UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Other Discussions > Homebrew Equipment

Notices

Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members.

Closed Thread
 
Thread Tools
Old 21st May 2021, 8:54 pm   #1
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,638
Default New valve amp build - 30w 6146 AB2 PP.

Over the last few months I've been fettling away on a new valve amp build. I swore this time I'd build something more simpler but as one does I've got a bit carried away and ended up going a bit off piste'. As all valve amp designs start I started experimenting first with the OP stage trying to find a good fit a few odd valves I've had stashed away for a while and mating them with some clones of Dynaco ST70 OPT's that Ed Dinning wound for me.

OP stage - After much experimenting with various odd valves including QQV03-20's, 6BG6GA's & 6DQ6B's I settled on 6146's running in class AB2. The OP valves are really wasted here as I could get over 70w with them in AB2 however 30w is plenty enough and I'm more interested in the direct connection of the driver stage to the OP stage that AB2 gives me rather than raw power.

Driver stage - The OP stage is driven by 12BH7 cathode followers run off a CCS this gives plenty of oomph at low-ish distortion. I'm not after ultra low THD, I'm after "sculpturing" the sound, adding just the right amount of 2nd harmonic, if I wanted low THD/straight wire with gain I'd use trannys and loads of NFB.

Phase splitter - again I'm using transistors for a CCS to get good AC balance using an ECC82 LTP, the only slight set back being the LTP has a bit of 3rd harmonic distortion,not to worry about though.

The part of the circuit that has been vexing me the most is the tone control/IP gain stage. I tried various circuit modifications to an active Baxandall tone circuit by for instance trying to increase gain by subbing an ECF82 for the 12AU7/12AX7 OP "opamp" not realising that than gain has to be spot on else the circuit won't work, see att. I'd painted myself into a corner by not having a clear design right from the start and so only had 1 1/2 B9A valve bases to play with as I fettled the chassis before having a clear idea of what I would build.

Lastly - power supply... again, like all my amp builds I tend to base my design around what bits I have, rather than the best or correct design,so with this amp as with the last one, the 807 amp, I used a choke I had as well as four 700u 250v caps, two in series to give 350u before and after the 15H choke. I had to wind two power transformers- the original ST70 clone was very loud and badly wound so I re-wound it to give me 0v, 450v, 470v, 500v HT, the original had a 330v 0v 330v HT sec with a 55v bias tap. With the extra room on the bobbin I overwound the primary and secondary with heavier gauge wire. The original pri was 220v, I rewound for 245v and put some extra turns on to give a higher V/T ratio. I also rewound a Japanese tfmr to give me a 100v bias supply & 100mA, a 3A heater winding and a 250v supply for the g2 supply. With AB2, the OP valve screen grid's have to be fed by a regulated PSU with a +/- 5% spec, the seperate tfmr and board gives a well regulated 175v with bullet proof SC protection.

. I built a circuit board cutting the traces with a Dremmel using the circuit Dave Cox kindly knocked up for me, see - https://www.vintage-radio.net/forum/...=167901&page=3 for details. This caused me loads of grief, it took me weeks to de-bug the circuit as well a small pile of dead Q's and mosfets, a lot of this was down to the construction method, Dremmeling a PCB just isn't good enough, it is a quick way to make a board,but just isn't accurate enough. In future I intend to get some PCB's made up, spending weeks of frustrated de-bugging is a waste of time and is bad for the soul.

Lastly, lastly, the chassis also took quite a bit of time to make, this is another aspect that calls for software design & professional manufacture, I came across this firm - https://www.meface.co.uk/bespoke-anodise...nt-panels/ in future I'll get the top panel and perhaps the front and back done to get a decent aesthetic result and hopefully save a lot of time mucking about with an angle grinder and files as well as lots of prep & painting.

The amp isn't quite finished but the left channel is built, it'll need testing then hopefully the rest won't take long. Anyhoo,enough of my waffle, here's some pics.....Andy.
Attached Thumbnails
Click image for larger version

Name:	layout prior to painting 1.jpeg
Views:	327
Size:	89.4 KB
ID:	234559   Click image for larger version

Name:	DSC00146.JPG
Views:	304
Size:	162.2 KB
ID:	234560   Click image for larger version

Name:	component layout.JPG
Views:	292
Size:	161.8 KB
ID:	234561   Click image for larger version

Name:	cathode follower IP stage x 2.JPG
Views:	283
Size:	162.5 KB
ID:	234562   Click image for larger version

Name:	de-coupling caps.JPG
Views:	290
Size:	161.6 KB
ID:	234563  

__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is online now  
Old 22nd May 2021, 12:33 am   #2
joebog1
Banned
 
Join Date: Jun 2015
Location: Mareeba, North Queensland, Australia
Posts: 2,704
Default Re: New valve amp build - 30w 6146 AB2 PP.

