Prática de Shadowing: Deep Dive: Investigating Thermal Cycling Failures Cracked MSOP Solder Joints - Aprenda a falar inglês com vídeo
Escolha um modo de prática
Cada modo abre uma sala de prática em tela cheia.
306 frases
Se as frases estiverem muito curtas ou longas, clique em Edit para ajustá-las.
1
You know what really gets my gears turning?
2
These real world puzzles.
3
And we've got a pretty juicy one for you today popped up on an online forum.
4
Oh yeah What's the story?
5
OK, picture this.
6
You've got an integrated circuit, MSOP package, soldered onto a PCB, standard stuff.
7
But then it goes through this grueling test, 1,000 cycles.
8
1,000, wow.
9
Yeah, extreme temperature swings.
10
We're talking minus 40 degrees C, way up to 85 C.
11
Bone chilling to scourging, basically.
12
Exactly. And the temperature ramps up and down pretty quick, 4 degrees a minute.
13
And they let it sit at each extreme for like 15 minutes.
14
Talk about stress.
15
Yeah, that kind of cycling, that's a serious test.
16
Puts significant strain on all the materials with the expansion and contraction.
17
Totally. And the person who posted, they even shared a cross-section image.
18
You could clearly see it, a crack right there in the solder joint.
19
OK, so failure analysis time.
20
Right, but here's the kicker, the part that makes you kind of scratch your head.
21
Despite this crack, this visible crack, the product, still working.
22
Still functional.
23
That is interesting.
24
Incomplete failure.
25
Exactly. So the big question we're tackling today is why?
26
Why did the solder crack, but the whole thing didn't just die?
27
Good question.
28
So our mission in this deep dive is to explore the potential reasons behind this MSOP solder joint cracking.
29
And we're going to really focus in on the role that differences in the coefficient of thermal expansion, CTE, between the materials might play.
30
It's a fascinating case.
31
Really highlights that subtle interplay of materials, science, and electronics.
32
And like you said, the fact it's an MSOP package, that's a bit unusual for this kind of severe cracking, isn't it? Right, that's what I thought, too.
33
Now, the original post, it gave us a good starting point, but well, it was missing some key details.
34
Always the way, isn't it?
35
Yeah. We don't know the exact solder paste used or the specifics of the PCB type, thickness, surface finish, nada.
36
And the location on the board?
37
Or if other parts failed?
38
Nope, none of that.
39
Or even if other components nearby had similar issues.
40
It's like trying to solve a puzzle with half the pieces missing.
41
Those missing details are often so critical.
42
Solder alloy, for instance, tells you its specific CTE characteristics right away.
43
Exactly. And just to make sure everyone's on the same page, MSOP.
44
Mini Small Outline Package.
45
It's basically a smaller SOP, tighter pin pitch, like 0.5, 0.65 millimeters maybe.
46
Usually for lower pin count ICs, less than 16 pins typically.
47
Compact. Yeah. And what's interesting, like we said, is MSOPs aren't usually the first thing you suspect for solder joint cracks.
48
They're smaller.
49
Joints tend to be pretty resilient.
50
Compared to, say, big MLCCs, which hate bending or connectors that see physical stress.
51
Precisely. So this report of a cracked MSOP definitely got our attention.
52
It does suggest we need to look beyond maybe the usual suspects.
53
The small size normally means less leverage for stress.
54
OK, so the initial buzz on the forum, the first reaction was, CTE mismatch.
55
Got to be CTE.
56
The classic answer.
57
Right. Makes sense on the surface.
58
Different materials, different expansion rates when things heat up or cool down.
59
But as you know, just blaming CTE, well, it might be a bit too simple.
60
It often is.
61
I mean, CTE differences are pretty much always there in electronics.
62
The industry spent decades optimizing materials to handle it.
63
Especially since it was still working, right?
64
Suggests maybe something else, like mechanical stress, was a bigger player?
65
Possibly, or at least a significant contributor.
66
It's rarely just CTE unless the mismatch is huge or other factors line up badly.
67
OK, so for anyone listening who's maybe not deep into CTE, let's break it down simply.
68
Imagine two materials stuck together.
69
Temp goes up, they grow.
70
Temp goes down, they shrink.
71
But if they grow and shrink at different rates, different CTEs, that puts stress right where they meet.
