18W "Dual Lite" build

[Amped]

Hi guys.  After a lot of help in this forum I've managed to build my first amp - well the main part anyway.  Phase 2 is the cabinet (got most of the bits I need), but thought I'd post what I've managed to do so far. 

The amp is built, it works pretty darned well - I'm sure there is a couple of small issues (hopefully small) but I'm hoping that it will be easy to troubleshoot/fix (I'll post some voltages and questions after this initial post).

The amp is a Dual Lite from Steve's page: http://sluckeyamps.com/dual_lite/dual_lite.htm The idea behind his amp design was exactly what I was looking for, something clean and a little quirky (i.e. not a direct clone) and I'm glad Steve pointed me to his site (thanks again mate). 
Due to how I approached it I made a couple of changes (hope that's OK with you Steve), the chassis, turret board and transformers I used was from an off-the-shelf TMB 18W package: https://www.evatco.com.au/18w-tmb-kickstarter-pack  This seemed like a perfect fit for the project, and I wouldn't have to make the chassis.  I altered the turret board, and there was a few extra holes I needed to fill so the changes I made were; I added a PPIMV (design from Steve - as a possible "mod" to his Dual Lite), and changed the input from 1 jack to 4 (2 per channel):

• Marshall 18W: Marshall 18W parallel triode/single triode but with 33K Grid Stoppers
• Vox AC15: Fender/Marshall style Hi/Low with 68K grid stoppers

It was a hard decision this one, I liked the switching input (very smart design) and didn't really know how it would react but there was 4 spare holes looking for input jacks, what was I to do ?

I've learned a lot, and I know there's heaps more to learn too, its been a real experience.  There's a couple of things I'd do different, I built the turret board and then installed it (I think it would have been easier to mount the board all wired up and then add components), and use liquid flux for soldering wires to turrets (suckers just wouldn't stick otherwise, had heaps of trouble with dry solder joints - hopefully I fixed them all,there's some messy ones in there...).  My soldering iron was pretty rubbish too, once I borrowed a 1/2 decent one from a friend it was a lot easier.  The chassis was quite small too, I don't regret doing it with this chassis (yet) but I think if I have to go back in (I still may have to) it may be difficult.

As mentioned it works, but I've got a small concern re: the voltages (bit high, I think because of the rectifier - a Sovtek 5Y3GT which from my reading can pump out voltages in the order of ~50V above a normal 5Y3 which is roughly what I'm seeing).  The Marshall 18W circuit tone control doesn't appear to do anything, and there's also some funky-ness around all of the pots related to hum too which I'll elaborate on a bit later... 

I've injured my hand and won't be able to play it (or make the cabinet) for another 3 weeks or so minimum which is frustrating, but maybe I can get it going better before then 

Attached is a couple of pics (a gut shot and a running amp).  I'm happy for any comments and advice for the future

Thanks again to the forum, and everyone who's helped me, its been a blast so far 

[Amped]

Voltages in attached .pdf (next to original reference voltages from Steve, bold and larger font...)

Seems a bit hot(?).  As mentioned earlier it may be due to rectifier being a Sovtek 5Y3GT(?).  It works, any issues running these voltages (especially EL84's)?  Also the EF806 grid being hotter than the anode - seems a bit weird to me, will it cause damage?(?).

[Amped]

There's a weird issue with the pots; depends on setting of the pots but they can cause the speakers to "pop" at about level 2 (nothing seems to happen until ~2 either) it seems to happen when most controls are at "0" if others are above 2 it doesn't seem to happen. 

The easiest to reproduce is all dials at '0', then turn PPIMV up, and when it passes 2 the "pop" occurs.  Hum steadily increases until it hits "10" then hum dramatically stops.

All dials at '0', if you set PPIMV to anything above 2 (some hum) and turn any other pot past 2, a "pop" is heard and the hum decreases slightly (even seems to change?).  Continuing to rotate dial up and hum slowly increases.  When you get to 9/10, another small "pop" and hum slightly decreases.  Slight differences if plugged into Vox or Marshall channel while doing, Vox channel is more noisy - is in general too).  If you have a control above "0" the amp works pretty well, it only seems like this issue is really if everything is set to "0"(?).

NOTE:  When this is happening, controls don't seem to work until after popping sound at ~2. 