Another nice build Andy!!! The Tektronix solder tags work nicely!!! I wonder how you hide the "nipples" that hold them on top of the chassis.

Joe
joebog1 is offline  
Old 22nd May 2021, 6:46 am   #3
Electronpusher0
Nonode
 
Join Date: Jun 2016
Location: Bognor Regis, West Sussex, UK.
Posts: 2,288
Default Re: New valve amp build - 30w 6146 AB2 PP.

Looks really good Andy, I love the finish.
Metal bashing is where I always struggle with my projects.
Unfortunately the link to Meface does not work.

Peter
Electronpusher0 is offline  
Old 22nd May 2021, 8:18 am   #4
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,638
Default Re: New valve amp build - 30w 6146 AB2 PP.

Thanks Joe
Quote:
I wonder how you hide the "nipples" that hold them on top of the chassis.
I'm not sure there is a way so I don't try mate.

Thanks Peter,try this link - https://www.meface.co.uk/bespoke-ano...m-front-panels

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is online now  
Old 25th May 2021, 7:01 am   #5
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,638
Default Re: New valve amp build - 30w 6146 AB2 PP.

Things are getting interesting as the amp shows it's peccadilloes, EG odd LF instability. It's started ticking like a clock which is a new one on me, the cause* is the screen grid power supply which has gone unstable. I powered the amp up with a bench PSU in place of it, problem gone. * Could be a valve though causing the supply to got into SC protection.

These odd faults are quite entertaining and fun to solve... sometimes. What gets me is that one minute everything is fine, the next not. It's easy to believe in Gods and Gremlins hurling thunderbolts and hiding screwdrivers.

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is online now  
Old 29th May 2021, 11:04 am   #6
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,638
Default Re: New valve amp build - 30w 6146 AB2 PP.

Can't find the fault, it appears to be very low frequency instability causing the HT, bias etc to go haywire. Here's a link to a video - https://youtu.be/Tsd859OvDDI of a scope trace monitoring the HT. It's a bit hard to see but there is periodic rising and falling of DC, followed by what looks like static, odd.

One other indicative symptom is the screen grids or at least one of them is drawing a hell of a lot of current, causing a stopper resistor yesterday issuing smoke. There is also a periodic clicking from the OP valve/s followed by "fizzing",this is maybe what we are seeing on the scope trace? Could it be there is a anode/g2 or similar inter electrode short?

I've swopped OP valves, moved grounds and tried another screen grid supply,no improvement.Any ideas?

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is online now  
Old 29th May 2021, 12:41 pm   #7
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
Default Re: New valve amp build - 30w 6146 AB2 PP.

6146 are quite frisky beasties with gain well into VHF. Even for HF radio use they need some care in preventing self-oscillation at high frequencies.

What looks like low frequency oscillation may be a high frequency oscillation which is squegging.

True RF bottles can get into plenty of mischief with just the lengths of wire of typical audio construction. Usually, they force you into non-inductive stoppers right at the grid pin and at the anode top cap. Care is needed in really good VHF decoupling of G2. RF decoupling on heater lines can sometimes be needed.

The need for stoppers at anodes is nasty because you'd really like to have nice ceramic-insulated TC connectors for safety, but having components and a soldered joint hanging out of them is somewhere on the spectrum between manslaughter and a Darwin award.

RF stuff with these things is normally in a screwed-together perforated screen, or if there is an openable lid, it's got an interlock switch.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 29th May 2021, 1:00 pm   #8
G3PIJpeter
Hexode
 
Join Date: Nov 2011
Location: Chippenham, Wiltshire, UK.
Posts: 319
Default Re: New valve amp build - 30w 6146 AB2 PP.

