|
Vintage Audio (record players, hi-fi etc) Amplifiers, speakers, gramophones and other audio equipment. |
|
Thread Tools |
3rd Dec 2021, 7:47 am | #1 |
Octode
Join Date: Oct 2019
Location: Stockport, Greater Manchester, UK.
Posts: 1,612
|
Olson audition amplifier
Just wanting to pick your brains
I'm looking for a good easily replicatable circuit for an all octal 6v6 PP build on a chassis I've acquired. The original lineup was a pentode to concertina phase splitter to 6v6 PP. The obvious choice is a Mullard 5-10 type build with an ef37a -> 6sl7/6sn7 phase splitter -> 6v6 PP with feedback. Has anyone got a good working circuit for this, as I cannot seem to find one? I am however really intrigued by this Olson audition amplifier build. It is a zero negative feedback build with allegedly excellent operational parameters (see attachments) and it's simple enough for me to replicate, even though it'll probably be a bit too sensitive. The only difference is that I won't run a quad 6f6 triode in the output - I may go down the standard 6v6 pentode or ultralinear mode and possibly apply some feedback to decrease the input sensitivity. We'll see. However, the more I look at the circuit, the more concerns I've got. The transformer is a 390-0-390v This means that: 1. The voltage on the plates of the 5y3 vastly exceed maximum values 2. The current required by the valves exceeds the maximum the valve could provide 3. The voltage on the plate and screen of the 6f6s is expected to be well in excess of 400v which is much more than the 6f6s can handle 4. The 6sn7 concertina is starved of current with the massive 5600ohm bias resistor. It needs 9mA, with that setup it will barely draw 1-2mA. Will it even work? So, am I missing something? Has anyone built this? Can anyone suggest a good all octal amp build? Thanks Gabriel |
3rd Dec 2021, 10:11 am | #2 |
Heptode
Join Date: May 2017
Location: Konongo, Ghana
Posts: 510
|
Re: Olson audition amplifier
A 5Y3 can deliver 115 mA at 2 x 390 V. What makes you think that this amplifier will draw more than 115 mA?
In this RCA Application Note from 1935 on class AB triode operation of the 6F6, note 3 on page 6 states that the zero-signal plate voltage used for the summary table is 350 V. The plate voltage in the Olson amplifier will not be so far off of that (I estimate the voltage drop over the cathode resistor to be around 30 V). http://www.one-electron.com/Archives...%20Triodes.pdf Although not hard evidence, I would think that RCA/Olson would not publish an article like this if it would destroy the valves used in them.
__________________
Robert |
3rd Dec 2021, 10:28 am | #3 |
Octode
Join Date: Oct 2019
Location: Stockport, Greater Manchester, UK.
Posts: 1,612
|
Re: Olson audition amplifier
I would agree with you but I cannot get my head around it. They should know what they are doing with their own valves.
Maximum signal plate current for a pair of 6k6s in triode mode is 60ma. There are 2 pairs so 120ma. 6j5 will draw 9ma approximately, so will the non cathodyne half of the 6sn7. The cathodyne half of the 6sn7 has a massive cathode resistor which will limit the current through it to about 2mA according to the valve graphs. I can't remember the exact figure but I calculated it earlier. So current at max signal estimated at 140mA, unless I'm missing something. 6f6 screen voltage tap is prior to the resistor, directly out of rectifier. 390v in will give about 425 v out, very conservatively, with a 40uf reservoir cap, which is also overrated for that valve, but probably ok. The only drop is the output transformer so I estimate 400v on those 6f6s. Last edited by Gabe001; 3rd Dec 2021 at 10:37 am. |
3rd Dec 2021, 12:23 pm | #4 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
|
Re: Olson audition amplifier
Indeed they do!
But you should question why they publish a design in the first place. Inevitably it's done to aid sales. Usually there is some characteristic they want to showcase..... how cheap, how powerful, how trendy. So expect valve manufacturer designs to be somewhat distorted to show off something. Sometimes this aspect is enhanced at the cost of other issues. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
3rd Dec 2021, 2:39 pm | #5 |
Heptode
Join Date: May 2017
Location: Konongo, Ghana
Posts: 510
|
Re: Olson audition amplifier
At zero signal, the current is 25 mA per 6F6.
I think that the 6J5 will draw less current than 9 mA. If it would draw 9 mA, the cathode voltage would be 16.2 V, while the voltage drop over R4 + R18 + R17 would be higher than 294.3 V (higher because I calculated with only 9 mA running through R17 so I didn't include the currents of the 6SN7 that also run through R17). This would leave only about 100 V over the 6J5. It's clear from the datasheet for the 6J5 that 9 mA at Vac = 100 V and Vg = - 16.2 V is impossible. Assuming that the voltage at the point where R17 and R18 connect is about 330 V, I calculate the current to be about 4 mA. I agree with your estimates of 9 mA and 2 mA as the currents through the 6SN7. I don't agree with your estimate of 425 V out from the 5Y3. The datasheet for the 5Y3 states that at 2 x 350 V and I = 125 mA, the output voltage is 350 V. So I would estimate the output voltage in the Olson amplifier to be 400 V or even less. So that would give a total of about 115 mA under no signal conditions. At full power the current draw could go up to about 135 mA. I would think that as long as the amplifier is used with music or speech, this is not a problem because peaks will be of short duration. But on the other hand: The (sometimes bass-heavy) music of today is probably not something Olson had in mind when designing this amplifier.
