The results of heat on PCB mounted tube sockets...

[tubetwang]

I used boards when first exposed to d.i.y. hi-fi tube building 5 years ago.
Was complete newbie then. Apprentice buider today.

Some boards were junk and some were great (Uncle Ned's (Triode Electronics) Shannon Parks's (diytube.com). I built 5 or 6 amps using boards (Dynaco ST-70 and ST-35). I may turn them into Dumble's and pay off the morgage.

But step on it long enough (tweak, re-tweak and re-re tweak) and you start having problems. The kind of problems that most newbie want to avoid...(but left alone, it does a great job).

Pcb's are great for newbies and others who want an easy build.

Today, i'm the proud owner of a point to point (rat nested and long hours) AA761 Champ, 2 x MA1, 3 x 5C3 Deluxe, 5B6 Bassman + wood and metal shop. I do not think i'd be there without that first pcb, Tino Zottola, Internet community, Freedom 50 and all the tube amp's books and videos purchased through the years.

PS. Interesting IAG boards at Partsconnexion.com (no flux problem on those...)

PS.. i would like to thank my 9 year old and fiance for putting up with me and believing in my talent...(':roll:')

[scottosan]

"DISCLAIMER: This is not intended to start a flame war >> It's informational only."

Not when the title of your post is "The results of heat on PCB mounted tube sockets...".

One board in one Soldano amp does not represent EVERY PCB with on-board tube sockets. Its ignorant to even remotely assert that. There are dozens of PCB guitar and hi-fi amps with tubes mounted on the board that have been around since the 1960's that are fine. There are also poorly manufactured guitar and hi-fi PCB amps that have failed miserably.

Posting a picture of a poorly manufactured PCB with a meltdown and shouting "ALL PCB'S ARE BAD!" doesn't prove anything, except that your particular Soldano Astroverb had a poorly manufactured PCB. But at least now we know you're a troll.

How many PCB have you had to replace components on? Seriously? The main problemis that the traces can peel away form the board due to heat form desoldering even with the most experienced techs.

[zaphod_phil]

Pcb's are great for newbies and others who want an easy build.

And trust me, there's nothing wrong with either of those...

[jersey_aaron]

>How many PCB have you had to replace components on? Seriously? The >main problemis that the traces can peel away form the board due to >heat form desoldering even with the most experienced techs.

How many boards have you worked on that were designed by us? After over a dozen re-solders on several components on our PCB, not a single trace has lifted.

Most people only have experience with PCB's for mass produced products. Most of them are made cheaply and soldered by machines. But a well designed board will not exhibit the typical problems of mass production (i.e. boards made for cost savings).

[scottosan]

>How many PCB have you had to replace components on? Seriously? The >main problemis that the traces can peel away form the board due to >heat form desoldering even with the most experienced techs.

How many boards have you worked on that were designed by us? After over a dozen re-solders on several components on our PCB, not a single trace has lifted.

Most people only have experience with PCB's for mass produced products. Most of them are made cheaply and soldered by machines. But a well designed board will not exhibit the typical problems of mass production (i.e. boards made for cost savings).

Your boars have not been around long enought to show signs of aging and natural chemical settling that occurs in all electronocs, PTP or PCB. What is the thickness of your boards? did you use 1oz. copper clad? Regardless how well you design the board, certian componts will/can fail. IF a diode goes on a PCB, its usually a mess, screen resistors, the same. I am not saying your product is bad, it just the nature of PCB. Not to mention the whole premise that a newbie will learn alot more when they have to learn how to layout components and see how they interact with each other. If I was to PCB DIY, then I would go a step furhter and etch my own board.

[jersey_aaron]

>Your boars have not been around long enought to show signs of aging and natural chemical settling that occurs in all electronocs, PTP or PCB.

Oh, I thought we were talking about how many times we can re-solder a well made PCB. You said that was the biggest problem, right? But I guess we can switch the topic to aging..

>What is the thickness of your boards? did you use 1oz. copper clad?

I used the cheapest material possible to maximize my profit and screw the builder.

>Regardless how well you design the board, certian componts will/can fail. IF a diode goes on a PCB, its usually a mess, screen resistors, the same.

Screened FR4 is highly resistant to heat. You actually have to blow-torch it to make a mess of it. As shown here:



>I am not saying your product is bad, it just the nature of PCB.

No, its your experience with cheap PCB's made for mass production that makes you think that. List all of the PCB guitar amps made for DIY'ers that you have worked on.

>Not to mention the whole premise that a newbie will learn alot morewhen
>they have to learn how to layout components and see how they interact >with each other.

Since you are new to 18watt.com, I suggest you read up on the TMB.

>If I was to PCB DIY, then I would go a step furhter and etch my own board.

Then go to it big guy!!! Get that photocopier and go to town! Woohoo!!

[scottosan]

>Your boars have not been around long enought to show signs of aging and natural chemical settling that occurs in all electronocs, PTP or PCB.