72
Yeah, like two people holding hands, one taking giant steps, the other tiny ones.
73
There's going to be tension.
74
Perfect analogy.
75
So in our MSOP case, there are three main spots where the CTE mismatches could matter.
76
First, between the solder itself and the IC lead frame, usually copper.
77
Right, the lead frame.
78
That's the metal pathway from the tiny chip inside out to the PCB.
79
Copper is a standard.
80
Good conductivity.
81
Exactly. Now, the forum user didn't say, but we can probably assume it's SAC 305 solder.
82
It's super common, lead-free.
83
Good assumption, industry standard pretty much.
84
OK, so SEC 305 has a CTE in the ballpark of, what, 22 to 25 ppm degree C, parts per million per degree Celsius.
85
Around there, yeah.
86
And the lead frame, copper, often tin-plated for solderability.
87
Copper's CTE is lower, maybe around 16.5 ppm degrees.
88
So there's a difference.
89
The solder expands and contracts more than the copper lead.
90
Right, and that difference creates sheer stress at that interface, especially when it gets cold, right?
91
The solder shrinks more.
92
Exactly. That low-temp part of the cycle is often where the strain really racks up.
93
Over many cycles, that repeated stress that leads to fatigue cracks forming.
94
Makes sense, like bending a paper clip back and forth eventually snap.
95
And we should probably remember, these solders, like SAC 305, they aren't just tin, silver, copper.
96
They have tiny amounts of other stuff, too.
97
These elements, yeah.
98
They fine-tune the property.
99
Yeah, I saw a blog post about it.
100
A bit of silver for toughness, fatigue resistance, a bit of copper for stiffness, strength.
101
It's a real balancing act.
102
Absolutely.
103
Those craze additions can really impact long-term reliability under thermal and mechanical loads.
104
OK, so that's solder to lead frame, interface number two.
105
Solder to the IC package body itself, that black plastic mold compound.
106
Ah, yes, the mold compound.
107
What's the CTE story there?
108
Generally, mold compounds are lower, maybe 10 to 20 ppm degrees, so usually lower than the solder.
109
OK. And its main job is protection, right?
110
Protect the chips, support the leads.
111
But its own expansion and contraction, well, it can push sideways on the solder joints, lateral forces.
112
Ah, so it could concentrate stress maybe at the edges where the solder curves up the fillet.
113
Potentially, yeah.
114
Yeah. Especially at the heel toe of the fillet.
115
Yeah. But you know, these mold compounds, they've been used forever.
116
Their properties are pretty stable, well known.
117
OK. Often the lowest CTE in the whole stack up, actually.
118
OK, good point.
119
That brings us to number three.
120
And this is often the big one, right?
121
Solder to the PCB itself.
122
Assuming standard FR4.
123
Yeah, this is usually the most critical interface for a CTE mismatch.
124
FR4, it's tricky.
125
Oh, so? Well, it expands differently depending on the direction.
126
Yeah. In the x and y plane, you know, along the surface of the board, the CTE is maybe 14 to 17 ppm degrees C, not too far off from copper or solder.
127
OK. But in the z-axis, through the thickness, below its glass transition temp Tg, it's already much higher, like 70 to 120 ppm degrees.
128
Whoa, OK, that's a jump.
129
And then when you go above Tg, which definitely happens during reflow soldering, that z-axis CTE, the alpha 2 CTE, it can skyrocket, 190, 200, even 300 plus ppm degrees.
130
Jeez. But the source mentioned cracking is less likely during reflow because the solder's melted, right?
131
Correct. Solder's liquid, it just flows.
132
The stress builds up later during operation, during those repeated temperature cycles.
133
Even if the product only operates, say, below Tg.
134
Right, even below Tg, that repeated cycling causes expansion and contraction.
135
And that big z-axis movement, that up and down stretching and squashing of the board thickness, that pulls and pushes on the solder joints connecting the components.
136
Tearing forces, basically, leading to fatigue over time.
137
Exactly. Even if the operating temps aren't super high, the board remembers that high temp reflow state.
138
And cycling stresses it relative to that.
139
So the XY expansion across the board surface is more uniform, maybe less direct stress on one tiny MSOP joint, unless you have really uneven heating.
140
Generally, yes.
141
The glass fibers in the FR4 do a decent job controlling XY expansion.