__________________
Robert |
3rd Dec 2021, 5:09 pm | #6 |
Dekatron
Join Date: Sep 2015
Location: Southwold, Suffolk, UK.
Posts: 8,302
|
Re: Olson audition amplifier
Noting the circuit diagram in Post #1, is that also known as a "Parallel Push-Pull" output stage?
__________________
Edward. |
3rd Dec 2021, 8:06 pm | #7 |
Dekatron
Join Date: Jan 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 8,171
|
Re: Olson audition amplifier
Hi Gabriel, search the forum for the Lola amp, this was publishes a few years ago and was a nice amp using 6V6 amd other octals
Ed |
3rd Dec 2021, 9:09 pm | #8 |
Octode
Join Date: Oct 2019
Location: Stockport, Greater Manchester, UK.
Posts: 1,612
|
Re: Olson audition amplifier
Hi ed, yes I am well aware of Lola. My first amp build was broadly similar to 'lola', but with different valves (ecc83 paralleled->6c4 concertina->6v6 PP). It works very well. Lola uses an ecc81 with one section for gain and another as a concertina, if I recall correctly. I also have a Mullard 5-10 in the pipeline, so I'm looking for something a bit different.
Robert, thank you very much for your detailed feedback and analysis. Excluding the four 6f6s (I'd replace these with a single pair of 6v6s), what do you think of the rest of the circuit? |
4th Dec 2021, 6:50 am | #9 |
Heptode
Join Date: May 2017
Location: Konongo, Ghana
Posts: 510
|
Re: Olson audition amplifier
The value of the anode resistors of the 6J5 and the first half of the 6SN7 is 22K. That value only just obeys the rule of thumb holding that the anode resistor of a triode should be at least 3 x the internal resistance. Maybe 22K was chosen for the 6J5 in order to keep the value of the volume pot low (for AC-signals R4 and R6+R7 are in parallel)?
Thats about all I can come up with. Probably Olson and Morgan had good reasons for their design choices, reasons that are outside my line of (in)sight. Addition: For modern sources the gain of this amplifier, which has no GNFB, may be a bit high.
__________________
Robert Last edited by Robert Gribnau; 4th Dec 2021 at 6:53 am. Reason: Addition |
4th Dec 2021, 7:06 am | #10 |
Heptode
Join Date: May 2017
Location: Konongo, Ghana
Posts: 510
|
Re: Olson audition amplifier
My understanding is that "Parallel Push-Pull" in the old(er) days meant a topology like in the attached schematic. But I see the term being used more and more for push-pull amplifiers with 4 (or more) power tubes with pairs (or more) of 'truly' paralleled power tubes, like in the Olson amplifier.
__________________
Robert |
4th Dec 2021, 8:12 am | #11 |
Dekatron
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,636
|
Re: Olson audition amplifier
Found this one over on DIY audio - http://diyaudioprojects.com/Schemati...Tube-Amplifier, can't get much simpler than that. Tell a lie, loose the GZ34 and pop in a silicon bridge rectifier and use a 240v 1:1 isolation tfmr and a small filament tfmr.
By the number of times I've seen this same schematic tweaked and built I'd say it was a potential candidate. Another here - https://antiqueradios.com/forums/viewtopic.php?t=233121 one 6V6 wired grounded grid, looks interesting Andy.
__________________
Curiosity hasn't killed this cat...so far. |
4th Dec 2021, 10:09 am | #12 | |
Dekatron
Join Date: Sep 2015
Location: Southwold, Suffolk, UK.
Posts: 8,302
|
Re: Olson audition amplifier
Quote:
__________________
Edward. |
|
4th Dec 2021, 12:06 pm | #13 |
Dekatron
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
|
Re: Olson audition amplifier
The Olson design is not exactly hi-fi and the cynic in me says maybe it was to help RCA shift excess 6F6 stock as most manufacturers would have moved to miniature valves by 1950.
|
4th Dec 2021, 3:00 pm | #14 |
Octode
Join Date: May 2009
Location: Stevenage, Herts. UK.
Posts: 1,515
|
Re: Olson audition amplifier
I did wonder that, this is a bit of a marketing exercise as they're also pushing their brand new 45s only changer.
