Deleted
Deleted Member
Posts: 0
|
Post by Deleted on May 24, 2017 10:46:09 GMT -6
So, the esteemed drbill posted this in the Stam thread. "They are probably the same folks who decry surface mount components." And it made me wonder, what, if any, effect is perceived by the presence of surface mount components in a pro audio circuit? I get it that a lot of folks, especially some purists online, prefer through-hole components because they're easier to service and mod, but in general why is there a bias against surface mount? I've read how some people think the Vintech stuff with surface mount sounds less "vibey", and I'm admittedly guilty of thinking along those lines, but let me throw it out to the experts: If implemented properly, surface mount components are sonically just as good, right?
|
|
|
Post by Martin John Butler on May 24, 2017 10:58:59 GMT -6
I'm curious about this too. Bad Cat amps has a new amp that instead of point to point has I think, surface mounted circuit boards. They claim it's more consistent and sounds as good as the hand wired amps.
|
|
ericn
Temp
Balance Engineer
Posts: 14,919
|
Post by ericn on May 24, 2017 11:06:20 GMT -6
This is one of those it's more about how it's implemented than blanket Bad or good!
|
|
|
Post by jazznoise on May 24, 2017 11:07:50 GMT -6
The components are obviously fine. The layout on the PCB is more likely to have an effect, if any.
|
|
|
Post by EmRR on May 24, 2017 11:12:16 GMT -6
Many are technically better, some are worse. Have to know your ingredients. Physical effects can differ, thermal effects, layout relationships, etc.
|
|
|
Post by swurveman on May 24, 2017 11:18:26 GMT -6
So, the esteemed drbill posted this in the Stam thread. "They are probably the same folks who decry surface mount components." And it made me wonder, what, if any, effect is perceived by the presence of surface mount components in a pro audio circuit? I get it that a lot of folks, especially some purists online, prefer through-hole components because they're easier to service and mod, but in general why is there a bias against surface mount? I've read how some people think the Vintech stuff with surface mount sounds less "vibey", and I'm admittedly guilty of thinking along those lines, but let me throw it out to the experts: If implemented properly, surface mount components are sonically just as good, right? If you're talking about preamps, I don't know why you singled out Vintech, but the original 1073 included a Printed Circuit Board.
|
|
|
Post by swurveman on May 24, 2017 11:26:36 GMT -6
The components are obviously fine. The layout on the PCB is more likely to have an effect, if any. This is interesting, because Geoff Tanner claims there's a flaw in the design of the Neve 1073 PCB, which when the EQ is engaged results in a drop of -4 dB at 20Hz. I wonder if Rupert's 1073 was released today whether people would criticize it for its faulty design!
|
|
|
Post by sll on May 24, 2017 12:03:31 GMT -6
The potential downside to SMD is heat dissipation. They are not good for power devices unless dealt with accordingly with proper heat sinks and layout. SMD can be better than through hole in terms of less parasitic capacitance and inductance, shorter paths for the connections and the overall circuit.
Like anything, it comes down to the proper design and layout.
