I think I go decent parts inside, and tin can looks on outside. I am not trying to impress anyone. Just want it to sound good. I use whatever recycled metal I find at work or yard sales. Some sheet aluminum for fit and finish.
I never try to cheat myself on resistors or caps. Too many cheap caps in the computer industry. They blow up people. Resistors are an inexpensive upgrade typically, and I only go above dale in the signal path.
Well, the proto builders get to weigh in and we can make a decision. I think gold has a different point of overheating. A lot of builders dont like gold. We could go thicker as well, but for a pre I think it's overkill. It's a sensitive area on the LEDs.
With my iron at 900F on a gold pad, I can burn it off
So this is a good construction note concern anyways. This area is likely a place to be careful.
My opinion is low-melting point lead solder is a way to go here. I have used so many crap solders. I love cardas. Worth the money.
Th board price increase by a $ is not going to hurt the current GB lot as I see lot more price on power supply PCB alone while this one is much more than PSU board. But this issue is not very critical and can be managed by direct soldering leads. I use nice temp controled pencil iron which does not heat the pad that much . But still found one or two pads coming off in the first time soldering. I ignored it thinking that I might have done something wrong but I remember now as Alan reported. Is it related to tinning or base copper adhesion to board ?
Kannan
The problem might be the traces and pads are too thin. I haven't seen the layout or the boards, so I don't know. I know this, that most pcb layout software has very thin traces and pads by default, which are suitable for board machine stuffing but not for regular diy. I usually set the traces much wider than the default, and use bigger pads. It's an easy solution which should not bring the cost of boards up.
Just my 2c.
Just my 2c.
Tea-Bag:
Watch the 4X220R mainly, the 2X220k, the pot and the electrolytics. All other stuff do not have to do match with transparency and tone.
Grufti's is a decent one, with a good pot but not expensive. So its a good indication of some logical but good grade Mezmerize cost. Then he was very thorough in to telling us his subjective evaluation, seems not verging to some custom tonal preference if I got it right. That is why I ask. Indicative cost for an indicative build.
Watch the 4X220R mainly, the 2X220k, the pot and the electrolytics. All other stuff do not have to do match with transparency and tone.
Grufti's is a decent one, with a good pot but not expensive. So its a good indication of some logical but good grade Mezmerize cost. Then he was very thorough in to telling us his subjective evaluation, seems not verging to some custom tonal preference if I got it right. That is why I ask. Indicative cost for an indicative build.
I only mentioned cases to indicate that the pcb is peanuts compared to the rest - I also spend more on the components and stick the pcb's on a nice block of wood!
I am using silver solder - I will try Cardas if it makes a difference.
Alan
You are right Tea-Bag, there are advantages and disadvantages to each of those choices. I actually don't have a really strong opinion on this matter at all from personal preference.
"Robust" and "reasonably easy to work with" seem like two good ideas for any diy project. I don't claim to know where these two best meet for these two pcb's.
Unchanged [cheaper, good enough] - gold [more expensive, sometimes no fun to work with] - thicker [more expensive, more robust, better heat dispersion, electrically overkill] ... we can probably live with all of them.
"Robust" and "reasonably easy to work with" seem like two good ideas for any diy project. I don't claim to know where these two best meet for these two pcb's.
Unchanged [cheaper, good enough] - gold [more expensive, sometimes no fun to work with] - thicker [more expensive, more robust, better heat dispersion, electrically overkill] ... we can probably live with all of them.
Well, the proto builders get to weigh in and we can make a decision. I think gold has a different point of overheating. A lot of builders dont like gold. We could go thicker as well, but for a pre I think it's overkill. It's a sensitive area on the LEDs.
With my iron at 900F on a gold pad, I can burn it off
So this is a good construction note concern anyways. This area is likely a place to be careful.
My opinion is low-melting point lead solder is a way to go here. I have used so many crap solders. I love cardas. Worth the money.
As I hold and I am checking out the two PCBs right now, I see that the Hypno has less exposed pad in the 3 Led position than the the Mez as well as the later being gold. My guess it is its down to the area. Hypno's like rings, when Mez's have a little more.
But those are small boards to keep it neat for signal path and noise. Use a pencil 15-20W and low melting point good solder for all soldering. If you know it, you will avoid it. That is why proto testing rules.
You are right - i also feel it is not the gold or tin but the pad area causing this. If less area is exposed like the Hypno board - probably we can avoid this
kannan
I'll gladly add up my parts cost sometime this week. My build cost very little considering how good it is, I can tell you that without going through any mouser invoices. No expensive caps, that's for sure, but connectors add up of course with seven pairs for the Mezmerize. On the other hand, if you only need 4 inputs, you don't really have to install all 6.
Oh, and another vote for Cardas eutectic solder.
Oh, and another vote for Cardas eutectic solder.
Suggestion for final package: instructions...