I notice that your title says the 6146s are running in class AB2 (i.e. with grid current flowing during part of the driving cycle). There is little advantage to driving them beyond AB1 (where no grid current flows) so far as power output is concerned. If you have a beefy driver capable of standing some extra resistive loading, then that might reduce the impedance of the control grid circuit and stop what sounds like parasitics.

Peter G3PIJ
G3PIJpeter is offline  
Old 30th May 2021, 7:03 am   #9
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,638
Default Re: New valve amp build - 30w 6146 AB2 PP.

Sussed it, partially, it was a cathode follower way back at the start of the circuit that was oscillating, not entirely sure why, more investigation needed.

I pulled the OP stage, instability still present, went through the whole amp this way pulling out valves till oscillation stopped. I'll attach a schematic later so folk can see what's what.

Noted David. In all my amp's I try as best I can to observe good practice, EG short as poss connections, etc.

Not really after high power Peter, rather the direct coupling of OP stage and driver/rest of amp this gives very good frequency response, maybe a bit of LF boost, yet to do a proper sweepso not 100% sure.

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is online now  
Old 30th May 2021, 8:08 am   #10
Robert Gribnau
Heptode
 
Robert Gribnau's Avatar
 
Join Date: May 2017
Location: Konongo, Ghana
Posts: 510
Default Re: New valve amp build - 30w 6146 AB2 PP.

Hello Andy,

Impressive, very nice and neat.

Are you indeed using Tektronix solder tags? I read somewhere that these tags contain silver and therefore need solder that contains silver (and because of that Tektronix included some of that solder for servicing purposes with their new equipment). Are you using special solder? Or is this story false/exaggerated (I have some of these tags too and would like to use them)?
__________________
Robert
Robert Gribnau is offline  
Old 30th May 2021, 8:40 am   #11
vinrads
Dekatron
 
Join Date: Feb 2009
Location: Chesterfield, Derbyshire, UK.
Posts: 3,737
Default Re: New valve amp build - 30w 6146 AB2 PP.

Quote:
Originally Posted by joebog1 View Post
Another nice build Andy!!! The Tektronix solder tags work nicely!!! I wonder how you hide the "nipples" that hold them on top of the chassis.

Joe
Like this, make a top plate which is removable ,held on with the side pieces .Mick.
Attached Thumbnails
Click image for larger version

Name:	IMGP0011.jpg
Views:	166
Size:	71.9 KB
ID:	235079   Click image for larger version

Name:	IMGP0018.jpg
Views:	152
Size:	84.0 KB
ID:	235080  
vinrads is offline  
Old 31st May 2021, 8:11 am   #12
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,638
Default Re: New valve amp build - 30w 6146 AB2 PP.

Thanks Robert.Yes they're Tek ceramic wafers and yes you need to use "special" solder. Tek say there is a potential for the silver shoes to be become un-bonded from the ceramic if you use normal lead solder, however I use normal solder occasionally to tack components in place either prior to filling with 3% silver solder or for tacking on wires after,with no issues.

I've a few rolls of the Tek solder, but it doesn't last long, there's only about 2ft, however I've found a cheap source of 3% solder at CPC, it stinks and is lead free I think, so the joint looks dull, unlike to Tek stuff. You need to use a reet hot iron to solder these. I've seen new ceramic wafers on ebay recently, how good they are being Chineseum I've no idea.

Good idea Mick if you like metal bashing and are accurate, my holes have a habit of wandering from where I thought they should be, also that method looks ok for B9 type valves, bit tricky for octal with my 30mm step drill,but yes,the method is good, I'm quibbling.

Back to the fault, I thought I'd got to the bottom of it, a ticking ECC82 cathode follower. I've re-biased it, made the bias point more hot, which seems to have helped,but I think there's more going on than one unstable valve. One 6146 seems to be twitchy or maybe has an intermittent anode/screen grid SC. I think the pot at the IP is somehow altering the bias point of the CF, even though it's de-coupled by a cap, see attached schematic.

I've wondered about moving the volume pot from the IP, to after the tone section, to just before the triode gain stage,however this would mean adding another RC circuit, not something I want to do.

Anyhoo,onwards and upwards, will test again today to see if the changes I made last night have stopped the LF instability. TFL, Andy.
Attached Thumbnails
Click image for larger version

Name:	DSC00125.jpg
Views:	93
Size:	63.4 KB
ID:	235113  
__________________
Curiosity hasn't killed this cat...so far.

