SMT Vs. Thru-hole? (signal quality)

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Kent, luketeaford, Joe.

User avatar
Laughing
Common Wiggler
Posts: 140
Joined: Sun May 29, 2016 3:43 pm
Location: Calialialialifornia

SMT Vs. Thru-hole? (signal quality)

Post by Laughing » Tue Jun 19, 2018 1:44 pm

Is there some debate as to why one is better than the other in terms of signal quality or something? Is it some audiophile nonsense? If it is, I'd like to hear the nonsense so I can make my own opinion on whether it's nonsensical or not.

So far, my experience with SMT has been actually considerably better than thru-hole because of all the extra stuff in assembly that you no longer have to do. It's part of the reason that my re-layout of the MFOS EchoFXXXX has mostly surface mount components on it.

I understand one bit is that film caps don't really exist in SMT, so that's one consideration.

User avatar
pre55ure
Super Deluxe Wiggler
Posts: 1438
Joined: Tue May 01, 2012 7:15 pm
Location: SoCal
Contact:

Post by pre55ure » Tue Jun 19, 2018 2:37 pm

The consensus from people who actually know things, is that for the most part there is no difference. Most currently constructed IC's transistors etc... are only built with one die and so SMD and TH parts are literally the exact same part inside. Only the casing differs.
You are correct about the film caps, though I have noticed that COG ceramic caps are getting used alot in places that years ago would have been film. This is a trend that has been happening in TH builds as well, so not exclusive to SMT.

I recently have started converting all the stuff I do to SMT, and have done LOTS and LOTS of listening tests as well as a lot of other testing and have been able to find no differences (in sound or any other tests) between the basic resistors, ceramic caps, and jellybean transistors.

That being said - I still use a fair amount of TH parts, simply because most of the stuff I do requires a fair amount of parts that are not available in SM packages.

User avatar
StillNotWorking
Veteran Wiggler
Posts: 654
Joined: Thu Aug 22, 2013 3:29 pm
Contact:

Post by StillNotWorking » Tue Jun 19, 2018 4:10 pm

PCB layout with short signal path is what can benefit most from SMD design, with theoretical lower distortion and noise. But we are here to make noise so...

User avatar
gbiz
Super Deluxe Wiggler
Posts: 1076
Joined: Tue May 07, 2013 11:20 am
Location: Sussex, UK

Post by gbiz » Tue Jun 19, 2018 5:17 pm

I built a couple of System 100 filter boards up in SMD recently, one with C0G & one with PPS film caps (Panasonic ECHU) in the audio path, mostly in the ladder. Side by side the one with the PPS caps sounded smoother, less buzzy. I removed the C0G caps & fitted PPS to the other one. I'll be using them again in my next build.

I can't recall seeing them being used any other SDIY module. I don't know if there's obvious reasons why they don't get used in more designs, though TBH i've not looked into it. They cost slightly less than the equivalent C0G for values greater than 10n. The footprint isn't much larger.

User avatar
mskala
Super Deluxe Wiggler
Posts: 2731
Joined: Thu Mar 27, 2014 7:33 am
Location: Toronto
Contact:

Post by mskala » Tue Jun 19, 2018 6:14 pm

Capacitor dielectric is a much different question from surface-mount versus through-hole. I myself like through-hole for ease of construction and repair, but not everyone finds it easier.
North Coast Synthesis Ltd.
Latest Web log posting: Passive multiples and friends

If you tell me that your goal is systemic change toward radical acceptance, and I see that you treat those you perceive as lesser-than with the same kind of scorn and derision that pushed me toward this insular little subculture where I feel comfortable [. . .] then you’ve successfully convinced me that your acceptance is not radical and the change you want not systemic. - Meredith Patterson

User avatar
gbiz
Super Deluxe Wiggler
Posts: 1076
Joined: Tue May 07, 2013 11:20 am
Location: Sussex, UK

Post by gbiz » Tue Jun 19, 2018 6:33 pm

mskala wrote:Capacitor dielectric is a much different question from surface-mount versus through-hole.
The OP suggested film caps don't really exist in SMT, so that's one consideration.. I was just sharing my experience that they do exist & IMO sounded better in that application. I fail to see how that's offtopic ;)

User avatar
emmaker
Ultra Wiggler
Posts: 769
Joined: Sat Mar 10, 2012 5:07 pm
Location: PDX

Post by emmaker » Tue Jun 19, 2018 7:07 pm

gbiz wrote:I built a couple of System 100 filter boards up in SMD recently, one with C0G & one with PPS film caps (Panasonic ECHU) in the audio path, mostly in the ladder. Side by side the one with the PPS caps sounded smoother, less buzzy. I removed the C0G caps & fitted PPS to the other one. I'll be using them again in my next build.

