t****1 发帖数: 827 | 1 the original question says: if Vgs<0.7, then cut off. did the devices in your simulation really cut off? no leakage current? Of course not, if you understand what is weak inversion.So, your simulation condition is different from the assumption in that question.
Ok, now let's assume it really cuts off. no leakage current. from your
answer,if vin=0, then vout=0.7, and if vin=0.1then vout is about 0.8v. is
that correct?
My question: when vin increases from 0 to 0.1v, why Vout can not stay at 0.7v. ... 阅读全帖 |
|
f****3 发帖数: 502 | 2 首先十分感谢你的多次回帖,谢谢!先说明一下,我只是请教问题,没有对任何人的回
复有任何的态
度问题,如果我哪句话让你不爽了,请谅解!
下面说一下这个问题,我又仔细看了一些wiki里面关于crossover的阐述,wiki那张图
上面接
Vdd,下面接Vss,只有负载接Gnd,这样就比较好理解了。
如果|Vin|<0.7,两管截止,输出为0,当Vin>0.7时,上面N管导通,Vout=Vin-0.7,当
Vin<-
0.7时,下面P管导通,Vout=Vin+0.7。当相应于我这到题来讲,上面依然是Vdd,但是
下面接的
GND,所以将GND等效为Vss时,那么输出2.5V即是相应wiki的输出为0,这样就很好理解
为什么不
是一直follow,而在中间需要转换,这是我想问的问题。
但是不好意思,你一直纠结于有无负载,我承认你的思维很严谨,对问题的分析很透彻
,但你并没有
领会我的困惑,呵呵,或许我的困惑很弱智,无需回答,但同时你也并没有说出电路真
正工作的过
程,即使你一开始就给出了很合理的答案。
Anyway,依然非常感谢你的回复,再重申一遍,我对你的回复没有任何的态度上面的问
题,请... 阅读全帖 |
|
c*****v 发帖数: 839 | 3 On Pg4 of the datasheet there's Vout UVLO. On Page 26, An undervoltage
lockout circuit on the USB PowerPath VOUT pin shuts down and prevents both
buck and LDO from enabling when Vout pin voltage drops below 2.6V. I am not
sure whether this is to protect the battery. One other situation for Vout
ULVO is that the battery is being removed.
detect |
|
E*****a 发帖数: 757 | 4 你的输出电压是Vout还是Vfout?
如果是Vout的话,乍一看就有可能会动吧
你要下面那个opa227做什么用?
Vout是Vin的一个比例,还是Vout要稳定在一个固定值?
如果是个固定值,你必须有个固定值的reference,然后搭loop
如果是Vin的一个比例,下面那个opa没有用吧。直接比例输出就好了 |
|
a****y 发帖数: 255 | 5 它有LXpin做电压输入, OUTpin做电压输出, CEpin是使能控制.
如果CE接高电平, 芯片工作, Vout就是变换后的电压.
可是如果CE接低电平, 芯片不工作, Vout是多少呢?
在Vin是直流3V左右,
Vout是等于Vin减去二级管的正向导通电压(这个情况 下, 是0.3V) 吗? |
|
S****e 发帖数: 1063 | 6 I got it, the IC has two parts, powerpath controller and charger. Vout is
the output voltage of the powerpath controller, and the input voltage of the
charger, like a midpoint between two cascaded stages.
the Vbus UVLO prevents the controller from operating when bus voltage is too
low, the Vout UVLO prevents the chargers from operating when their input(
which is the controller output) is low. both UVLO are protection against low
input voltages, not output. It's the term Vout that makes it confus |
|
x*****9 发帖数: 3 | 7 我要测量该电路的输出电压与温度的关系,该输出电压是数字电压,所以,一定是大信
号。可是,对于switch cap电路,不可以run dc,因为对于这种电路,dc是开路的,所
以,得不出任何有用的信息。就我所知,对于这种电路,只能run transient,而且为
了得到不同的温度下的值,可以run temp parametic analysis。可是,这种情况下画
出的图是一组不同温度下的Vout和time的关系,而不是我希望的Vout和Temp的关系。
急盼回复,谢谢! |
|
a****y 发帖数: 255 | 8 多谢. 可是如果看 Vin -- 电感 -- 二级管 -- Vout 这个联线,
这时候(CE disabled), Vout上的电压应该是Vin减二级管的正向导通电压吧. 虽然此时
芯片没有提供输出电压. 我的理解对吗? |
|
|
f*****0 发帖数: 489 | 9 if you read the datasheet more carefully, you will know that the Vout pin on
the chip isn't to output voltage. the schottky diode / capacitor is. read a
little bit more about boost converters and you will understand why.