Last edited by Diabolical Artificer; 31st May 2021 at 8:31 am.
Diabolical Artificer is online now  
Old 31st May 2021, 8:53 am   #13
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,638
Default Re: New valve amp build - 30w 6146 AB2 PP.

This is what I think may have been the cause of the instability...Yesterday I pulled at the suspect CF and re-biased it a tad more positive which seems to have cured the issue, I also got rid of the diode. I think the pot ( see schematic) has been affecting things somehow,together with the diode. I think at power up as the heater warms up and current starts to flow, the cathode has been rising in voltage but somehow the grid has not been settling at it's bias point, could it be the pot has been pulling the grid down and the diode switching on & off? This has caused the valve to re-cycle through a "start up" condition, thus causing the whole amp to oscillate at LF. see attached image for use of diode.

Hope that makes sense.
Attached Thumbnails
Click image for larger version

Name:	CF diode.jpg
Views:	101
Size:	23.7 KB
ID:	235117  
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is online now  
Old 2nd Jun 2021, 2:24 pm   #14
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,638
Default Re: New valve amp build - 30w 6146 AB2 PP.

Well,after sort of solving the instability I tried moving the volume pot from before the first cathode follower, to after the tone circuit, EG after the last CF, I also fitted a tone defeat SW. With the tone control switched out the instability is back,but not as bad,with the tone control switched in,the instabilty is gone. This is a bit mad as with the TC SW'd out, there's one coupling cap/RC circuit,with the TC SW'd in, there's four coupling caps.

Anyhoo, I think at the heart of all this is the power supply, perhaps it's resonant at VLF. In an effort ot analyse the power supply I've looked at all my references and cannot find how to calculate whether a power supply is potentially resonant at LF or how to design for stability. I've looked in Radio Designers, Morgan Jones Valve amps, Crowhurst and loads more reference books as well as online.

As I have it, we want a PSU that has a low OP Z and not resonant at VLF, how one works that out I can't find out. My HT PSU consists of an IE tfmr with a Rpri of 7.5r, and an Rsec of 32r. Vsec is 325v AC RMS into a fuse, then a bridge rectifier then into a CLC filter of two 700u caps in series (350u) then a 5H choke, then two more 700u caps in series. All caps have 180k series resistors across them. HT is 410v. Load is 2562r as far as I can make out (410v HT/0.160mA) The Idc of 0.160mA is quiescent, I'd expect it to be about 1.5 times that at full whack.

I tried adding some small caps in parallel with the 2nd cap, a 10u and a 1u but no improvement. I worked out Xc and XL - which is Xc = 454r and Xl = 3140 at 100hz and the PSU source resistance as 45r, but after that am not sure where to go next. Any ideas welcome.

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is online now  
Old 9th Jun 2021, 7:27 pm   #15
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,638
Default Re: New valve amp build - 30w 6146 AB2 PP.

Well, for now the Gods are appeased and are looking down kindly. I stopped the LF oscillation by doing two things. First I re-routed the ground bus; as was both OP stages and all stages downstream shared the same ground bus. I whipped this out, grounded each OP stage with a thick black wire to the main grounding point, IE 2nd cap after the choke. All stages after this, EG cathode followers/drivers, LTP phase splitter, gain stage etc, I grounded in that order to a wire gnd bus coming after the OP stage.

Second I re-jigged the HT de-coupling with individual de-coupling for each stage apart from the hi-gain stage and CF that form an "op amp" on the OP of the tone control. Lastly I moved the volume pot from the just after the IP,to after the tone stack/before the gain stage. Result.

Before all this I stripped the amp right back to OP stage, drivers/CF's and LTP PS, I then added one stage at a time after ensuring there were no shennanigans. I recommend this approach if anyone ever gets instability issues and gets bogged down running round in circles trying to fault find. By stripping it all back and starting again,having a good look at your layout and basic stuff it's easier to get your head round the problem.

Anyhoo,here's a couple of pics of the some latest work,at present I'm about half way to building the other channel which should be a hell of a lot quicker now I've ironed out the wrinkles.