142
It's that z-axis movement through the layers that's harder to constrain and often causes more trouble for the joints.
143
OK, let's sum up the CTE likelihood, then.
144
You generally need, what, a difference bigger than 10 or 15 ppm / ˚C for really significant stress and cracking from CTE alone?
145
That's a reasonable rule of thumb, yeah.
146
So looking at the numbers, solder to copper lead frame, maybe 6 to 9 ppm difference.
147
Solder to mold compound, maybe 2 to 15 ppm difference.
148
Solder to FR4 XY, maybe 5 to 11 ppm difference.
149
They're mostly kind of borderline or within that range, not screamingly high mismatches, assuming standard materials like SAC305.
150
If they used, say, a brittle, low temp solder, that's a different story.
151
Right. So CTE is definitely a factor adding stress, but maybe not the sole smoking gun here, especially since it still worked.
152
The community thought CTE first, but maybe the evidence points elsewhere.
153
It seems plausible.
154
That functional aspect really pushes us to consider other things more strongly, like mechanical stress.
155
OK, so if it's not just CTE, how do we figure out what really happened?
156
Time to put on our detective hats, I guess.
157
Indeed. When you face a failure like this, you need a systematic approach.
158
Experience helps, sure, but structure is key.
159
OK, walk us through it.
160
Step one. Step one, information gathering.
161
Always. You can't solve anything without the facts.
162
We use the 5W1H framework, who, what, when, where, why, how.
163
We need more on the product's application, the exact test profile, manufacturing details, all of them.
164
More info equals better chance of finding the root cause.
165
Makes sense.
166
The forum post gave us the what crack joint and some when we're 1,000 cycles, but the why and how are big question marks, plus those missing details like solder type, PCB specs.
167
Exactly. So after gathering what we can.
168
Step two. Get the actual failed sample and ideally related devices, too.
169
Ah, the physical evidence.
170
Yeah. Photos are OK.
171
Descriptions are helpful, but nothing beats having the actual cracked board.
172
Avoid misinterpretation and grab any bits that fell off.
173
Got it. Handle with care.
174
OK, samples in the lab.
175
What's next under the microscope?
176
Step three.
177
Check solderability and crucially, inspect the IMC, the intermetallic compound.
178
IMC. OK, remind us what that is again.
179
It's that thin layer that forms right where the solder meets the metal pad on the PCB and where it meets the component lead.
180
It forms during soldering when heat makes the metals react.
181
And looking at it tells you.
182
It tells you about the initial soldering quality.
183
Was the reflow profile right?
184
Any process control issues, oxidation on the parts, a good IMC is key.
185
So what does a good IMC look like?
186
Uniform continuous layer, both at the PCB pad and the component lead.
187
If it's patchy or way too thick or has voids, that suggests problems in the initial soldering.
188
But you usually need to cross section to see it properly, like the forum user did.
189
Right. And based on that image they shared, what did the IMC look like in this case?
190
The description said the IMC down at the PCB pad looked good, uniform.
191
So probably means the reflow process and the board finish were OK.
192
OK. What about at the lead itself?
193
Harder to tell precisely because of the crack right there.
194
But apparently, there was evidence of IMC at the base of the lead.
195
And the growth on the sides of the lead was normal, so just as likely OK there too initially.
196
So initial soldering probably wasn't the root cause.
197
But you said IMC, even if formed well, can be a weak point.
198
Yes, absolutely.
199
It's essential for the bond.
200
It is the bond, metallurgically speaking, like mortar in a brick wall.
201
But it's also typically more brittle than the bulk solder alloy itself.
202
So when the joint is stressed, mechanically or thermally, that brittle IMC layer is often where cracks like to start.
203
It's the weakest link in the chain, mechanically speaking.
204
Gotcha. OK, so even with good IMC, stress finds the weak spot.
205
What's step four in our investigation?
206
Step four, examine the crack morphology, the shape, the path it took.
207
What did the photo suggest?
208
It looked like the crack probably started right at that IMC interface at the bottom of the MSOP lead.
209
And interestingly, the complete break, the full fracture, seemed to be only at the bottom.
210
The cracks on the sides hadn't gone all the way through, which, as we said, explains why it still worked.
211
Current could still sneak through the partly cracked sides.
212
Makes perfect sense.
213
And reinforces the idea that the IMC was the point of failure initiation.