I can't recall seeing them being used any other SDIY module. I don't know if there's obvious reasons why they don't get used in more designs, though TBH i've not looked into it. They cost slightly less than the equivalent C0G for values greater than 10n. The footprint isn't much larger.
One issue with the PPS caps is that they are very fragile and hard to solder by hand. You can melt them pretty easily if you are not careful. If you are soldering with real reflow, reflow oven or hot plate then it gets more doable.

User avatar
NV
Super Deluxe Wiggler
Posts: 2578
Joined: Tue Jul 08, 2008 7:40 pm

Post by NV » Tue Jun 19, 2018 8:23 pm

There is some difference in components relative to size. You'll have an easier time finding higher wattage ratings in larger resistors for example. At a point it's better to go TH for mechanical reasons, rather than SMD jacks/headers and praying they never encounter any shearing forces.

In terms of signal quality however, a large part of any typical synth circuit's character will come from active components like transistors and ICs, and in DIP and SMD variants they're going to be using the exact same die. Any difference you hear between a TH and SMD circuit will come from choices in design and any number of other factors before it comes from something like the difference in stray capacitance on leads.

From a design and manufacturing standpoint, SMD has massive benefits over TH. Tighter circuits with improved routing options is a boon for layout. Part sourcing and assembly are both significantly easier and cheaper, which means potentially more budget to put towards things that would make a quantifiable difference. That could be better component specs or just more time and funds towards R&D, which (should) lead to improved design.

User avatar
Laughing
Common Wiggler
Posts: 140
Joined: Sun May 29, 2016 3:43 pm
Location: Calialialialifornia

Post by Laughing » Tue Jun 19, 2018 8:44 pm

SMD Film caps sound interesting! A pity they can't get larger capacities in smaller surface mount packages, but that might be worth a shot in the future. I've had no troubling soldering polystyrene caps so far with the warning that they'll just melt if you heat them too long. And yes, I plan not to do mechanical components with SMT, and probably not IC's quite yet.

flts
Super Deluxe Wiggler
Posts: 2989
Joined: Tue Mar 04, 2008 7:01 am

Post by flts » Wed Jun 20, 2018 1:54 am

I'm just reiterating some of the stuff already said here, but I think the main thing that people normally completely forget (not in this discussion but a lot of "audiophile" level DIY debates) is, that we aren't talking about magic or some completely different way of manufacturing components from ground up. We are simply comparing two different kinds of component packages:

1) those that have leads that go through PCB and

2) those that have leads that are soldered on the same side as the component.

In most cases, there needs to be no other difference between a component in SMT and in THT, than that one has longer round leads and the other shorter flat ones, and both are enclosed in plastic / metal cases of different sizes.

The puzzling point is that someone claiming a TO92 part would sound radically different from a PDIP part of same vintage and manufacturer is usually treated with skepticism, whereas someone claiming exact same part in PDIP vs SOIC package sounds very different is sometimes actually taken seriously, at worst with nonsensical replies like (exaggerating a bit but only a bit) "yeah, SMT is that digital stuff, lifeless and mass manufactured, the through hole equivalent has more mojo".

The main counter-points to consider regarding perceived differences (again reiterating what others have said) would be, that

1) Some particular type or brand of component may not be available with both mounting styles, which could often contribute towards any larger differences - a bit like if you substituted a through hole component with another completely different through hole component with completely different specifications.

2) The mounting styles allow for different kinds of layouts, which may affect performance one way or another.

In the first case, I don't think we're talking about "difference between through-hole and SMT" per se - rather difference between using completely different components, some of which are simply not available as SMT equivalents. If you eg. substitute vactrols with something completely different, or poorly specified vintage TH capacitors with modern tightly specified C0G SMT ceramics, then the difference isn't about "SMT vs THT" anymore, but differences between designs that use different types of parts altogether.

In the second case, it makes sense in some critical parts of some circuits that the different trace lengths and layouts indeed make an audible difference. I assume this becomes a bigger contributing factor with high speed digital circuits, but might affect many analog circuits as well. Here's where my knowledge ends, as I'm not an actual EE and only know a bit about the digital design side which isn't that relevant here.