Olson himself was no slouch and invented isobaric speaker loading amongst other things. |
4th Dec 2021, 5:07 pm | #15 |
Octode
Join Date: Oct 2019
Location: Stockport, Greater Manchester, UK.
Posts: 1,612
|
Re: Olson audition amplifier
Thank you all for your comments.
PJL/WDAddict - I agree with you re 6f6s. Probably everyone had migrated to the 6v6s too earlier on. Andy: I am familiar with that circuit. It's a dynaco a 140 (or clone). The Chinese have perfected the art of migrating that to a PCB, there are about 100 minor variants! I have no doubt it's great, but it's probably too low gain for me. I need an input sensitivity of around 400mv so I can use it with a BT receiver, tablet, phone etc. That one (just a quick back of envelope calculation) probably runs at 4v. The problem with the Olson is that, as Robert pointed out, it will probably be too sensitive, but we'll see. |
4th Dec 2021, 8:39 pm | #16 |
Dekatron
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,636
|
Re: Olson audition amplifier
Yep, one circuit a Dynaco clone. I followed your last amp build and I think you have it in you to design a valve amp from scratch.This may sound scary but it's very easy to do. I'll qualify that last sentence, it's easy to build a valve amp that sounds good but doesn't have fantastic distortion/ frequency response specs. IMHO a bit of distortion or colour enhances the listening experience.
Inna nutshell you start with the OP stage lifted straight off the specs or parameters from an OP valve datasheet, that gives you your HT (and hence what mains tfmr you need), your OPT specs, HT current etc. That's 75% of the amp design more or less. Next bit is working out the circuit to drive the OP stage, not hard at all. Chuck it all together on a bench prototype and tweak, I think you'll be surprised how easy it is and the great results you can achieve. The upside is you'll learn far more than reading 100's of books or watching 100's of video's. The downside is you'll have difficulties, and feel like jumping off a bridge sometimes, but once you overcome that particular issue, your walking on air. Following a schematic is ok, but it's no fun, no challenge. I'll finish by saying designing from scratch isn't for everyone, and as other members will point out amplifier design can be a complex and involved process requiring a certain depth of knowledge to achieve a high specification. I'd counter by saying it's the process, the journey and the satisfaction at the end and it's possible to design and build a grand sounding amplifier, if your willing to have a go. Andy.
__________________
Curiosity hasn't killed this cat...so far. |
6th Dec 2021, 1:29 pm | #17 |
Octode
Join Date: Oct 2019
Location: Stockport, Greater Manchester, UK.
Posts: 1,612
|
Re: Olson audition amplifier
Thank you for the vote of confidence Andy. My first build on the thread you referred to was a complete homebrew job. I did this intentionally and as as you said it provided invaluable experience, although there is much to learn.
I thought I'd use an established circuit for this one, just to see how it sounds compared to the homebrew one, but it'll probably end up being a hybrid of this with an UL OP stage and possibly a bit of NFB. Still not 100% convinced though - I'll keep looking. |
9th Dec 2021, 8:22 am | #18 |
Dekatron
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,636
|
Re: Olson audition amplifier
Check out some old RCA valve books and books from transformer makers like Acrosound etc, the published amp designs. Another good place to look is here - 100 amplifiers - http://lilienthalengineering.com/100...iers-chapter-1 lots of schematics/amps discussed. Another source is to look for EL84 PP designs and sub 6V6's instead.
Andy.
__________________
Curiosity hasn't killed this cat...so far. |
9th Dec 2021, 10:04 am | #19 |
Octode
Join Date: Oct 2019
Location: Stockport, Greater Manchester, UK.
Posts: 1,612
|
Re: Olson audition amplifier
Hi Andy,
Thanks very much for the link. I see that the Olson is discussed on page 2 and they speak very highly of it. I didn't know that using the same valve in sequence improves the harmonic distortion. I note that the rectifier in the circuit shown on the page you linked is a 5v4 as opposed to the original 5y3, which makes more sense. Gabriel |
29th Dec 2021, 11:49 am | #20 |
Octode
Join Date: Aug 2008
Location: Rayleigh near Southend-On-Sea, Essex, UK.
Posts: 1,852
|
Re: Olson audition amplifier
Hi,
Here we go the info I talked about in the Bulletin thread elsewhere on the forum. There is 6V6 circuit from the Data Publications book, it’s only the circuit as the full scan won’t fit on the forum. The other is from the early Caxton Radio (dark blue covered) book – I’ve changed the feedback to g2 as I feel it wouldn’t have been the right phase. I nearly included these in the Sterns article in the Bulletin, but I decided not to as I must keep them focused and concise on the piece of equipment being featured, however I ‘ll use these in a subsequent article notably a real oddball amplifier using push pull KT81’s, which is the late 1940’s GEC relative of the 6V6. I’ve recently built a guitar amplifier squeezing 18 watts out of a pair of 6V6’s running in fixed bias with an HT of just over 300V. Terry |