I did take a shortcut I wonder if this is causing the issue(?).  I earthed the PPIMV to star ground where power supplies and EL84 cathodes is earthed rather than directly on the EL84 cathode earth, and I also earthed the shielded cables for the PPIMV on the same wire.  Actually it was a mistake but thought I might get away with it.  Thoughts?  Should I separate the shield garbage earth from the PPIMV, and change that to earth closer to the power valves on the turret board?  Its about a 3" run as opposed to a 1" shared wire...

NOTE:  All input wires and PPIMV are shielded wire (PPIMV was long runs and pretty close to the PT).

Also, the tone control on the Marshall 18W channel doesn't seem to do anything at all.  I'll go over the circuit and see if I made a mistake...

[Amped]

Hmmmm...

After a few hours of mucking around with it, things seem to have settled down a bit.  Is that even possible - is there a "burn-in" period with a new amp?

I'll try again tomorrow when its cooled down and see if it is actually better, or I'm imagining things.

[choosebronze]

I can't speak to the pot problem, but the Sovtek 5Y3 is nothing but problems. Word on the street is they are not actually a 5Y3, and they definitely give higher voltage.

[purpletele]

Nice work!!

[Amped]

I can't speak to the pot problem, but the Sovtek 5Y3 is nothing but problems.

Yea I wish I did a bit more research before I committed.  140mA compared to other brands seemed like a good idea, if the model is the same must be compatible right?  Came with a glowing reference from someone too which didn't help.  Ah well, maybe a JJ might be better.

[Amped]

Nice work!!

Thanks mate 

[Amped]

Hi again.  Here's a schematic I made with the alterations added.

It sounds good (mostly) - It feels like I'm 99% there, but I'm having some issues that need some troubleshooting and as I'm an amateur I'm not quite sure where to start. 

A thump is heard when all knobs set to "0" and you move any other knob past ~2/3.  Also when moving Various other random sounds and anomalies turning the various controls, they don't seem tied to any specific channel or control - although the issue seems slightly more pronounced on some controls than others (namely the EF86 Vox channel).  It is really weird.  If you turn a knob past 2 (past the "thump") and advance volumes on the 12AX7 channel (Marshall) it seems clear as a bell, but if you advance the Vox channel volume it gets noisey (must be a problem with the Vox channel?  Interference/lead dress?).

Reading around I don't think the EF86 screen voltage being slightly higher than the plate is an issue.  Pretty sure I didn't mess up the components but will go over them to see if there's a wrong value somewhere.  Only other thing I can think of is lead dress - some interference due to wires or earth issues(?).  However the Vox channel is definitely more noisy than the Marshall channel. 
NOTE:  I am testing in a quiet room and it might be possible that I'm being a bit picky (apart from the thumping thing of course) but it can sound real clean at some settings (12AX7 especially).  The EF86 might just be a bit more noisy than the other too (?).  Anyway some advice would be awesome.



Would anyone be able to provide some advice on where I should be starting to look?

Thanks guys.

[shooter]

provide some advice

since the amp is a "proven design" I would focus there, what makes yours diferent than original.
there are 3 yellow-ish tabs at the top for troubleshooting, read 'em, try 'em

it might be possible that I'm being a bit picky

my rule of thumb;
if it goes away with guitar on 1 and amp knobs set to 5, I leave it for now and fix the "real" problems

[PRR]

> Only other thing I can think of is lead dress

+1

The "thump" at some point on the dial is typical of the amplifier chain going into oscillation because signal in later stages is sneaking-back to earlier stages. This throws-off DC bias, and you can probably see that on a meter.

I once loosely copied Fender's Champ layout and was astonished how sensitive the tone-stack leader were to layout.

[Amped]

provide some advice

since the amp is a "proven design" I would focus there, what makes yours diferent than original.
there are 3 yellow-ish tabs at the top for troubleshooting, read 'em, try 'em

it might be possible that I'm being a bit picky

my rule of thumb;
if it goes away with guitar on 1 and amp knobs set to 5, I leave it for now and fix the "real" problems

Thanks Shooter, I'm follow your train of thought.  I'll also read through those troubleshooting articles and see if I can find anything that will help.

[Amped]

> Only other thing I can think of is lead dress

+1

The "thump" at some point on the dial is typical of the amplifier chain going into oscillation because signal in later stages is sneaking-back to earlier stages. This throws-off DC bias, and you can probably see that on a meter.

I once loosely copied Fender's Champ layout and was astonished how sensitive the tone-stack leader were to layout.