214
Exactly. And ideally, you'd want to look at other leads on that same MSOP, see if they show similar early signs of cracking.
215
Is it just one pin or systemic?
216
Right, get the bigger picture.
217
Now, switching gears slightly, that number, the 1,000 cycles, the source material pointed out that's pretty extreme for typical commercial stuff, yeah.
218
Oh, absolutely.
219
Standard reliability test, maybe simulating shipping, you might do 5, 10 cycles, maybe 100 for some things, 1,000 cycles like that.
220
That's more like component qualification or high-rel stuff, automotive, aerospace, military.
221
So it makes you wonder about the product's intended use.
222
Why such intense testing?
223
It does. And it really makes you think maybe something beyond just inherent material properties was at play.
224
Was the board itself flexing a lot under those conditions?
225
Good question.
226
And the source author also wondered why they did a destructive cross-section after finding it still worked.
227
Maybe just routine inspection found it.
228
That's plausible.
229
Sometimes non-destructive tests, like X-ray, might misfind cracks like that.
230
Cross-sectioning gives you the definitive answer.
231
OK, so let's recap.
232
CTE mismatch, probably a factor, but maybe not the main driver alone.
233
Initial solder quality IMC looked OK based on the limited info.
234
Seems that way.
235
So where does that leave us?
236
The source material strongly suggests mechanical or bending stress as the most likely culprit.
237
Why that? Well, if the usual suspects aren't clearly guilty, you look for other forces.
238
And PCB warpage during temp cycling is a huge one.
239
Warpage? How does that happen?
240
Multilayer boards.
241
They're complex structures.
242
Copper layers, insulation, via maze, they don't all expand and contract perfectly evenly, especially with big temperature swings.
243
This unevenness causes the board to, well, warp, bend, flex, twist slightly.
244
And that slight bending stresses the components sitting on top.
245
Exactly. The components are rigid.
246
The board beneath them is flexing.
247
That force gets transmitted straight into the solder joints, holding them down.
248
And over 1,000 cycles of that bending back and forth, that's classic fatigue.
249
Causes cracks, often with those telltale branch patterns as the crack finds the easiest path.
250
And thinner boards would warp more, right?
251
Less rigid?
252
Generally, yes.
253
And location matters, too.
254
Components far from mounting points, far from where the PCB is screwed down to the chassis, they often see more stress.
255
Because the board flexes more out in the middle.
256
Right. Or if the chassis itself deforms with temperature, it can pull or push on the PCB.
257
And those components out on their own become the stress relief points.
258
Their solder joints take the hit.
259
OK, that sounds like a very plausible explanation for this case.
260
So if PCB bending mechanical stress is the likely villain, what can engineers do to stop it?
261
Countermeasures.
262
Several options.
263
First, design phase.
264
You can use CAD simulation, finite element analysis, FEA.
265
To predict the stress.
266
Exactly. Simulate the temperature cycles on the whole assembly, PCB, and chassis.
267
See where it deforms most.
268
Then you can modify the design.
269
Maybe change screw locations, add more mounting points, maybe move sensitive components like this MSOP to a less flexy part of the board, or add stiffeners to the chassis.
270
Or even choose a more rigid PCB material, though that costs more.
271
Makes sense.
272
Address the root cause in the design.
273
What about manufacturing process fixes?
274
The source mentioned underfill.
275
Yes, underfill.
276
Basically, squirting epoxy under the component between it and the PCB.
277
To reinforce the joints.
278
Right. It adds mechanical support, helps distribute the stress away from just the solder.
279
It can definitely help.
280
But as the source wisely pointed out, it often just delays the inevitable if the underlying warpage is still happening.
281
It treats the symptom, not the disease.
282
Eventually, the underfill itself might crack under enough stress.
283
So it might get you through warranty, but maybe not solve the fundamental problem.
284
It can be like that sometimes.
285
It's a valid technique, often used.
286
But it's best when combined with addressing the root cause of the stress.
287
So the best approach is always figure out why it's cracking the root cause, verify it, and then fix that fundamental issue.
288
Absolutely.
289
That's the path to a truly robust and reliable solution.
290
Well, this has been a really fascinating deep dive.
291
It's amazing how complex even a simple solder joint failure can be, so many interacting factors.
292
It really is.
293
And we have to remember, our whole discussion here is based on assumptions, given the limited info from that forum post.