Finally, the mechanical difference is of course a big thing which people have opinions on - ease of repair, sturdiness of mounting panel components on PCB and so on. However, that's a separate point from signal quality.

toneburst
Ultra Wiggler
Posts: 893
Joined: Fri May 02, 2014 2:41 pm
Location: London

Post by toneburst » Wed Jun 20, 2018 3:21 am

I hate soldering SMT electrolytic caps.

That's my contribution to this topic.

User avatar
gbiz
Super Deluxe Wiggler
Posts: 1076
Joined: Tue May 07, 2013 11:20 am
Location: Sussex, UK

Post by gbiz » Wed Jun 20, 2018 5:23 am

emmaker wrote:
gbiz wrote:I built a couple of System 100 filter boards up in SMD recently, one with C0G & one with PPS film caps (Panasonic ECHU) in the audio path, mostly in the ladder. Side by side the one with the PPS caps sounded smoother, less buzzy. I removed the C0G caps & fitted PPS to the other one. I'll be using them again in my next build.

I can't recall seeing them being used any other SDIY module. I don't know if there's obvious reasons why they don't get used in more designs, though TBH i've not looked into it. They cost slightly less than the equivalent C0G for values greater than 10n. The footprint isn't much larger.
One issue with the PPS caps is that they are very fragile and hard to solder by hand. You can melt them pretty easily if you are not careful. If you are soldering with real reflow, reflow oven or hot plate then it gets more doable.
Yeah, i'd noticed the Panasonic product doc states "reflow only", & visually they do look fragile. I was fully expecting them to melt at the first touch of the iron tip. But they actually hand soldered quite easily, needing minimal heating to get the solder to fully flow across the joint. For an experiment i removed a few with a hot air rework station & reused them, with no noticeable damage. (I'd half expected them to melt). I reckon they'd be OK for a competent builder to hand solder provided they're careful. And at least with the larger package size, they're easy to handle.

One thing i won't be doing again with them. I've got into the habit of pushing the device that's in the tweezers down onto the board with the tip of the iron. A bad move with these. As i discovered, the top surface melts quite easily :hihi:

User avatar
EATyourGUITAR
has no life
Posts: 6079
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Post by EATyourGUITAR » Wed Jun 20, 2018 6:10 am

some guitar pedals that were originally manufactured with through hole carbon comp resistors established a specific signature sound that creates a demand many years later for guitar pedals with that exact PCB layout using those exact resistors. metal film removes a lot of thermal noise associated with resistors and so to a guitar player the higher quality component is shit. I know a tube amp builder who only uses new old stock mil spec (made in the USA) carbon comp resistors. then he grades them by hand. I'm not sure if he is bullshitting me with marketing hype or maybe because in the NAVY they only do things one way. this was the same guy who argued that guitar players need NASA spec soldering joints on everything. you have to draw the line somewhere on what is real and what is hype but I think there is a sliding scale based on your own personal standards and your time/money budget.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

User avatar
StillNotWorking
Veteran Wiggler
Posts: 654
Joined: Thu Aug 22, 2013 3:29 pm
Contact:

Post by StillNotWorking » Wed Jun 20, 2018 7:18 am

Some tube amp builder do shit you with all sorts of marketing hypes — as many boutique pedal builders do. They make their living similar to most of the hi-end hi-fi marked where the discussions can come down to which material sounds best when used as floor standoff for the speaker cables.

I believe there are web pages around showing some insight to carbon resistors so called magic to tone only apply to high voltage circuits. Meaning they're are wasted in stomp boxes and most of the circuit inside a tube amp.

"then he grades them by hand" — genius of course take in account those old components, caps, tubes and even mains power will drift out of his predefined speck and make his fragile dream castle dispear in the wind.

Not saying its a bad idea to sort and test your typical 20% off the shelf component values. It's been done for transistors and op-amps for decades in synth design. But then you never seen Roland marketing it as something special. It's just part of the design to ensure functionality.

flts
Super Deluxe Wiggler
Posts: 2989
Joined: Tue Mar 04, 2008 7:01 am

Post by flts » Wed Jun 20, 2018 7:34 am

Just a note, that as I kind of suggested in my post, now we're out of "SMT vs THT" discussion and into "using rare old mojo parts vs using modern easily available parts" discussion.

The simple fact is that there are no rare tropical fish caps or vintage mil spec carbon comp resistors available from the store in SMT, so it's kind of, well... not an useless discussion by any means, but entirely another thing than simply comparing different component packages. You could as well say "through hole sounds different from through hole".