|
|
|
Post by svart on May 24, 2017 12:06:26 GMT -6
So, the esteemed drbill posted this in the Stam thread. "They are probably the same folks who decry surface mount components." And it made me wonder, what, if any, effect is perceived by the presence of surface mount components in a pro audio circuit? I get it that a lot of folks, especially some purists online, prefer through-hole components because they're easier to service and mod, but in general why is there a bias against surface mount? I've read how some people think the Vintech stuff with surface mount sounds less "vibey", and I'm admittedly guilty of thinking along those lines, but let me throw it out to the experts: If implemented properly, surface mount components are sonically just as good, right? You're going to get quite the gamut of opinions on this.. But as a hardware design engineer who also dabbles in audio work, I feel I have a unique perspective on both. The short answer is that it doesn't matter which one you use, as long as you understand why you're using the parts you are using. Most active devices like transistors or opamps are the same exact part inside the plastic package and that at audio frequencies, the lead lengths don't matter by themselves. It boils down to something I've mention before here.. Stability. Active parts have a wide range of operational bandwidth, and audio frequencies rarely inhabit more than a fraction of the bandwidth of the parts. Because of this, the parts need to be stabilized by feedback or other gain/bandwidth limiting so that you don't have issues with stability. One thing that can happen is that the parts are operated at much wider bandwidths than needed under the belief that wider bandwidth will improve fundamental audio signal performance. In most cases the operational bandwidth is much, much greater than the needed bandwidth with the side effect of being sensitive to small effects of trace/lead inductance and capacitance forming "tanks" or frequencies where the parasitic effects can resonate with the stimulation of the gain/bandwidth of the active part. This results in oscillation as the part starts to feed power into the tank, and the tank then feeds it back to the part. This is, of course, the simplified version, but the effect is well known by most folks. So, in reality, shorter leads on SMD parts, coupled with lack of leads on caps/resistors, etc, generally allow circuits to operate at higher frequencies without the parasitic effects of the legs of the parts. I guess you can say that SMD parts are going to be more stable for the same gain/bandwidth. Audio wise, unless you happen upon an effect of the parasitic components of the circuit actually helping you out, then there is no audio difference or benefit with using leaded parts vs. SMD.
|
|
|
Post by Martin John Butler on May 24, 2017 12:12:34 GMT -6
Cool posts guys.
|
|
ericn
Temp
Balance Engineer
Posts: 14,919
|
Post by ericn on May 24, 2017 12:16:08 GMT -6
So, the esteemed drbill posted this in the Stam thread. "They are probably the same folks who decry surface mount components." And it made me wonder, what, if any, effect is perceived by the presence of surface mount components in a pro audio circuit? I get it that a lot of folks, especially some purists online, prefer through-hole components because they're easier to service and mod, but in general why is there a bias against surface mount? I've read how some people think the Vintech stuff with surface mount sounds less "vibey", and I'm admittedly guilty of thinking along those lines, but let me throw it out to the experts: If implemented properly, surface mount components are sonically just as good, right? You're going to get quite the gamut of opinions on this.. But as a hardware design engineer who also dabbles in audio work, I feel I have a unique perspective on both. The short answer is that it doesn't matter which one you use, as long as you understand why you're using the parts you are using. Most active devices like transistors or opamps are the same exact part inside the plastic package and that at audio frequencies, the lead lengths don't matter by themselves. It boils down to something I've mention before here.. Stability. Active parts have a wide range of operational bandwidth, and audio frequencies rarely inhabit more than a fraction of the bandwidth of the parts. Because of this, the parts need to be stabilized by feedback or other gain/bandwidth limiting so that you don't have issues with stability. One thing that can happen is that the parts are operated at much wider bandwidths than needed under the belief that wider bandwidth will improve fundamental audio signal performance. In most cases the operational bandwidth is much, much greater than the needed bandwidth with the side effect of being sensitive to small effects of trace/lead inductance and capacitance forming "tanks" or frequencies where the parasitic effects can resonate with the stimulation of the gain/bandwidth of the active part. This results in oscillation as the part starts to feed power into the tank, and the tank then feeds it back to the part. This is, of course, the simplified version, but the effect is well known by most folks. So, in reality, shorter leads on SMD parts, coupled with lack of leads on caps/resistors, etc, generally allow circuits to operate at higher frequencies without the parasitic effects of the legs of the parts. I guess you can say that SMD parts are going to be more stable for the same gain/bandwidth. Audio wise, unless you happen upon an effect of the parasitic components of the circuit actually helping you out, then there is no audio difference or benefit with using leaded parts vs. SMD. Great post Chris
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on May 24, 2017 12:35:41 GMT -6
swurveman: A couple years back I was researching online about the differences between the Vintech X73 vs X73i. The X73i allegedly being considerably cheaper due partly to the automated build process. My takeaway was that the experiences people described with both left the surface-mount built X73i as sonically less desirable. svart: Thank you for your reply. You are exactly the type of expert I was hoping would lend some input.