Oh, I thought we were talking about how many times we can re-solder a well made PCB. You said that was the biggest problem, right? But I guess we can switch the topic to aging..

>What is the thickness of your boards? did you use 1oz. copper clad?

I used the cheapest material possible to maximize my profit and screw the builder.

>Regardless how well you design the board, certian componts will/can fail. IF a diode goes on a PCB, its usually a mess, screen resistors, the same.

Screened FR4 is highly resistant to heat. You actually have to blow-torch it to make a mess of it. As shown here:



>I am not saying your product is bad, it just the nature of PCB.

No, its your experience with cheap PCB's made for mass production that makes you think that. List all of the PCB guitar amps made for DIY'ers that you have worked on.

>Not to mention the whole premise that a newbie will learn alot morewhen
>they have to learn how to layout components and see how they interact >with each other.

Since you are new to 18watt.com, I suggest you read up on the TMB.

>If I was to PCB DIY, then I would go a step furhter and etch my own board.

Then go to it big guy!!! Get that photocopier and go to town! Woohoo!!

It doesn't matter how well the PCB is built, it will never be as easy to work on, repair, mod and a handwaired amp. Again your product has its place, but I think it has more limitations than benefits in the DIY community. Ask any experienced tech and they will choose work on a handwired amp over a PCB any day. Can you change the layout of a PCB? sure, you may have to cut a trace. I just don't see how one can learn painting by numbers. And you argument about quiteness of PCB over PTP. I would be happy to take anyone up on that challenge.

[scottosan]

I would also like to say that I am not totally anti PCB. The soldano SLO is one of the best laid out boards of all time. But no board mounted sockets and potentiometer and input jack

Its not just about marketing. Have you compared and original Plexi to a reissue? A real 59 Bassman to a reissue? Original twin reverb to a reissue?

[jersey_aaron]

>And you argument about quiteness of PCB over PTP. I would be happy > to take anyone up on that challenge.

Then take up StevesUk from the forum here, our first customer (who ordered two more of our boards). To quote him:

"This is the first of perhaps fifteen 18W and 36W TMBs, except one, I have built without squeal and hum when first switched on.. You can literally turn all volume controls and tone controls to the MAX and there is no feedback, squeal or oscillation..It is so silent, when the guitar is removed it could be left on without a problem As this one is intended for a recording studio, this is perfect."

What do you say about that, my newbie friend?

>Have you compared and original Plexi to a reissue? A real 59 Bassman >to a reissue? Original twin reverb to a reissue?

I don't make reissues and I am not Fender or Marshall. I make a JA TMB PCB. It does one thing and it does it very well. They are quiet, made well and are easy to assemble. They're also cheaper than any other TMB kit on the market and you can put one together in a day. Are they for everybody? No. Some people like PTP. Personally, my favorite amps were PCB, so I could care less. I'm all about tone.

[rjgtr]

There's probably an interesting psychological study in all this PCB vs PTP debate.

Both methods, properly done, have merits. I have a couple Riveras I've ben using for over 5 years with no marks on the PCBs and it uses PCB mounted sockets. So it can be done if the PCB is of the right quality. And if you read the past threads you'll see JA did a lot of work to perfect his PCB.

For my own builds I find PTP more convenient, but for a mass produced amp or "build and use" type of kit, PCB has a lot of merits. There's no need to be totally polarized.

JA wrote - "I'm all about tone" - that's what it's about.

[csbradley]

I was doing safety testing on one of our PCB's a little while ago. I was intentionally shorting components to see how it would fail, and make sure it would fail safely. Had a resistor in the power supply that was supposed to fail open just sit there and heat up.

Ok, the point of the story, the component got above 360 degree C, that was the limit of the thermal camara. Trace temperatures wasn't far behind. Test times were about 15 minutes per test. (That would be plenty of time to blow a fuse in normal conditions.) Solder melted, of course. And yes, the board scorched under the component. But the traces never lifted, and the PCB was usable after, and was able to continually replace the component and perform additional tests without a problem.

>The main problem is that the traces can peel away form the board due to
>heat form desoldering even with the most experienced techs.

Also, I'm sure its probably already been discussed, but if you continually solder and desolder on PCB's, you need to use the proper soldering iron and tip for the job. I wouldn't use the larger tips I use when working with Turrets and Eyelets. Using too large a soldering iron and too hot an iron will cause you to lift traces. In most cases, using the proper tools keeps these problems from happening. If you have problems with traces lifting, try using an iron with a temperature control.

If I lift a trace, its usually because I'm trying to desolder a component with multiple leads, and I can't heat all the pins at ones, so I have to work it out, causing mechanical stress on the PCB and traces. In these cases I should use the iron with hot air and suction, but thats just not always available, or I don't want to take the time to go get it. The proper tools really make a difference.

CSB

[tubetwang]

I'm no expert but all these points are valid.