Side note: I'm still not sure how many guitar players have done something like proper ABX tests for pedals built with eg. different types of resistors matched to roughly same values. Even the thought of controlled double blind usually seems to bring bad vibes and much disdain among musicians and vintage gear hobbyists. The power of psychology is a wild thing - if one knows something has Cool Vintage Mojo parts inside and it matters to him/her, one tends to think it sounds better regardless of level of musicianship or ear training.

But that's neither here nor there, I suppose that wasn't the point of the discussion - unless to remind everyone that "I think x sounds better than y" doesn't mean much yet without knowing about testing methodology, and talking about possible differences between same types of components in different packages is probably the "safer" way to go.

User avatar
fuzzbass
Super Deluxe Wiggler
Posts: 2089
Joined: Thu Feb 06, 2014 6:08 pm
Location: Florida, USA

Post by fuzzbass » Wed Jun 20, 2018 8:25 am

toneburst wrote:I hate soldering SMT electrolytic caps.

That's my contribution to this topic.
Me too. I have been hating it a lot lately and I am getting better at it. Also, tropical fish capacitors sound warmer because they come from the tropics.
Wired for weird

flts
Super Deluxe Wiggler
Posts: 2989
Joined: Tue Mar 04, 2008 7:01 am

Post by flts » Wed Jun 20, 2018 10:48 am

Rest assured, the hate for manually soldering SMT electrolytics is well shared among the connoiseurs in the community.

User avatar
emmaker
Ultra Wiggler
Posts: 769
Joined: Sat Mar 10, 2012 5:07 pm
Location: PDX

Post by emmaker » Wed Jun 20, 2018 11:32 am

gbiz wrote:
emmaker wrote:
gbiz wrote:I built a couple of System 100 filter boards up in SMD recently, one with C0G & one with PPS film caps (Panasonic ECHU) in the audio path, mostly in the ladder. Side by side the one with the PPS caps sounded smoother, less buzzy. I removed the C0G caps & fitted PPS to the other one. I'll be using them again in my next build.

I can't recall seeing them being used any other SDIY module. I don't know if there's obvious reasons why they don't get used in more designs, though TBH i've not looked into it. They cost slightly less than the equivalent C0G for values greater than 10n. The footprint isn't much larger.
One issue with the PPS caps is that they are very fragile and hard to solder by hand. You can melt them pretty easily if you are not careful. If you are soldering with real reflow, reflow oven or hot plate then it gets more doable.
Yeah, i'd noticed the Panasonic product doc states "reflow only", & visually they do look fragile. I was fully expecting them to melt at the first touch of the iron tip. But they actually hand soldered quite easily, needing minimal heating to get the solder to fully flow across the joint. For an experiment i removed a few with a hot air rework station & reused them, with no noticeable damage. (I'd half expected them to melt). I reckon they'd be OK for a competent builder to hand solder provided they're careful. And at least with the larger package size, they're easy to handle.

One thing i won't be doing again with them. I've got into the habit of pushing the device that's in the tweezers down onto the board with the tip of the iron. A bad move with these. As i discovered, the top surface melts quite easily :hihi:
My first experience with these was a long time ago before I had developed good SMT soldering skills. We used them in the filters for audio ADCs and DACs. I went through a few before I got them down on a board. Fortunately my skills have gotten better over the year.

User avatar
EATyourGUITAR
has no life
Posts: 6079
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Post by EATyourGUITAR » Wed Jun 20, 2018 5:02 pm

flts wrote:Just a note, that as I kind of suggested in my post, now we're out of "SMT vs THT" discussion and into "using rare old mojo parts vs using modern easily available parts" discussion.
actually my comments assume that the reader has at least the basic knowledge that most people are buying metal film when they buy SMD resistors. guitar players tend to avoid metal film SMD components for two reasons. I gave both reasons since it is not ONLY a question of SMD metal film vs carbon comp through hole but also a discussion of SMD vs through hole since that is a big determining factor in the difference in sound between SMD metal film and carbon comp through hole. maybe I should have clarified that point down to the explicit detail. I will try harder next time.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

User avatar
spotta
Needs more wigglin' time
Posts: 933
Joined: Thu Jul 28, 2011 7:57 am
Location: Essex. UK

Post by spotta » Wed Jun 20, 2018 5:05 pm

flts wrote:Rest assured, the hate for manually soldering SMT electrolytics is well shared among the connoiseurs in the community.
Someone should come up with a dual footprint for SDIY designs that has two pads for SMT that both feature holes so you could fit a TH where height allows.
Spotta
--

User avatar
pre55ure
Super Deluxe Wiggler
Posts: 1438
Joined: Tue May 01, 2012 7:15 pm
Location: SoCal
Contact:

Post by pre55ure » Wed Jun 20, 2018 5:55 pm

Someone should come up with a dual footprint for SDIY designs that has two pads for SMT that both feature holes so you could fit a TH where height allows.
Edit: (oops.. Didn't see that you were referring to the electrolytics. That should be easy to do as well.)