Andy.
Attached Thumbnails
Click image for larger version

Name:	DSC00157.JPG
Views:	113
Size:	161.5 KB
ID:	235653   Click image for larger version

Name:	DSC00158.JPG
Views:	102
Size:	160.8 KB
ID:	235654  
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is online now  
Old 9th Jun 2021, 9:17 pm   #16
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
Default Re: New valve amp build - 30w 6146 AB2 PP.

With the cathode follower, try a few hundred ohms in the anode connection, right at the valve holder pin. If necessary go up to a few k Ohms.

Cathode followers (like emitter followers and source followers) can be right devils to get stable. With some C to ground in a couple of places, they degrade into colpitts oscillators implemented by their strays.

Grid resistors help, but anode ones are a bit more powerful at taming them.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 9th Jun 2021, 10:43 pm   #17
Restoration73
Nonode
 
Restoration73's Avatar
 
Join Date: Jun 2013
Location: Surbiton, SW London, UK.
Posts: 2,801
Default Re: New valve amp build - 30w 6146 AB2 PP.

Many years ago I acquired a marine MF transmitter/receiver/dynamotor combo made
by Coastal Radio (Edinburgh). The AM transmitter used a single 6146 driven by 5763
oscillator, with a pair of 6146's used for high level modulation (mod transformer) and the results were excellent.
They probably thought it was better to use a single type of valve for RF and Audio to limit stocks needed.
https://radiohistory.uk/Coastalradio.htm
Restoration73 is offline  
Old 9th Jun 2021, 10:58 pm   #18
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,528
Default Re: New valve amp build - 30w 6146 AB2 PP.

Power supplies resonant at (V)LF.....

Are you using chokes in the filter circuit? These can cause very nasty resonance problems with large electrolytic capacitors if the choke is a good one with low series resistance. There's a neat trick with a series RC damping circuit in parallel with the dc output of the filter that can be handy at the expense of another gurt big electrolytic. R in series with the choke also works but generates extra heat and voltage drop.

Simulating your filter's response to a step input voltage can be quite an eye opener!
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is offline  
Old 10th Jun 2021, 12:42 pm   #19
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,638
Default Re: New valve amp build - 30w 6146 AB2 PP.

Thanks for the tips David.In a lot of Telequipment scopes and other similar gear of that period I've seen, as you say, all the CF's with 100r on the anodes.

The 6146 is a really nice valve, they're really under used in this amp. It'd be far better with higher wattage OPT's. Acrosound supplied a schematic for one of their OPT's and the Ampeg SVT used four in their SVT bass guitar amps. This amp though is a hodge podge of bits I had and is by way of an experiment also.

I did a load of calculations regarding the PSU Chris which showed the PSU's as being resonant at 3.8hz.Over on other threads it was recommended to try a 100 - 120r resistor to damp it, also suggested was an RC damping circuit. By my reckoning I'd need a 3400u cap, I don't have one of those but do have a big 4700u 400v jobbie,however there's no room inside the chassis and it's also overkill. If this was a more powerfull amp using the 6146's to their full potential I might consider it,however as is everything's stable so there's no need.

Thanks for all your IP & suggestions, Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is online now  
Old 10th Jun 2021, 3:18 pm   #20
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
Default Re: New valve amp build - 30w 6146 AB2 PP.

As noted, the 6146 is quite a 'frisky' valve for use in audio service: they are rated for RF power-amplifier service at up to 150MHz! - you should definitely include grid-stopper resistors [1K in each control-grid, 100 Ohms in each screen-grid suggested - ideally the screen-grid oned should be wirewound so they act as a lossy choke at VHF].

Inductance in the cathode-leads can be a problem too: the 6146 is provided with three pins connected to the cathode - these are all intended to be earthed to the chassis directly - if you have a few inches of wire from the cathode tags to a 'ground bus' those few inches can look like a resonant line at VHF and you can be well on your way to building an oscillator.

Whatever you do, don't allow the valves to operate with the screen-voltage supply connected but no anode-voltage - the resulting excess current can destroy the screen-grids in seconds. 6146s are particularly sensitive to this issue, I've destroyed a few in the past.

Also remember that though an amplifier may be stable under static conditions, once it is driven it may develop oscillations as the drive moves the operating-point of the valves into different parts of their characteristic curves. This can be spectacularly frustrating to identify!

(I used to build guitar-amplifiers using 807s and they were on occasions just as difficult to tame - though guitarists generally didn't seem to care that much!]
G6Tanuki is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 9:32 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.