Hey all, the 35 watt Class A Cary SLA-70 seems to be a simple build. I was about to start an amp build based on the Quicksilver 8417 amp (with KT88) and came across the Cary Schematic (attached). I have purchased a 4.3k edcor tranny for the 8417 and my B+ can be right around 450V - 550V. I have the parts for the 8417 and the Cary SLA-70 (maybe not the exact OPT)...
Can anyone offer an opinion on which would be a better design and thus better sounding amp?
The Cary is designed around EL34s and a tranny to match. Don't know if the 4.3k OPT and KT88 with all else same would be workable...
I kinda like my Class A in SS, and would like to see what a Tube Class A would sound like.
Thanks in advanced.
Can anyone offer an opinion on which would be a better design and thus better sounding amp?
The Cary is designed around EL34s and a tranny to match. Don't know if the 4.3k OPT and KT88 with all else same would be workable...
I kinda like my Class A in SS, and would like to see what a Tube Class A would sound like.
Thanks in advanced.
Somehow pentode amps haven't been in vogue for a long time. And they tend to be more fussy about loadlines, so better get a perfect transformer - tube match.
As for the circuit, i quite like it but will choose a better supply for the screens - either a separate transformer/rectifier/choke/caps/the works or a good high voltage regulator. From past experience this makes a big difference.
As for the circuit, i quite like it but will choose a better supply for the screens - either a separate transformer/rectifier/choke/caps/the works or a good high voltage regulator. From past experience this makes a big difference.
The SLA-70 design is hideous, especially the power supply. 1100uF is WAY, WAY, WAY too large for use with hollow state power diodes, and paralleling them for a 36W amp is unnecessary. It's just asking for trouble. I'd also lose that "standby" switch, as breaking the DC center tap connection can expose you to high voltage if that switch breaks down.
It would be much better to use a smaller reservoir capacitor and clean up the ripple with an LC LPF after it. Either that, or stick silicon diodes in front of the 1100uF reservoir capacitor. Hollow state diodes aren't gonna last very long looking into a load like that.
The phase splitter is horrible, suitable as a quick 'n' dirty design used for a PA, plate modulator, or possibly a guitar amp. That's what this design looks like: modulator, PA or guitar amp.
It would be much better to use a smaller reservoir capacitor and clean up the ripple with an LC LPF after it. Either that, or stick silicon diodes in front of the 1100uF reservoir capacitor. Hollow state diodes aren't gonna last very long looking into a load like that.
The phase splitter is horrible, suitable as a quick 'n' dirty design used for a PA, plate modulator, or possibly a guitar amp. That's what this design looks like: modulator, PA or guitar amp.
Yes, I saw that - a 1100uf cap after the 5AR4 - max is 60uf? These must have arced over like crazy. Is that whey there are two in parallel?
Anyway, I have been playing around with TubeCAD and modeling the output tube section with Cathode bias. 525V B+, common cathode resistor of 330ohm bypassed with 120uf cap, 4.2k 60W Edcor output tranny with around 80ma per KT88 tube (36 watts/tube which is 90% dissipation) .
This gives 68wpc @4ohm, and around 10w class A...also has nice linear response too....
Need to look into PSUdesigner for the 525V B+. I have a pair of Edcor 360-0-360 250ma power transformers and a choke. I can go SS or I have a pair of 5U4 rectifiers (these will be monoblocks).
And looks like (per Miles) I need to do a little work on the front end...is a 6DJ8 / 12BH7 concertina like the Quicksilver 8417 better than the 6SL7 phase splitter of the Cary SLA70 (that bad, huh?)?
Anyway, I have been playing around with TubeCAD and modeling the output tube section with Cathode bias. 525V B+, common cathode resistor of 330ohm bypassed with 120uf cap, 4.2k 60W Edcor output tranny with around 80ma per KT88 tube (36 watts/tube which is 90% dissipation) .
This gives 68wpc @4ohm, and around 10w class A...also has nice linear response too....
Need to look into PSUdesigner for the 525V B+. I have a pair of Edcor 360-0-360 250ma power transformers and a choke. I can go SS or I have a pair of 5U4 rectifiers (these will be monoblocks).
And looks like (per Miles) I need to do a little work on the front end...is a 6DJ8 / 12BH7 concertina like the Quicksilver 8417 better than the 6SL7 phase splitter of the Cary SLA70 (that bad, huh?)?
A cathodyne/concertina splitter is better than that paraphase. It lacks any internal balance (it does include balancing resistor, but you'd need an o'scope to balance the thing) and doesn't have the same harmonic distortion per phase. It's strictly in the Q 'n' D category of design.