Awesome thanks for the advice.  I'll put a meter on it and see if I can see anything going on, it does seem like that would be problematic.  There's a few control wires - and an earth passing close by the 6 way switch that I'll investigate as a starting point. 

Quick question (something that has been bugging me), the OT wires, I've twisted the 2 input wires and they go straight to the power valve anode (pin7) close to the other wires around the socket (cathode, grids etc.).  Also the 3 output wires (4,8,16 taps) I've plaited together and they are pretty close to the sockets and switch above.  Question; can these wires live together nicely, or do they need separating?  Can't seem to find any references to correct OT lead dress anywhere...

[bmccowan]

There's a couple of things I'd do different... and use liquid flux for soldering wires to turrets (suckers just wouldn't stick otherwise, had heaps of trouble with dry solder joints - hopefully I fixed them all,there's some messy ones in there...).

Due to this quote from your initial post, I am wondering about your solder joints. Soldering to turrets should be no problem. If you use the right solder there is no need for additional liquid flux. Having heaps of trouble with solder joints is very likely associated with the issues. Having a couple of bad solder joints is not unusual, and reflowing them fixes them unless there is dirt/grease. So I'd consider if you are using the right type solder and the right technique, the instructions on Doug's site are good. If you do have bad solder joints, you will be chasing gremlins for a long time.

[dunner84]

+1 about soldering; however, I have read on other forums about bad batches of turrets. I have stumbled across 2 or 3 myself.
Also, having an iron that suits the turret is important! I will reuse turret boards out of old Hammonds, etc, and they have these nice thin, easy to solder turrets. The ones I buy today online, are MUCH heavier, and require a bigger soldering iron.

I didn't think much on this until recently. I set a friend up with parts for a small amp project, and he called me over because he was having issues... The thing looks terrible! Big globs if solder everywhere, and none of it stuck to the turrets..
His iron was simply not powerful enough to heat the turrets... Well it was, but it took a lot of time per turret. I lent him my iron, and problem solved!

[Amped]

Due to this quote from your initial post, I am wondering about your solder joints. Soldering to turrets should be no problem. If you use the right solder there is no need for additional liquid flux. Having heaps of trouble with solder joints is very likely associated with the issues. Having a couple of bad solder joints is not unusual, and reflowing them fixes them unless there is dirt/grease. So I'd consider if you are using the right type solder and the right technique, the instructions on Doug's site are good. If you do have bad solder joints, you will be chasing gremlins for a long time.

+1 about soldering; however, I have read on other forums about bad batches of turrets. I have stumbled across 2 or 3 myself.
Also, having an iron that suits the turret is important!

The turret board I used had some turrets already and I re-used them if they were in the right place.  Those were the worst for soldering - maybe some contaminant on the surface(?), the new ones I added were much better (That and using a slightly higher power iron helped too).  Adding the liquid flux to the equation made a lot of difference also, using extra flux made soldering them dead simple.  I was surprised with the difference it made. 
I've looked over my solder joints with a magnifying glass, they look OK (putting aside my ugly soldering) but I will definitely have another look just to be sure (there was a cap over a turret hole that looked OK but when I pulled on it and it broke away). 

Thanks to you both for the advice.

[Amped]

Hi all, it's been a while but thought I'd update the forum with what I've been up to.  I re-built the amp much closer to the original design to simplify and start from something that was proven to work.  From there I made changes to layout and various components, it's still not perfect but pretty bloody close.  Sounds awesome, taken it out a couple of times now and it works well.

The only outstanding problem is with all dials at "0", if you turn the presence knob up it squeals (oscillation).  If any other knobs are turned up past say 2/3, problem goes away.  It's never used in that position so I can live with it - but if anyone has any suggestions I would be interested.

There are some changes I've made to the original to get things under control, I'll list them here:
* The PT is smaller in output voltage than the one in Steve's diagram but B+ was actually higher after rectification.  Was a bit hot so I tried to address that, after a few different changes I finally settled on an 18K dropper to replace the 8.2K between B and C, and a 5R4 rectifier.  A=339V, B=315V, C=255V, D=252V.  To be fair there wasn't much difference between 5Y3 and 5R4, both seem to work fine, just a bit hotter (A=361V, B=338V, C=270V, D=269V).
* Changed the screen resistor from 100R to 1K.  If I was to do it again I'd have 2 1K resistors to PV screens.
* EF86 was the biggest problem with this build - very noisy and sensitive to tapping (the switch in particular).  Swapped EF86 Tone switch and Volume controls around so wires routed better.  Used hotglue to secure caps on switch and covered legs (made them short and used teflon sleeve).  The biggest difference was the earth.  Moved the valve internal shield earth (2, 7) from right beside the socket to the input jack earth using the coax shield for the input wire - this helped a lot.
* Changed the layout slightly due to using axial caps as opposed to radial. 
* Chanel switch moved beside input jack to solder directly together (removing first wire run) and make input wires shorter also.  Front panel layout isn't as good as Steve's original, but wiring was easier and wire runs shorter.