I would like to suggest that the final package for the boards come with a sheet (maybe we can collectively put that together here) with some of the gotchas and recommended practices. The thread is long enough now that it's a real pain finding (and not overlooking!) important information.
Here are some suggestions:
- issues with potential shuttering problems: list of suggested solutions
- how to measure IDSS for the sk170 to put "the right one" in the "right spot"
- how to measure the LEDs (and mixing is allowed, for example) to get the right total
What others along these lines do we have?
My experience with the revC GB was that lots of questions like this came later because we didn't put together a sheet like that.
Thoughts?
peter
I would like to suggest that the final package for the boards come with a sheet (maybe we can collectively put that together here) with some of the gotchas and recommended practices. The thread is long enough now that it's a real pain finding (and not overlooking!) important information.
Here are some suggestions:
- issues with potential shuttering problems: list of suggested solutions
- how to measure IDSS for the sk170 to put "the right one" in the "right spot"
- how to measure the LEDs (and mixing is allowed, for example) to get the right total
What others along these lines do we have?
My experience with the revC GB was that lots of questions like this came later because we didn't put together a sheet like that.
Thoughts?
peter
In my phono thread a member compiled scattered info in a pdf, and then the repetitive building specific questions cut down dramatically. So its a good idea. A combined pdf with BOM and some building remarks would be swell. I agree.
I have the Hypno board. Soldered a lot in the past couple days. Waiting for a few more parts that I thought I had but dont.
I thought that some of the resistors pads were super small. That coupled with the fact that the holes were very tight for the leads means that the solder did not want to wick away from the iron even with a decent amount of flux. I felt that I had the iron on the resistor leads for to long and especially on the TO-92 leads for to long for this reason. I always prefer the oblong pads that are thin and narrow so I can heat the pad, then roll the iron over to the lead and apply solder quickly rather than heating pad and lead at the same time.
I had no pads remove even though I did remove and replace 4 resistors and one capacitor. A good solder wick and a hot iron help here.
I prefer plated through holes. I believe they can be ordered without ordering double sided and are nice because they allow for the solder to wick up the lead and into the hole so we have a large contact area for the lead.
Uriah
I thought that some of the resistors pads were super small. That coupled with the fact that the holes were very tight for the leads means that the solder did not want to wick away from the iron even with a decent amount of flux. I felt that I had the iron on the resistor leads for to long and especially on the TO-92 leads for to long for this reason. I always prefer the oblong pads that are thin and narrow so I can heat the pad, then roll the iron over to the lead and apply solder quickly rather than heating pad and lead at the same time.
I had no pads remove even though I did remove and replace 4 resistors and one capacitor. A good solder wick and a hot iron help here.
I prefer plated through holes. I believe they can be ordered without ordering double sided and are nice because they allow for the solder to wick up the lead and into the hole so we have a large contact area for the lead.
Uriah
I would agree that it is an area issue and not material. I found the pads to be very small all around. I use a pretty small tip on my iron and some of the pads were hard to get contact with. I had to remove and replace my matched j-fets couple times. I didn't have any issues with losing pads. But I changed some LED's and had a pad lift after one change. I think it is also a matter of a single sided board that doesn't have the via (thru hole) to hold things together better. If the pads were made bigger, I think it would be nice and easier to work with. If they stay as they are, there shouldn't be any really big issues, just that things are a bit more temperamental.
P.S I have the mesmerize board
P.S I have the mesmerize board
on that note, would it be worth 'advertising' this to the rest of the board as a shunt gb also? perhaps there are many overlooking this thread because they don't want a direct-coupled b1 but would love to have a preamp-worthy shunt or ten. I'd imagine #'s for shunt gb would be huge
And Tea-Bag would become the supersonic postman. Better no.
I thought of that only after i posted. I certainly don't want more work for him
Just comming off several SMD boards I did not have any issues soldering.I have a Mezmerize, I can see folks having an issue though. I use a temp controlled iron and Alpha .25 63/37 solder. I would say the the costs being so low for this build that the board price will not be an issue, however I would not want to have to make any changes that would cause a delay or hassle for those respnsable. With care this is an easy build, at least I found it to be.
I think at this stage, the current boards are workable, albiet not perfect. So adding gold might help a bit here. I think changing the layout is a choice more for CRT to consider. He has done much work here already. We have to re-tool at this point anyways for better lettering. So cost's their will not change.
I have worked on worse boards than this.
If a retool would be done, it might be to make the hyno a little more robust, but the Mezmerize had a problem with Dave too.
My opinion is we march forward with what we have.
People make mistakes soldering, and occasionally have to do their own mods to get the circuit right.
I say we march forward with the current GB, and consider a 2 layer setup, industrial version next year.
I have worked on worse boards than this.
If a retool would be done, it might be to make the hyno a little more robust, but the Mezmerize had a problem with Dave too.
My opinion is we march forward with what we have.
People make mistakes soldering, and occasionally have to do their own mods to get the circuit right.
I say we march forward with the current GB, and consider a 2 layer setup, industrial version next year.