294
There could easily be other factors we haven't even touched on.
295
Absolutely true.
296
And to you, listening if you have other ideas, alternative takes on this cracked MSOP, we genuinely love to hear them.
297
Solving these engineering puzzles really benefits from different perspectives.
298
Definitely.
299
And hey, if you enjoy digging into electronics, manufacturing, reliability, stuff like this, make sure you subscribe to our channel.
300
Please do. And if you found this chat valuable, maybe give us a like, subscribe, share it around, a five-star review on your podcast platform.
301
That always helps, too.
302
Helps others find us.
303
Yep. You can find links to anything related we mentioned down in the show notes or episode info.
304
Thanks so much for tuning in for this deep dive.
305
Yeah, thanks, everyone.
306
Until next time, keep asking questions, keep learning.
Baixar aplicativo
Tudo o que você precisa para falar fluentemente
Pronúncia com IANota em cada frase
Prática de IPADomine cada som
VocabulárioMonte seu banco de palavras
Jogo de vocabulárioAprenda brincando
Mesmo canal
B23
8D Reports Made Simple: How to Solve Problems at Work—and at Home
Business EnglishWorking Adults
B25 Key Mindsets Before You Start Writing an 8D Report
Business EnglishWorking Adults
C11
什麼是屏蔽罩(shielding can)？工程師對屏蔽罩的愛恨情仇
Daily ConversationUniversity StudentsWorking Adults
C12
What is IMC (Intermetallic Compounds) part I? The role of IMC plays in the soldering
Daily ConversationWorking Adults
C12
What is IMC (Intermetallic Compounds) part II? The different between IMC, alloy, eutectic
Daily ConversationUniversity StudentsWorking Adults
C14
What is IMC (Intermetallic Compounds) part III? Explain Alloy, Eutectic in detail
Daily ConversationUniversity StudentsWorking Adults
C15
QC 7 tools – Cause and Effect Analysis (Fishbone) Diagram Introduction
Daily ConversationWorking Adults
C12
What Are PCB and PCBA? What’s the Difference Between Them?
Business EnglishUniversity StudentsWorking Adults
C1How to evaluate whether an 8D report is well written
Business EnglishWorking Adults
FOR YOU
Vídeo recomendado
C11
174 frases
Via-in-Pad Reliability: When Does Your PCB Need It? #pcbmanufacturing #pcbdesign
Business EnglishUniversity StudentsWorking Adults
C18
70 frases
When open-sourcing your code goes wrong...
American AccentBusiness EnglishUniversity StudentsWorking Adults
C12
245 frases
Power Plant | How it Works Power Plant Steam Cycle
Business EnglishUniversity StudentsWorking Adults
C13
85 frases
Material Properties 101
Business EnglishUniversity StudentsWorking Adults
C19
265 frases
How are Microchips Made? 🖥️🛠️ CPU Manufacturing Process Steps
Business EnglishUniversity StudentsWorking Adults
C111
176 frases
A Psychiatrist Explains "Quiet Cracking" (While Your Work Still Looks Fine)
American AccentBusiness EnglishUniversity StudentsWorking Adults
C11
250 frases
Embedded Insiders: Qualcomm's Acquisition of Arduino
Business EnglishUniversity StudentsWorking Adults
C12
280 frases
Siemens S7-1200: Signal Modules - Wire, Program, Test
Business EnglishUniversity StudentsWorking Adults
C12
244 frases
Real-Time Digital Twins Using CFD, Sensors & AI - Haris Kokkinos | Podcast #165
Business EnglishUniversity StudentsWorking Adults
B24
200 frases
The Chopsticks get Chopped - Starbase Summary
Business EnglishUniversity StudentsWorking Adults
C11
19 frases
I Made an EMP Device... It Actually Works
Daily ConversationUniversity StudentsWorking Adults
Vocabulário e dicas de fala para esta lição
Esta aula de conversação de nível C1 usa o vídeo “Deep Dive: Investigating Thermal Cycling Failures Cracked MSOP Solder Joints”. As palavras que mais se repetem: solder, Cte, stress, crack, lead. Este vídeo tem 306 frases e 3067 palavras para praticar shadowing. A fala dura 17:13. O falante fala em ritmo natural, cerca de 178 palavras por minuto, próximo de uma conversa do dia a dia. Apenas 77% das palavras estão entre as 3.000 mais comuns do inglês, por isso o vocabulário é exigente.