The 6SL7 makes a good LTP, but you lose half your gain when using it as a splitter. It can, however, be a bit thin directly driving finals, but pents operating as pents aren't difficult loads, so it may work OK. Grid dirivers are better, especially when it comes to handling overdrive. Once the grids start pulling current, those 6SL7s roll over and die on you.
Hey, I have been using this Triode calculator to calculate voltage gain, and output impedance from plate or cathode - used it from basic preamps designs...nifty for a novice like me (I know I need to read Morgan Jone's book)...but just noticed that all results are independent of B+ voltage...is this true?
Adam's Amplifiers: Triode Calculator
Anyway, the pair of 6SL7 (220k/2k) gives a +/- 50v swing. If I put a 6SL7 with 47k/470 gives a +/-17v swing, which is probably not enough to drive the KT88...
Maybe a 6SL7 driving a 6SN7 LTP like a Williamson circuit?
Adam's Amplifiers: Triode Calculator
Anyway, the pair of 6SL7 (220k/2k) gives a +/- 50v swing. If I put a 6SL7 with 47k/470 gives a +/-17v swing, which is probably not enough to drive the KT88...
Maybe a 6SL7 driving a 6SN7 LTP like a Williamson circuit?
Last edited:
Sorry, this was to be a 6SN7 with 47K/1.8k gives a +/- 17V swing...
Instead of the 6SL7 on the Cary front end, how about a LTP with 6SN7 or 6N1P? Concertina still better?
If you're talking about a whole redesign, then the end result won't be a Cary anymore. If that's the case, might as well scrap the whole idea, and go for a Le Renard. You'll get the same power out, maybe a bit more even. You already have a 4K3 OPT, and that's close to the 4K4 load for a pair of PP 6BQ6GA's. The DC rail voltage is a good deal lower.
Better front end: cascode LTP that eliminates an extra gain stage, even with gNFB, and a grid driver to take care of overdrive gracefully. Better power supply as well. None of that excessive reservoir capacitor/current surge problems, and one 5U4GB can supply all the DC required.
Yes, it is a basic design, but I used to have one and if correctly tuned it worked quite well. I would built it, though, there are million of better designs out there.
i for one favors huge caps in the psu, but only after a choke not after a rectifier cathode, i still use the recommended values...
since i wind my own power traffos and chokes, i do not use center tapped rectifier arrangements, instead i use a hybrid rectifier in case of conventional tubes like the kt88 and voltage doublers for tv tubes...
this is a good read....powersupplies
since i wind my own power traffos and chokes, i do not use center tapped rectifier arrangements, instead i use a hybrid rectifier in case of conventional tubes like the kt88 and voltage doublers for tv tubes...
this is a good read....powersupplies
The Le Renard - is there a schematic?
As far as a PS, I found a few monster 450V 2200uf caps in my stash (Apex Jr). Thinking a 5U4, 40uf cap (bypassed with .22 PP), 5H Choke, and then the 2200uf cap...still need to put it all in PSUdesigner.
I have quite a few other 450V caps, but these were just laying around....got four, was going to use one per monoblock.
Also if needed, can put a pair of 1N4007 in front of the 5U4, which essentially makes the 5U4 a soft start rather than a rectifier...kinda like the attached (this is from another amp).
I often hear of amps on this site that have poor design (ie - a Jadis JP200 linestage - with three 12AX7 per channel? Is it possible for this to sound good as a linestage?), and get confused thinking a good tubed front end slapped onto a good tubed output stage should make a great tubeamp. Not necessarily correct...nor that simple...
As far as a PS, I found a few monster 450V 2200uf caps in my stash (Apex Jr). Thinking a 5U4, 40uf cap (bypassed with .22 PP), 5H Choke, and then the 2200uf cap...still need to put it all in PSUdesigner.
I have quite a few other 450V caps, but these were just laying around....got four, was going to use one per monoblock.
Also if needed, can put a pair of 1N4007 in front of the 5U4, which essentially makes the 5U4 a soft start rather than a rectifier...kinda like the attached (this is from another amp).
I often hear of amps on this site that have poor design (ie - a Jadis JP200 linestage - with three 12AX7 per channel? Is it possible for this to sound good as a linestage?), and get confused thinking a good tubed front end slapped onto a good tubed output stage should make a great tubeamp. Not necessarily correct...nor that simple...
Attachments
Last edited:
- Status
- Not open for further replies.