yes, if you the CE pin low, it would disable the chip (which by the way is a
PFM chip, good for low-load or standby situations). When that happens, the
LX pin is left float and the current will flow from Vin to Vout through the
inductor and the schottky diode. So the rough volta |
|
f*****0 发帖数: 489 | 10 "当chip不工作时,内置FET不工作,外面那个电感saturate,怎么可能还有输出。"
when the chip is disabled, the gate goes low, and the drain goes high
impedance. When that happens, the current goes through the inductor and the
diode to the load.
another way to look at it: notice that the Out pin supplies Vdd for the chip
and is connected to Vout? if Vout went low when the chip is disabled, how
would the chip have started?
BTW, the chip can be easily converted to a gated PFM regulator, aka LT1073
style. |
|
f*****0 发帖数: 489 | 11 the widlar bandgap recognized the positive but small TC for delta Vbe and
tried out to lay it on top of a Vbe (which has a negative TC).
so the voltage drop over R2, V2=I2*R2=(delta Vbe / R3)*R2. by adjusting R2/
R3, you can give it a tempco that is "roughly" +2mv/c.
the Vout = V2+Vbe (for Q3), and by trimming R2/R3, you get a Vout that has a
0 TC.
the brokaw design is roughly the same except that it runs both legs of the
unbalanced "current mirror" at the same current - it is driven by a curren |
|
b***r 发帖数: 149 | 12 问这个circuit的gain
我当时的回答:
if it is ideal op-amp
=〉equal potential at + and - terminals
=> + terminal virtual ground
=> (Vin-0)/R1 = (0-Vout)/R2
=> Vout = - (R2/R1) * Vin
then the interviewer asked me if i was sure that the gain is negative?
i said if the op-amp is ideal, then yes. however, this configuration forms a
positive feedback. if the output voltage increases a bit, the + positive
also increases a bit, then the output increases more. therefore, this
circuit can oscillate.
he put it aside an |
|
b***r 发帖数: 149 | 13 thank you for the replies
i agree that the circuit is in positive feedback
my confusion is:
does this equation still hold for "ideal" op-amp? (Vin-0)/R1 = (0-Vout)/R2
clearly i can't derive Vout=infinity from this equaiton
the |
|
ET 发帖数: 10701 | 14 可这你没解释为啥amplifier transistor contribute more non-linearity呀。
你也说了,transconductance是nonlinearity source,
amplifier的transcondutance, gm = f(vgs, vth), 即使考虑short channel effect,
vds, gm = f(vgs, vth, vds)
对于既定的biasing point, vgs, vth look quite constant for me.
即使vgs, vth变化,那对gm的影响也是线性变化 (level -1 model).
从输入到输入的gain可近似为gm1/(gm2+gmb2) ,如果考虑body effect of cascode tr
ansistor,
而vout = gain * vin 即使vin是small signal sin. 如果gain是个常数,或者线性变化
的,vout的变化也是线性的。
对gm1/(gm2+gmb2) 求导,with respect to bias. 带来影响的我只 |
|
ET 发帖数: 10701 | 15 这里的一个重要区别是考虑high frequency, capacitive load
要看impedance,而不只是低频下的resistance.然后用 r*c来近似一个pole.
实际上,也可以假设output resistance 很大,无穷来简化。
你是从vin->vout ,在m3 的gate端的断开的吗?
比如你的boosting amplifier,有Gm, smallest Rout & infinite Rin , ideal opamp
.
从vin, 在m1的输出有 v2, v2=gm1*vin/sc1, 可以画一个current source:gm1*vin //
c1
v2控制boosting amplifier & cascode amplifier
boosting amplifier, 可以画成gm3*v2 //c3, 它有输出电压:v4=gm3*v2/sc3
再看cascode transistor, 它的current source是gm2*vgs , here vgs=v4-v2, 这个cu
rrent source // c3
所以,vout= |
|
ET 发帖数: 10701 | 16 理论上是这样(或不是这样),但只是定性的描述。如果这样解释,voltage mode和cur
rent mode就没区别了。
另外,以你这样的解释,是谁控制谁的问题了。首先,peak current是哪来的? peak
current是根据需求的output, 折算出的duty cycle,还有设定的current ripple, in
ductor确定的。 根据这些来设定稳定态的duty cycle. 这个peak current都是事先设定
了.