|
|
|
Post by drbill on May 24, 2017 13:23:57 GMT -6
Implementation. Parts choice. It's what separates the men from the boys so to speak. (No offense to the boys, I'm one of em.... ) One thing that comes to mind - for the DIY style designers especially, it's much easier to implement thru hole components than SMD components. Another important point is size, and the fact that some thru hole components are becoming more and more difficult to source as the tech world moves into smaller and smaller implementation of electronics. Like computers in our wrist watch?
|
|
|
Post by BradM on May 24, 2017 13:27:57 GMT -6
I'm just going to jump in and second everything that sll , EmRR , and svart said. Since these are audio circuits we are talking about, at the end of the day you just have to listen to what you've built regardless of whether the parts are big or small. Just think of all the shitty sounding gear Behringer made with through hole parts. Then think of all the awesome sounding stuff Rupert Neve makes with SMT stuff. I would like to add the following. I find modifying / repairing gear with SMT parts a bit easier to do than through hole. You don't have to disassemble much except for a top cover in most cases. With a hot iron and a blob of solder you can flick off the old part. I've definitely messed up more traces removing large through hole components. Companies that cling to their through hole designs in the long-run are going to have a hard time finding parts. Eventually we all will have to adapt. In the short amount of time my company has been manufacturing gear, I've already seen through hole parts I use become obsolete. It's much easier to design for SMT from the beginning than it is to redesign after a bunch of stuff becomes obsolete. If anyone lives in or gets a chance to visit the San Francisco Bay Area / Silicon Valley, I highly recommend checking out the Computer History Museum in Mountain View. They have excellent exhibits that tell the story of how we went from tubes to tiny ass microscopic SMT parts. It's truly fascinating if you are an electronics nerd. www.computerhistory.org/Brad
|
|
|
Post by johneppstein on May 24, 2017 13:47:06 GMT -6
So, the esteemed drbill posted this in the Stam thread. "They are probably the same folks who decry surface mount components." And it made me wonder, what, if any, effect is perceived by the presence of surface mount components in a pro audio circuit? I get it that a lot of folks, especially some purists online, prefer through-hole components because they're easier to service and mod, but in general why is there a bias against surface mount? I've read how some people think the Vintech stuff with surface mount sounds less "vibey", and I'm admittedly guilty of thinking along those lines, but let me throw it out to the experts: If implemented properly, surface mount components are sonically just as good, right? Yes. And I've had various people - including Geoff Daking - tell me about techniques for working on them without using specialized equipment - but they're still a PITA. But then I'm old, my close-up eyesight is nothing like it used to be, and my hand shakes a little sometimes... I believe that in cases where an SMD implementation sounds different from a through-hole version of the same circuit a lot of, if not all, the difference is due to layout, which is something that can be very important but is frequently ignored in audio gear.
|
|
|
Post by johneppstein on May 24, 2017 13:58:07 GMT -6
I'm curious about this too. Bad Cat amps has a new amp that instead of point to point has I think, surface mounted circuit boards. They claim it's more consistent and sounds as good as the hand wired amps. SMD has no business being in a tube guitar amp. The density is not necessary, even undesirable, and it makes service in the field more difficult. The only reason to use it is economy of assembly, and that's a bogus reason in a "boutique" product. Also see my above comment on the effects of layout in audio devices. Bad Cat are losing the narrative.