Vocabulário principal deste vídeo
As 15 palavras mais avançadas do vídeo, com pronúncia e significado:
| Palavra | Pronúncia | Significado |
|---|---|---|
| solder substantivo | /ˈsɒl.də(ɹ)/ | solda |
| mold substantivo | /moʊld/ | mofo, bolor |
| chassis substantivo | /ˈt͡ʃæsi/ | chassi, chassis |
| fatigue substantivo | /fəˈtiːɡ/ | fadiga |
| axis substantivo | /ˈæksɪs/ | eixo |
| brittle adjetivo | /ˈbɹɪtəl/ | frágil, quebradiço |
| contraction substantivo | /kənˈtɹækʃ(ə)n/ | contração |
| plausible adjetivo | /ˈplɑ.zɪ.bəl/ | plausível |
| thickness substantivo | /ˈθɪknəs/ | grossura, espessura |
| rigid adjetivo | /ˈɹɪd͡ʒ.ɪd/ | rígido |
| reliability substantivo | confiabilidade, fiabilidade | |
| puzzle substantivo | /ˈpʌzəl/ | enigma |
| deform verbo | /dəˈfɔɹm/ | deformar |
| fascinate verbo | /ˈfæsɪneɪt/ | fascinar |
| alloy substantivo | /ˈæ.lɔɪ/ | liga |
Phrasal verbs que você vai ouvir
| Palavra | Pronúncia | Significado |
|---|---|---|
| figure out verbo | descobrir, deduzir | |
| point out verbo | apontar | |
| cool down verbo | /kuːl ˈdaʊn/ | esfriar, arrefecer |
| fall off verbo | cair | |
| go down verbo | descer | |
| go up verbo | subir |
Frases que vale a pena repetir
Frases curtas e completas do vídeo que você pode usar na conversa do dia a dia:
- You know what really gets my gears turning?
- So the big question we're tackling today is why?
- Right, that's what I thought, too.
- Always the way, isn't it?
- What's step four in our investigation?
Gramática neste vídeo
As estruturas que o falante mais usa, com as palavras exatas do vídeo:
| Estrutura | No vídeo |
|---|---|
| Present perfect have/has + particípio passado — uma ação passada que ainda importa agora | they've been · has been · haven't even touched |
| Voz passiva be + particípio passado — o foco está no que acontece, não em quem faz | is screwed · is based |
Pronúncia para ficar de olho
O falante usa 25 contrações e formas reduzidas, como aren't, we're, they're. Diga-as na forma curta, do jeito que você ouve.
- Os sons de “th”: thickness /ˈθɪknəs/, thermal /ˈθɜːməl/, thumb /ˈθʌm/
- Os sons de “sh” e “zh”: contraction /kənˈtɹækʃ(ə)n/, assumption /əˈsʌm(p).ʃ(ə)n/, qualification /ˌkwɑlɪfɪˈkeɪʃən/, simulation /ˌsɪm.jəˈleɪ.ʃn̩/, sheer /ʃɪɚ/
- Palavras longas — acerte a sílaba tônica: mechanically /məˈkænɪkli/, ideally /aɪˈdi.ə.li/, qualification /ˌkwɑlɪfɪˈkeɪʃən/, automotive /ˌɔtəˈmoʊtɪv/, characteristic /ˌkæɹ(ə)ktəˈɹɪstɪk/
Como praticar com este vídeo
- Ouça o vídeo inteiro uma vez sem falar e anote as palavras que você não conhece.
- Comece na velocidade 0,75×, faça shadowing frase por frase e volte à velocidade normal quando ficar fácil.
- Grave a sua voz e compare com o original, prestando atenção a palavras como solder, mold, chassis.
O que é a Técnica de Shadowing?
Shadowing é uma técnica de aprendizado de idiomas com base científica, originalmente desenvolvida para o treinamento de intérpretes profissionais. O método é simples, mas poderoso: você ouve áudio em inglês nativo e repete imediatamente em voz alta — como uma sombra seguindo o falante com 1-2 segundos de atraso. Pesquisas mostram melhora significativa na precisão da pronúncia, entonação, ritmo, sons conectados, compreensão auditiva e fluência na fala.