• 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.

Big *** '50s Power Supply, now what? 300B SET ideas...

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello, All! The Tubes forum helped launch me into tube madness, and I have built and modded a CCS cathode follower buffer/line stage that I simply adore. I'm ready for something bigger, but this time I won't go it alone. I need your input!
I just 'inherited' a crazy '50s custom-built tube amp. It's so old, it's MONO! Built for a friend back in the day by an RCA engineer who was a friend of D. Hafler's, it has a separate power supply leading to a Heathkit preamp and a custom built 5881 P-P monoblock w/6SN7 driver+splitter tubes. It has sat in an attic since the '60s. Unfortunately, I can't get any sound out of it, and it looks like it's too much work to troubleshoot w/ the 3-chassis config. & multi-wire umbilicals.
I fired it up, first using the light-bulb trick, and then once tha caps had formed a bit, with full power briefly just to check voltages on the power supply. A healthy 495V coming off the CLCLC stage. Something like 585V before it? 545V?

I want to rebuild the power supply to feed a stereo 300B SET amp. There, I said it!

We're looking at the following parts: a HUGE power transfo, "GE Grade1 ClassA Family03" 800Vct 300mA with three heater sec. 5.1V@4.0A, 6.4V@6.25A, 6.4V@0.3A and this thing is potted and solid, big too at 5"x4"x7" inches high!
RCA 5U4G rectifier feeds two rather large chokes and a bank of four "cigarette pack" oil caps, 600V 4MFD, Cornell-Dubilier and Aerovox. I was disappointed in the low C values given the size of these, but they're not leaking and I will still use them. I do wonder how these would sound as a coupling cap!
There's another transformer on the chassis, a 25V sec. "filament transformer" it looks like he stepped it down thru yet another trans for the 12V tubes in the preamp? I plan to remove these in favor of dual 5V filament trafos for the 300Bs. the 5V sec. on the power trans can run the rectifier. I dunno what good that 6.4V@0.3A sec. is? Too weak to run more than one small tube? It is left idle on this amp, anyway.

Getting down to the design questions I have: I am strongly leaning towards a 300B "Monkey" or DRD-style amp. If I go this route, don't I want higher B+? Welborne uses 550V, J. Elliano uses 700V, Thorsten's (diyaudio ID:Kuei Yang Wang) design uses 620V. How could I get that 600V from this transfo? A different rectifier? Solid-state would get me there. I don't understand Thorsten's hybrid rectification very well... How about dual tube rectifiers? What are the benefits of this approach (lower supply impedance?), and can it raise B+ higher than one tube?

I know there are a lot of great NOS tubes out there, but I want to drive the 300Bs with something widely available, in a single gain stage. I looked at the JJ ECC99, but don't understand how this tube is designed to drive something like a 300B? It seems to be kinda low mu, high Rp? Probably useless in a DRD circuit. I have read that a triode-strapped pentode such as an EL84 is a great way to go. How about the 6V6? Can the EL84 as triode stage swing the 300B grids properly if I want to use the ~2Vrms output of my tube buffer as driven by CD output?

Basically, looking at Thorsten's Monkey design schematic, I would like to adapt it to a "currently available" driver tube, like the EL84: here. I tried to find tube info/charts on the d3a, but couldn't, to compare it to an EL84.

Then, the biggest challenges for me remain choosing the appropriate plate chokes and redesigning my power supply to drive it all. I can't wait to hear your suggestions!!!

Peace.
 
If you tie the CT of the rectifier winding off, hybrid bridge rectify, and choke I/P filter, you will get into the voltage zone you want. About 150 mA. of B+ will be available. Stacked pairs of UF4007s would form the connection to ground. The other side of the bridge would be a 5R4, as its high forward drop is an advantage in this situation. The B+ rail will come in somewhere around 650 V.

The ECC99 is mu 22, LOW Rp, and HIGH gm. If you want more gain along with HIGH gm, use a 5965 with its sections in parallel. The mu of the 5965 is 44.

ECC99 data sheet here. 5965 data sheet here.
 
BTW, the link in my first post goes to the schematic for Thorsten's power amp section. Here is the schematic for the power supply.
Is this the type of hybrid bridge rectifier you mention? In this circuit the output voltage seems to get boosted above AC levels, in my case would be close to 1000V. Hence the recommendation of a high-drop rectifier tube?

I still can't seem to get a handle on how the hybrid bridge works, but I believe that it does so I guess I might go ahead and use it. I feel I would understand it better if the intention was to generate a negative supply. How does tying those diodes to GND increase the voltage on the other side of the tube rectifier? I guess I should just pretend that the tube rec. is just two more diodes, but I still don't get it!

I wonder if it would be worth putting hexfreds in?

