audio, but not in a straight A1 triode mode, and no way they are replacements to 300B. Don't compare apples with oranges, they both are great, but not interchangeable.
 
Member
Joined 2002
Paid Member
Just stumbeled on this old thread, after notizing a few FU50 + 6N8P + 5Z3P Amplifiers/kits epay & AliX.
Is consensus still that GU50/FU50 is "useless" ?
Considering the price of a humble Shuguang 300B, even they are getting more expensive than a cheap/ok output-transformer...
 
Last edited:
I collected a few Russian GU50 tubes a while ago, but haven't gotten around to building with them yet.
I intended to use them for push-pull triode, class A. The idea was to make a PP triode amp that could make 15W per channel.
Then I started doing more research, drew some loadlines, etc.
The GU50-Triode is not anything close to a real 300B. But how can you say it's 'useless'? I think it can be pretty good for not too much $$.

From what I've read, the general consensus is that the best way to use a pair of GU50 is in pentode, using plate-grid local NFB ("Schade"). That should be done with an LTP driver stage made of high-gm pentodes, something like 6EJ7/EF184. I'm pretty sure Wavebourn used a 10k:VC OPT for his GU50 amplifiers, with a very high B+ and regulated Vg2 supply down around +200V.

As a triode, GU50 max plate-to-cathode voltage is going to be around +400V. To get close to class A operation the load impedance would need to be high, perhaps 10k:VC, or else you'll be running in class AB operation for push-pull. A high power 10k OPT capable for use with output tubes drawing 80mA each of plate current is a tall order. If you go back to a lower impedance OPT but stay up at Vp = 400V you'll be in class AB operation. THD would then be kind of high, so you'd need some NFB, at which point you'd have to ask why you're not using the GU50s in pentode in the first place.

Perhaps if you reduce your power output requirements to only 10Wpc, PP GU50-Triode can work with good damping, low THD, all without global NFB, with more easily obtainable high power OPTs rated 5k:VC or 6.6k:VC.

Yes, you'd be burning 50W in plate dissipation plus 20W of heater current to make 10W of audio output power. Damnable triodes...

I have a wreck of an old Dyna ST70, from which I was thinking of taking its OPTs and trying something like PP GU50-Triode.

Loadline
Rload = 2.5k
Vp = 350V
Ip = 70mA

GU-50_triode_2500ohm-LOADLINE.jpg
 
FU50 as triode

p { margin-bottom: 0.25cm; line-height: 120%; }a:link { } Pentode curves are irrelevant.



The advertisements for this amplifier used to make it clear that the FU50 (GU50, ГУ-50, Telefunken LS50) is being operated in triode mode. I see now that they no longer always do so.


The FU50 has a following among diy-ers in China as a triode-strapped output power valve in both SE and PP modes. It is often referred to as the “little 300B” because of its relatively short stature when side by side with a 300B. It is popular because when triode strapped the curves resemble that of the 300B and because it is widely available at low cost as NOS ex-military ($2 -$7 per tube). Much less than the baroque prices now commanded by new production 300Bs. An additional attraction is that the cathode is separate from the heater.


It is my understanding that the triode strapped FU50 (GU50, ГУ-50, Telefunken LS50) as SE and PP also has a following amongst Russian and German audiophiles, especially for use with LOMO Kinap, Tesla and BEAG speakers.


If you are interested in following recent Chinese diy FU50 circuit iterations you may want to look at this for a 6C16 (5842, PTT216, WE417A) + FU50 choke coupled instance.


6C16+FU50ÒôÉ«¶ÀÌصĸ߱£Õ浨»ú¹¦·Åµç·--6C16+FU50 POWER AMPLIFIER


Vailala
The circuit as drawn dumps 1/7th the aidio power into the 600R in the cathode. Error? Who knows.
 
It looks to me like a simple variation on a Loftin-White circuit, DC-coupled. Yes, extremely wasteful of energy, as that 2.4k ohm 30W resistor is needed to raise the voltage to that of the plate of the 6C16 driver triode. The 600 ohm resistor is the cathode load resistor for the FU50-Triode. Assuming something like Ip = 70mA for the FU50, the 600 ohm resistor would drop 42V, which seems about right for the grid-cathode voltage to bias an FU50. I guess they call for that to be a 20W rated resistor in case of things shorting out. In normal use it should dissipate approx. 42V * 0.07A = 2.94W.

The 2.4k resistor will have the plate currents of both the FU50 and the 6C16 going through it, so I expect about 80mA.

2400R * 0.08A = 192V

192V * 0.08A = 15.36W

The 30W rating for that resistor makes sense. It's not even excessive. That's gotta be one big honkin' resistor!