I'll just add that most newbie will start with a PCB or Tino Zottola's books.

That is how i started. If you do not know what a resistor is, you are not going to built an amp from scratch and may be reluctant to get your hands dirty and invest...toolings alone can run you towards a divorce.

If you get addicted you may want to move on to whatever...

I do think there's merit in PCB.

I know some of my amps are noisy and i tend to play the quiet ones.

[dotfret]

I have been trying to be even-handed here, but I'm going to say where I'm coming from this time. If you think this is over the top, let me know, but I have points to make.
I am primarily a chemist, with specialisation in high polymer chemistry (so I know about the composition of the boards). I also studied electrics and I have a post-grad qualification in electronics from the University of Middlesex. You don't want to know about the rest - it is not relevant.
30 years ago I worked in a music shop part-time as a guitar setter, and I was able to try out a very wide range of everything. At the same time I helped in the repair shop, and got to see a lot of what went wrong. My main job was as a spectographer, so I got to see some really high voltages in action on a daily basis. Later, I also spent some time with a monitor manufacturer, so I know about thermal damage to PCBs on a commercial scale.
The trouble with mounting sockets directly to a PCB is in the mounting point. You can reduce the effect by making sure that there is the largest amount of copper around that point, introducing metal to act as a heat sink, whatever. In time it will go wrong - that may take 4 years or 30 years, it depends on the engineering. There are factors here that are incompletely understood - no-one has a complete understanding of why steel rusts, but it does. The analogy is that the bimetallic nature of the soldered joint has something to do with the failure, coupled with a lack of understanding of the adhesion of the track to the board. What we can say is that putting sockets directly on the board is a mistake, but it may take 20 years to show up as such. My number one amp is forty years old and has needed one output resistor replaced in that time. I've had a couple of others with PCBs that needed a lot more work. My opinion is that it has something to do with the nature of crystalline tin - do you know the story of Napoleon's troops besieging Moscow - I think the solder "crystallises" to some extent and exhibits the heat generating properties associated with increased resistance, which becomes more noticable when it is only in contact with a thin film of copper.

[jersey_aaron]

Sounds like voodoo. But you know, 30 years was a long time ago and the PCB technology back then was crap. Circuits were still layed out on gigantic reams of paper because they didn't have PCB design software. They couldn't even take full advantage of the technology. Boards back then weren't made out of hi-temp fiberglass. Hell, some amp boards today aren't either!

But you know, from a manufacturers standpoint, if a component fails after 30 years, they don't call that "Going wrong", they call it "built to last".

My 1971 Ampeg V4 PCB-based amp is still kicking a$$ though!

[EBE]

This seems like a never ending debate. I have put the JA 2.0 TMB values in one of my builds (my own layout though) and it is a very pleasing amp. Good cranked tones. Still I will never do a PCB. Just me. More power to anyone who builds one. Different strokes for different folks. You are never going to convince the AVHs that this is something they need - so why try.

[dotfret]

Not voodoo, just an incompletely understood subject, and you display your own ignorance when you refer to it as such. I understand your desire to sell what you believe in, but be tempered by experience. I used to advocate Dan Armstrong's amps as a great sounding and undervalued design to my friends. They were cheap, they sounded good. The sockets were under-rated. Most of those amps went up in flames before those sockets could be changed to ceramics, which were hard to come by 25 years ago. That was my stupidity in the face of fact. It does take years before the stupidity of mounting sockets onto PCBs shows up. I have already said that I cannot give a complete explanation. I just know that IN THE LONG TERM IT IS STUPID, AND THAT THERE ARE ALTERNATIVES WHICH CAN STILL MAKE USE OF PCB technology.

[s2]

I still owe you guys some photos of the Fender. The one with the discolored steel chassis from having been used hard for 10 years, but with a pristine PCB holding the sockets.

I can't bring myself to call it a quality amp, but I see no problems at all with the PCB holding the sockets in this particular amp.

It is nice to know some PCBs will outlast the average car.

[bitsandvolts]

There are plenty of rock solid Peavey's out there that are 20+ years old...and still rockin....granted, some of those are solid states...but some of those old Peavey Mace amps would take a few bullets before crapping out...

[dotfret]

Hey, CS Bradley - you lift a track because you don't use enough power! The trick is to apply just enough heat to remove the connection in question - if the heat goes down the line, it is not sufficient in the primary location. To move a multi-pin and recover it, lift each leg a little at a time with leverage (eg insert a screwdriver under it and lever while unsoldering). If it is not worth recovering because it is dead, cut the legs off and take the pins out individually. On a commercial scale, you solder PCBs by putting all the components on the board and float them across a bath of molten solder, not taking too long about it! It is time versus energy. I make a really neat job soldering with a 65W gun using crocodile clips as heat sinks on thermally sensitive components (eg FETs), but I used to make a mess with a 15W iron trying to do the same thing gently. Localise that need for heat, and understand how that thing was created.