The 5687 with sections paralleled sounds like an NOS I would be willing to try. This tube seems to get pretty nice remarks all 'round, and it's not too hard to find.
 
The 5687 is very similar to the ECC99, heater draw is higher and mu is lower. For a 2 stage amp directly driven by a CDP, the 5965, which is inexpensive and GOOD sounding, is the way to go.

Ignoring losses, which are inevitable regardless of rectifier technology and filter topology, if a capacitor I/P filter is used, the DC voltage will be approx. 1.4X the AC RMS value. OTOH, if a choke I/P filter is used, the DC voltage will be approx. 0.9X the AC RMS value.

To form a hybrid bridge, each of the anodes of the vacuum rectifier gets connected to a separate end of the rectifier winding, in this case 800 V. The B+ is taken from the cathode, as is usual. A path to ground must be created. That's where the "sand" comes in. The free cathode of a stacked pair of UF4007s is connected to an end of the rectifier winding and the free anode is grounded. A second stacked pair of UF4007s connects the other end of the rectifier winding to ground.

In a hybrid bridge rectifier, the vacuum diodes dominate and block the creation of SS diode switching noise. So, FREDs are a waste of money.
 
Remember the KISS principle. For your first amp, why not use what you have got- build yourself a single ended stereo amp, using triode strapped 5881 with 6SN7 for driver tubes. This can sound really good with a pair of good SE output transformers

To start with I strongly recommend going to the download page at duncanamps.com and grabbing the power supply simulator and the tube database software.

800VCT should easily get you a decent 400 odd volts using a bog-standard rectifier tube, easily enough for your purposes. No need for exotic bridges to start with, IMHO.

Also I would strongly recommend using new power supply capacitors, there is a possibility that the can caps may contain PCBs that are not that good for you if they fail.

By the way, please read the safety thread if you are a beginner- 500V can kill you stone dead if you let it!
 
Sounds like a classic Williamson amp. I bet the output transformer is pretty nice. Probably used KT66s or 6L6GCs originally, as that high plate voltage is wayyy beyond the rated capabilities of a 5881.

Be aware that your measured voltages are going to be higher than 800VCT since it was probably rated for 115VAC line. You are probably getting more like 870VCT, hence the 495V measured B+ voltage. Normally that would be closer to 450V... well, if it were still the 1950's. Also take into consideration that your 300mA current draw rating applies for 115VAC/800VCT, and derate it accordingly if you plan to run it at full 125VAC line voltage. Still, that's a really big transformer!

BTW, saw your speakers in NJ ... very neat!
 
SHiFTY,
I like your idea for a single-ended 5881 amp. As it stands, I have already ordered up some 300Bs to play with, but starting out with a more traditional circuit and using a lot of the parts I already have may indeed be the best way to get up and running.
How does this design fit the bill? Schematic is at the bottom of the page: SE 5881 amp
This design is definitely running this tube flat out, not very conservative settings!
BTW, thanks for the link to the duncanamps site... I haven't been there in a while and I forgot what the name was.

Jon,
First of all, thanks for the comment on my speaker designs. Yeah, I think this old amp may well be a Williamson design; I haven't looked at it closely enough to say for sure. The output transformer is an Acrosound, but I am not impressed - it seems lightweight and possibly ring-y.
You know, you were right on guessing the VAC of the power trans with modern wall voltages here in the U.S. I measured it at exactly 870VAC!

Does anyone know if a typical 300B-type SE OPT would work well with the 5881 or 6L6GC, or are the optimal primary impedance requirements incompatible? I would love to but output transformers that I could later switch to 300B duty, if I end up making the 5881 amp.
 
That linked schematic looks perfect- you might want to test your 5881s if they look a bit old and tired, but generally these things last a long, long time. Maybe run them a bit more conservatively- increase the cathode resistors a bit or use a lower B+ voltage.

If you want to see what voltage your power supply is giving, rig up the transformer to the 5U4 to C-L-C and then a dummy load of power resistors. From that circuit you can work out roughly what current you will be drawing, and measure the B+ voltage. It will drop quite a lot under load.

With regards to an output transformer, most have taps on the secondary, so you can use them for different loads.

For example a 5K:4/8/16ohm would be a pretty standard choice. If you put an 8 ohm speaker on the 16ohm tap, it effectively turns into a 2.5K:8ohm transformer.
 
Using 300B to do a retifier tube each have 125 mA

This is my suggest!

Using 300B to do a retifier tube each have 125 mA

My amplifier is using 6 pcs 300B parallell (750mA) at 485 Volt after the choke coil

(R+L channel) total 12pcs


An externally hosted image should be here but it was not working when we last tested it.
 

Attachments

  • cimg4593_1.jpg
    cimg4593_1.jpg
    36.6 KB · Views: 228
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.