Like this? :hihi:

Mocked this up for doing my own prototypes, but I still haven't actually tried it out yet. It's a combo for 0805 resistors or caps or vertically mounted resistors.

Image

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 7307
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Wed Jun 20, 2018 7:11 pm

you guys talking about guitar pedals are full of shit. everyone knows that the only real pedals have jrc4558 opamps. cuz mojo. cuz engineering.

are surface mounted parts good for metel? no! ... and everyone knows that. only resistors and 4558 opamps can make teh brutals. cuz badd ass mojo!!!

stupid asses.

Image
5U MODULAR NORMALIZING PROJECT (for your entertainment) viewtopic.php?t=78836&highlight=
.. as of Dec 8th 2020 on a break for a bit .. contact me via bamco60@hotmail.com if needed.
WELCOME TO 2021 .. THE BEATINGS WILL CONTINUE UNTIL MORALE IMPROVES.

User avatar
Grumble
Super Deluxe Wiggler
Posts: 1813
Joined: Thu Dec 17, 2015 5:24 am
Location: U aint much if U aint Dutch
Contact:

Post by Grumble » Thu Jun 21, 2018 1:16 am

Glad someone cleared that up :nana:

ubiquiphilia
Common Wiggler
Posts: 184
Joined: Mon Apr 14, 2014 5:41 am
Location: Tasmania

Post by ubiquiphilia » Thu Jun 21, 2018 1:54 am

spotta wrote:
flts wrote:Rest assured, the hate for manually soldering SMT electrolytics is well shared among the connoiseurs in the community.
Someone should come up with a dual footprint for SDIY designs that has two pads for SMT that both feature holes so you could fit a TH where height allows.
been pleasantly surprised lately to find the standard footprint for TH electros in the low uF range; 10uF, 22uF et al fits a tantalum cap nicely across the pads.

flts
Super Deluxe Wiggler
Posts: 2989
Joined: Tue Mar 04, 2008 7:01 am

Post by flts » Thu Jun 21, 2018 2:05 am

EATyourGUITAR wrote:actually my comments assume that the reader has at least the basic knowledge that most people are buying metal film when they buy SMD resistors. guitar players tend to avoid metal film SMD components for two reasons. I gave both reasons since it is not ONLY a question of SMD metal film vs carbon comp through hole but also a discussion of SMD vs through hole since that is a big determining factor in the difference in sound between SMD metal film and carbon comp through hole. maybe I should have clarified that point down to the explicit detail. I will try harder next time.
I understood you and it is a valid point, sorry if I put my reply too bluntly.

I'll try to formulate my point in a bit different way: we really haven't defined WHAT it is we are exactly comparing - and that kind of affects which points / comparisons are relevant here.

Are we talking about

- How the exact same circuit will perform when using SMT vs THT parts with as close to identical specifications as possible? In that case it doesn't make any sense to compare eg. carbon comp vs metal film resistors, as metal film is available in both SMT and THT and using that will get rid of one extraneous variable in the comparison of different package types.

- The availability of different components in SMT vs THT, and hence the amount of choice one can make with component selections without mixing package types? In that case yours a relevant point to make - some specific components are not available either in through hole or in SMT, and someone may wish to use a specific component for a specific task (eg. a modern, powerful 32-bit microcontroller for a digital module, or a particular vintage cap / resistor for a particular analog circuit) because of how it performs.

There are probably other sensible angles to look at this from, but those are the most common ones I see. It may sound like silly semantics, but I think that's actually an important distinction to make for the discussion to be fruitful.

TL;DR: are we talking about how the package and mounting type affect the sound of a particular component, what kinds of components one can use while limiting oneself to just TH or SMT and how that may affect the sound of the end result, or something else?
Last edited by flts on Thu Jun 21, 2018 2:11 am, edited 1 time in total.

Post Reply

Return to “Music Tech DIY”