比如给定vg, ig, 要求弄个vout, iout(rms) = vg*ig/vout,
再说我可以有50%的ripple. 那么peak current就是iout(rms)+/ 25% iout(rms).
可问题是,current mode 和voltage mode都在的时候,output voltage 经过error am
plifer后的输出, vcontrol, 设定的peak current.
剩下的部分我就不明白了。
it
qua
up |
|
ET 发帖数: 10701 | 17 如果是prediction就先进了。可惜不是。
这个原理和pwm & pfm相似的意思。light load, pulse frequency modualtion, 是监
测到peak current的结果。这时候vout还是恒定的。
但是,同时也把vout给降低了,就这个意思。
当然,这个一般超过了buck converter的工作。这时候cpu可能不干了。所以这里就有一
个让cpu也工作的vdd.
具体的需要看Anthony Stratakos的phd 论文了。 |
|
g****t 发帖数: 31659 | 18 哪有他的论文下载?
这个公司很牛嘛,也就13年历史的小公司,今年第一季度销售额几千万了.
如果是prediction就先进了。可惜不是。
这个原理和pwm & pfm相似的意思。light load, pulse frequency modualtion, 是监
测到peak current的结果。这时候vout还是恒定的。
但是,同时也把vout给降低了,就这个意思。
当然,这个一般超过了buck converter的工作。这时候cpu可能不干了。所以这里就有一
个让cpu也工作的vdd.
具体的需要看Anthony Stratakos的phd 论文了。 |
|
l*****i 发帖数: 296 | 19 depends on the input impedance of opamp.
suppose the input impedance is Zin
then
A(s)(vin-Zin/(Zin+Z(s))vout)=vout
H(s)=A(s)/(1+Zin*A(s)/(Zin+Z(s)))
(1 |
|
l***g 发帖数: 1035 | 20 by clamping i meant Vout = 1/(1-D)Vin, I set a Dmax so that Vout won't excee
d the Vmax under concervative Vin. |
|
j******e 发帖数: 526 | 21 如果|Vin|<0.7,两管截止,输出为0,当Vin>0.7时,上面N管导通,Vout=Vin-0.7,当
Vin<-
0.7时,下面P管导通,Vout=Vin+0.7。当相应于我这到题来讲,上面依然是Vdd,但是
下面接的
GND,所以将GND等效为Vss时,那么输出2.5V即是相应wiki的输出为0,这样就很好理解
为什么不
是一直follow,而在中间需要转换,这是我想问的问题。
=========================================================
负载很重要,没有负载的话即使Vin>0,7,上面的管子也不会导通,电流没有回路
而且负载所接点位也很重要,因为上下管都截止了,输出点悬空,由负载所接点位决定
,不知道说的对不对 呵呵 |
|
m****o 发帖数: 264 | 22 u sure the ckt hookup is all right?
ideal case first,
vout = (Vi + Vfout)/2*(1+R3/R4)=Vi+Vfout;
i_R = i_Load = (vout - Vfout)/R = Vi/R;
Vfout = Vi*R_load/R
am i missing something? |
|
x*****9 发帖数: 3 | 23 我要测量switch cap ckt simulation该电路的输出电压与温度的关系,该输出电压是
数字电压。可是,对于switch cap电路,不可以run dc,因为对于这种电路,dc是开路
的,所
以,得不出任何有用的信息。就我所知,对于这种电路,只能run transient,而且为
了得到不同的温度下的值,可以run temp parametic analysis。可是,这种情况下画
出的图是一组不同温度下的Vout和time的关系,而不是我希望的Vout和Temp的关系。
急盼回复,谢谢! |
|
i****2 发帖数: 1520 | 24 After one on-site interview this Tuesday, the flash memory company in San
Jose gave me the offer same day. but another chip company I was on-site
interviewed yesterday told me they need several days to notify me the
interview result.