|
|
|
Post by johneppstein on May 24, 2017 14:04:36 GMT -6
So, the esteemed drbill posted this in the Stam thread. "They are probably the same folks who decry surface mount components." And it made me wonder, what, if any, effect is perceived by the presence of surface mount components in a pro audio circuit? I get it that a lot of folks, especially some purists online, prefer through-hole components because they're easier to service and mod, but in general why is there a bias against surface mount? I've read how some people think the Vintech stuff with surface mount sounds less "vibey", and I'm admittedly guilty of thinking along those lines, but let me throw it out to the experts: If implemented properly, surface mount components are sonically just as good, right? You're going to get quite the gamut of opinions on this.. But as a hardware design engineer who also dabbles in audio work, I feel I have a unique perspective on both. The short answer is that it doesn't matter which one you use, as long as you understand why you're using the parts you are using. Most active devices like transistors or opamps are the same exact part inside the plastic package and that at audio frequencies, the lead lengths don't matter by themselves. It boils down to something I've mention before here.. Stability. Active parts have a wide range of operational bandwidth, and audio frequencies rarely inhabit more than a fraction of the bandwidth of the parts. Because of this, the parts need to be stabilized by feedback or other gain/bandwidth limiting so that you don't have issues with stability. One thing that can happen is that the parts are operated at much wider bandwidths than needed under the belief that wider bandwidth will improve fundamental audio signal performance. In most cases the operational bandwidth is much, much greater than the needed bandwidth with the side effect of being sensitive to small effects of trace/lead inductance and capacitance forming "tanks" or frequencies where the parasitic effects can resonate with the stimulation of the gain/bandwidth of the active part. This results in oscillation as the part starts to feed power into the tank, and the tank then feeds it back to the part. This is, of course, the simplified version, but the effect is well known by most folks. So, in reality, shorter leads on SMD parts, coupled with lack of leads on caps/resistors, etc, generally allow circuits to operate at higher frequencies without the parasitic effects of the legs of the parts. I guess you can say that SMD parts are going to be more stable for the same gain/bandwidth. Audio wise, unless you happen upon an effect of the parasitic components of the circuit actually helping you out, then there is no audio difference or benefit with using leaded parts vs. SMD. Except that SMD encourages a much higher density board with the traces much closer together. This is sometimes not a good thing in audio, and can negate the benefits of the absence of component leads - especially when tubes are involved.l
|
|
|
Post by jimwilliams on May 24, 2017 14:15:43 GMT -6
Guitar amp vibration plus heat will cause those parts to come loose or fail. NASA had problems with early surface mount from g forces. Glue the parts down and coat it all in conformal coating and it's pretty reliable, unless a part fails.
My issues are super thin pcb traces that don't carry current and ceramic caps used for audio. Quality film caps are not available in surface mount. WIMA tried to make surface mount polypropylene film caps but they would melt during assembly. Resistors vary from crappola to Vishay bulk foil at $20 each. Chances are you get the cheap stuff instead. Assembly houses don't like to do mixed signal through hole plus surface mount but that is the only solution if you want to use the best stuff.
|
|
|
Post by svart on May 24, 2017 14:27:43 GMT -6
You're going to get quite the gamut of opinions on this.. But as a hardware design engineer who also dabbles in audio work, I feel I have a unique perspective on both. The short answer is that it doesn't matter which one you use, as long as you understand why you're using the parts you are using. Most active devices like transistors or opamps are the same exact part inside the plastic package and that at audio frequencies, the lead lengths don't matter by themselves. It boils down to something I've mention before here.. Stability. Active parts have a wide range of operational bandwidth, and audio frequencies rarely inhabit more than a fraction of the bandwidth of the parts. Because of this, the parts need to be stabilized by feedback or other gain/bandwidth limiting so that you don't have issues with stability. One thing that can happen is that the parts are operated at much wider bandwidths than needed under the belief that wider bandwidth will improve fundamental audio signal performance. In most cases the operational bandwidth is much, much greater than the needed bandwidth with the side effect of being sensitive to small effects of trace/lead inductance and capacitance forming "tanks" or frequencies where the parasitic effects can resonate with the stimulation of the gain/bandwidth of the active part. This results in oscillation as the part starts to feed power into the tank, and the tank then feeds it back to the part. This is, of course, the simplified version, but the effect is well known by most folks. So, in reality, shorter leads on SMD parts, coupled with lack of leads on caps/resistors, etc, generally allow circuits to operate at higher frequencies without the parasitic effects of the legs of the parts. I guess you can say that SMD parts are going to be more stable for the same gain/bandwidth. Audio wise, unless you happen upon an effect of the parasitic components of the circuit actually helping you out, then there is no audio difference or benefit with using leaded parts vs. SMD. Except that SMD encourages a much higher density board with the traces much closer together. This is sometimes not a good thing in audio, and can negate the benefits of the absence of component leads - especially when tubes are involved.l It really depends on the adeptness of the person directing/doing the layout. If you've taken E-MAG classes or are familiar with the 2D/3D field principles, then you know that there are many, many ways to control crosstalk and signal coupling besides just using distance. In tube stuff, yeah the signals are much higher in amplitude, and inter-stage impedances are usually fairly high, which would lead to increased crosstalk in general. However, the rules of e-mag still apply and the crosstalk can be abated with careful layout work. One such trick is using field polarity and direction to position parts in ways that would not readily allow ingress or egress. People already make use of this when positioning power transformers in chassis, but tend to overlook that all parts have some degree of emission, especially at higher potentials.
