• 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.

PSU capacitance and Toobe Rectifiers

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi All

I've made the Andrea Ciuffoli SE-PP design using the E182CC driver through the Lundahl interstage tranny (1660) and KT88 outputs. Love it. Had some questions regarding the PSU capacitance and rectifier choice.

At the moment, it's running dual mono PSUs with 350V AC taps to the 5AR4s, which then goes to 50uf, then a 5H choke, then 500uf, then 11k2 dropping resistors to 100uf. Are these capacitance and Henry values about right? What's the effect of changing each - I understand the 50uf at the start is about the max suggested. Best types to use?

I do have some 220/220uf/350V BG's and some 100/100uf/500V BGs from a preamp project that never happened. If I put 440uf on the storage, and 200uf after the dropping resistor, would that work OK? Are BGs better in certain positions?

Lastly, I'm using the 5AR4 Sovteks as tube rectifiers. These sound pretty good, but what's the go on rolling others? Recommendations?

Thanks in advance for your help.

Cheers, Kendrick.
The Kendrick?

What kind of wire are you using for point to point wiring? See I gave you an opening, now go for it.....Black Gates, say you are serious at this. Try some soft recovery diodes with snubbers. What are your raw and load DC voltages and your load current? half wave or full wave rectifcation? African or European?


Yes KKKKKendrick!

Hi Harry, you cheeky devil you... :D

I think you just might know what I'm using for internal wiring in this amp, my Silver Stealth wire as used in the interconnect kit-set. Check out Hugh's site for more details and the harmonic discord forum for PE. I haven't yet released a package for people to make up themselves, but it wouldn't be difficult as I've the reels pre spooled onto 4.2m reels, and the tape raring to go. Contact me if you're keen to try it. Believe me, the wire inside the amp is just as critical as the bits you see on the outside. I changed less than a foot of wire in my preamp a few weeks back with a commercial alternative and it sounded completely different. Needless to say I preferred my own alternative!

OK, so at risk of frying my ***, I'll do some measurements on the amp, as it's running (playing some nice acoustic soul guitar by Dexter Moore http://www.dexonyx.com/news.htm ) by the way. Nice stuff.

The HT from the power tranny (also strip wired) is 340V AC. At the first cap, post rectifier, there's 382V DC. Out of the choke it's 367V DC, which is what the 500uf sees. Actually the dropping resistor is 8k8 and after this it's 186V, dead nuts on the requirement per Andrea's circuit. In biasing the KT88s, I've got 29V over 150 ohms on the cathode bypass resistors, keeping the suggested 220uf/35V (using Elna Starget currently, which is the same cap used in part of Hughs Nirvana upgrade). Other bypass caps to try? Let me know. I think this equates to around 96mA per tube and a touch over 35W. Might have to pull it back just a smidge.

I have used the BYW29E and STTA520 as soft recovery diodes in other projects with very good results, but as I'm toobe rectifying for the first time, there's no SS diodes anywhere to be found. BYs with an LM317T make a sweet low voltage tube DC heater supply (did this on the preamp, but the power is using the AC feed for now anyway). By the way, the interstage tranny keeps things QUIET!

I wouldn't mind upping the B+ a bit more to back off the current. Not sure which dictates tube life, plate wattage or current more...any thoughts out there? A CT tranformer around 36V should pep it up a bit more - will try soon.

The output tubes are Sovteks, which I think are excellent value for money.

The input E182CC is a NOS Phillips I found on ebay. Nicer than the JJ ECC99 I thought (more harmonically developed, better bass) - note the pin swaps on the left triode and heater between these tubes!

I was a bit apprehensive about sticking ANOTHER transformer in the signal path, but the Lundahl is impressive. Apart from looking very well made, and being relatively affordable (around $80US ea) the redesign from an Australian Contan circuit showed that the removal of the input board, coupling caps, etc gave me more leading edge and speed and bucket loads more transparency and depth. Yummy.

Andrea's circuit is very simple, and I heartily recommend it for someone looking for a robust, fine sounding toobe power amp in the 20-30W/ch variety. Seems to truly be the best of SE and PP.

Now I've just got to get around to designing mono block chassis machined out of billet, include a funky blue neon under the chassis (aka Hovland) and I'll be all set! Check out the new stock at Jaycar in Australia (www.jaycar.com.au).

Cheers, Kendrick.
I'll try to help out a fellow Aussie and tubophile. Greetings from sunny Sydney.

Power supplies are kind of like a black art. Easy to make something that works, but difficult to make something that sounds fantastic. There's no magic formula with Black Gates that will make any power supply sound good - in fact you have to understand the application before substituting brands or types.

The first cap is to make the filter a capacitor-input type. This makes the output voltage about 1.4x the AC voltage, though this drops a fair bit with the internal resistance of the rectifier and the current drain on the whole circuit. 50uF is the maximum for that particular rectifier - and this cap needs to be GOOD. Do NOT use an under-rated cap here - this cap sees the biggest AC swing out of any caps in your whole amp, so use something with 2x or more AC rating.

I recommend AXON or Solen polypropylenes, and the 630VAC caps should do well.

There are those who believe in a few basic principles in SE tube power supply design:
1. Choke input sounds better than cap input
2. Less capacitance is always better
3. Compensate the above with high value (but still low DCR) chokes, and 3 or more filter stages
4. High DCR at any stage is undesirable

Point 1 I strongly believe in, though it means you will probably have to buy a new power trannie, and also probably a new choke (since you need a choke capable of choke input designs, often called a 'swinging choke' designed with an air gap).

Points 2 and 3 are contentious, and those who practice this often use NOS paper-in-oil reservoir caps, or 'motor run' caps no larger than 20uF. Good quality chokes are very important here, such as 20H and DCR of less than 100 ohm.

In our application, I'd forget about piling on the Black Gates for that 500uF and instead upgrade your choke to 10H or more (ensuring low DCR). Then 100uF will probably be plenty, and you can use just one of your precious Black Gates. Not knowing the circuit design, I'm confused as to the 11k dropping resistor - if this power supply is for only one part of the circuit, then I'd probably try for a choke-input filter and very little DCR in the circuit. If you need multiple voltage taps for different parts of the circuit, then you could even try using a discrete power supply for that particular section. Be wary of direct-coupled circuits, though - since mains voltage variations could cause large swings in bias.

I hope this helps.

Robert Ang
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.