I also kinda ran out of heater wire, so these aren't as twisted as I would have liked but - heater noise isn't a problem so I don't really care (phew).

Still to do:  Front/rear panel and cabinet.  Still working on how to do a front/rear panel, but I've started on the cabinet using pine and Tasmanian oak (hardwood).

Some pics:

[Amped]

Oh and all resistors are 1W minimum.  The 1/2W ones were a bit noisy.

Modified diagrams:

[phsyconoodler]

Your build looks nice. Ive built about 2 dozen 18 watts of various types and your votages are near perfect. Not sure what youre complaining about there.
  Ive never needed flux when soldering

[sluckey]

The only outstanding problem is with all dials at "0", if you turn the presence knob up it squeals (oscillation).  If any other knobs are turned up past say 2/3, problem goes away.  It's never used in that position so I can live with it - but if anyone has any suggestions I would be interested.

When and how did you put a presence control in this amp? My original has none. The original has the Vox preamp feeding one input of the LTP PI and the Marshall preamp feeds the other input to the LTP PI. Doesn't leave any place to connect a typical Marshall style presence circuit.

But, assuming you have changed the circuit to allow a presence circuit in a NFB loop, then squealing may be caused by incorrect feedback phase. Swap the OT plate leads to see if the squeal goes away.

EDIT... I don't see any presence pot in your picture. Were you really referring to the CUT control? If so, forget about swapping OT leads.

[Amped]

Your build looks nice. Ive built about 2 dozen 18 watts of various types and your votages are near perfect. Not sure what youre complaining about there.
  Ive never needed flux when soldering

Thanks took a couple of goes to get it right but I got there. 
Nope, not complaining about voltages they are pretty good.  I had trouble with microphonics, feedback and oscillations and I was thinking could be attributed to voltage - especially V2 EF86 which I've heard can pretty sensitive to voltages over 100V.  Initially it was at 113V, but the change of rectifier and dropping resistor to 18K brought it down to 98V (with the 18K and 5Y3GT it was 103V which seemed OK too?).  In the end I don't know if it was the voltage drop or the earth change, or both that helped but it seems pretty good now.  Tricky little suckers those EF86's.
Just for reference; voltages stock Steve layout using the Hammond 290PAZ transformer and 5Y3GT rectifier are: A=356V, B=332V, C=301, D=298V which gave me 156V at V1, 113V at V2 (119V Screen), 222V V3, and 345V at V4/5 (332V screen).
Yea in hind sight didn't need the flux, was just my iron was too cold and turrets needed cleaning first.  Some steel wool first and a 60W iron - soldering=easy.

EDIT... I don't see any presence pot in your picture. Were you really referring to the CUT control? If so, forget about swapping OT leads.

Sorry Steve - yes Cut control; don't know where "presence" came from-there's no presence control it's exactly like your schematic (except the two minor value changes I made to the dropper and V4/5 screen resistor).  Thanks again BTW for putting me onto you build, it's been fun to make, and sounds really great.  Someone else I know is going to build it also now

I probably didn't explain the last issue I have very well either;  with all dials at "0", if you turn the "Cut" to ~9 it squeals (so almost no resistance between each grid input, just the .0047 capacitor).  Tried a few resistors in parallel with the cap (33K, to 1M), and larger cap values - no change.  Was thinking a resistor in series with the cap slightly larger than the value the potentiometer is (so setting the resistance permanently to above the value ~9 on the pot is so you effectively can never dial resistance lower than that) might work, but that would change it for the whole amp and as mentioned if any other control is dialed up even a bit this doesn't occur.  It would have been a work-around not a fix so didn't bother.  If anyone has any ideas what might be happening I'd be interested in your thoughts 

[Amped]

On another note Steve, is your site gone?  I went back to have a look at your other projects and got a 404: Page not found. 

Probably a bit premature, but was starting to think about what my next build might look like