|
|
|
Post by BradM on May 24, 2017 15:25:58 GMT -6
Guitar amp vibration plus heat will cause those parts to come loose or fail. NASA had problems with early surface mount from g forces. Glue the parts down and coat it all in conformal coating and it's pretty reliable, unless a part fails. My issues are super thin pcb traces that don't carry current and ceramic caps used for audio. Quality film caps are not available in surface mount. WIMA tried to make surface mount polypropylene film caps but they would melt during assembly. Resistors vary from crappola to Vishay bulk foil at $20 each. Chances are you get the cheap stuff instead. Assembly houses don't like to do mixed signal through hole plus surface mount but that is the only solution if you want to use the best stuff. Jim, I think in the year 2017 assembly shops are pretty well versed in doing both through hole and SMT on the same board. In my Silver Bullet design there is a mix of both stuff. All the SMT goes through the reflow process. Then some through hole is done using a selective solder machine. Then the remaining through hole stuff like transformers is done by hand. You are right--if you want the best stuff that's what you have to do. I do through hole for all audio path electrolytics and film caps. Brad
|
|
|
Post by stormymondays on May 24, 2017 16:13:18 GMT -6
As far as I can see, Bad Cat is not using SMC. They have a line of amps that use PCB boards instead of point-to-point. That's not SMC.
|
|
|
Post by joseph on May 24, 2017 17:18:14 GMT -6
Well not a preamp, but this is surface mount. Modularized Minimoog circuits. Pretty fucking vibey!
|
|
|
Post by Guitar on May 24, 2017 20:45:35 GMT -6
I'm just a hacker/noob, I have a very hard time working on SMD. My hand is not steady enough for those tiny parts. I think it's better left to a wave soldering machine than me. I know some guys in town though, that like Brad, are comfortable with the re-work.
Depending how much importance you put on your capacitors, I think your options would be very limited in SMD. As others have already pointed out. But caps are one of the small parts that make big differences to me in my audio constructions.
My DBX 560A compressors, and my MXR GT-OD, both are SMD heavy boxes that sound great, so I am certain that it can be done well.
|
|
|
Post by BenjaminAshlin on May 24, 2017 21:13:04 GMT -6
I'm curious about this too. Bad Cat amps has a new amp that instead of point to point has I think, surface mounted circuit boards. They claim it's more consistent and sounds as good as the hand wired amps. Hey Martin, Their new range of amps use PCB's but I'm pretty sure they don't use SMD (surface mount devices). Other companies do in their smaller amps.
|
|
ericn
Temp
Balance Engineer
Posts: 14,919
|
Post by ericn on May 24, 2017 21:13:15 GMT -6
I'm just a hacker/noob, I have a very hard time working on SMD. My hand is not steady enough for those tiny parts. I think it's better left to a wave soldering machine than me. I know some guys in town though, that like Brad, are comfortable with the re-work. Depending how much importance you put on your capacitors, I think your options would be very limited in SMD. As others have already pointed out. But caps are one of the small parts that make big differences to me in my audio constructions. My DBX 560A compressors, and my MXR GT-OD, both are SMD heavy boxes that sound great, so I am certain that it can be done well. You think you have problems working with SMD? 😎
|
|