more questions to share:
1: how to define the noise margin and check the noise margin of gates:
answer: high noise margin is NM high = Voh-Vih ; NM low = Vol - Vil. you
could use circuit level simulation too to get the 2-D graph of Vin and Vout.
when the k=-1 , tha |
|
b*****g 发帖数: 2727 | 25 我的fireplace是heat-n-glo AT-GRAND, pilot燃着没问题,但一开开关燃半分钟就连
pilot一起灭掉。测了pilot燃着时候thermopile open voltage 440mV, 开关打开掉到
170mV. thermopile型号2103-512。 thermodouple难拆一点所以还没有测。请问是不是
thermopile 的问题?谢谢。 |
|
i******l 发帖数: 268 | 26 让我高中时候在想一个问题,问了化学老师(当时她还是我的班主任),结果被呵斥:
别问我这样无聊的问题!可我心里一直没放下。
想:人一直在呼吸,肺活量是基本恒定的,所以我们吸进去的气体和呼出来的体积是一
样的(Vin=Vout),这意味着吸进去的气体的分子数(摩尔量)和呼出来的是一样的(n
_in=n_out);但是,吸进去的是氧气O2,呼出来的是二氧化碳CO2,这一进一出,就差
了一个C原子,同时分解出的水被排出体外;这样,体重就轻了,或者说:就减肥了。
详细计算:安静时,人每分钟呼吸7-8升气体,意思是每天呼吸7*60*24=11,000升气体
。吸进去的气体氧含量20%,呼出来15%,意思是人一天消耗氧550升。
为了简化计算,我们考虑一些量的对应关系:消耗一摩尔O2,对应一摩尔C,对应一摩
尔碳水化合物(CH2O)。意思是,如果产生的水分能够及时排出体外,每消耗一摩尔O2
,我们失去一摩尔CH2O的重量(20g)
一摩尔气体占22.4升,所以我们每天呼吸可以减肥550/22.4*20=500g,一斤。 |
|
q***z 发帖数: 934 | 27 Now I am analysizing data from experiment
plot Vout vs THD, etc :PP |
|
c****s 发帖数: 2487 | 28 here it is... though i do think there must be something else wrong
well, the whole story is, I had a curve printed out as .ps, then
converted the .ps to .fig with pstoedit, then edited it with xfig,
finally exported the .eps
I also tried another way around: imported the .ps into corel draw,
then exported the eps after editting. still the same beep.
%%%%%%%%%%
\begin{figure}
\centering
\psfrag{celsius}{\textcelsius}
\psfrag{Vout}{$V_{out}$}
\includegraphics[width=\textwidth]{fig.eps}
\c |
|
n*****y 发帖数: 134 | 29 forget one thing,
what is the headroom of vout for CS and CG (if voltage supply rail is 0V and
Vdd)? |
|
m*****t 发帖数: 3477 | 30 ur output voltage is a time dependent (DC biasing is not.), so how can u
plot Vout vs Temp w/o time axis?
run temp paramedic analysis, given a fix time instant, plot the voltage
points, then do interpolation to get a curve. |
|
g*****d 发帖数: 210 | 31 条件很简单
Vcc=10V, Vin=0 or 5 V, Vout=0 or 10V
When input in 5V, the output should be 0V.
When input in 0V, the output should be 10V.
可是CMOS process 的BREAKDOWN 是5V
有没有人知道任何REFERENCE PAPER...
Thanks |
|
f*****0 发帖数: 489 | 32 the output of the opamp will be either +max if the voltage on the non-
inverting end is greater than that on the inverting end (in this case, it is
zero), or -max if otherwise.
the voltage on the non-inverting end is the weighted average of Vin and Vout
. so the opamp will flip to +max if Vin is such that it causes the weighted
average to be positive; or the opamp will flip to -max if the weighted
average goes negative.
this causes the system to have a small hysteria (thus noise resistance)
depe |
|
d****o 发帖数: 1112 | 33 what is your requirment?
Vin=?
Vout=?
Iout=?
usually switcher and mosfet driver already integrited together.
high |
|
ET 发帖数: 10701 | 34 咱这纯粹是向您学习来着。我在看tony的那部论文。
现在想明白的就是application和designer的如何结合起来的。 听起来你好像是做apps
的吧?比如就这个buck converter,这个topology已经很明白了,vin, vout, iin, in
out都有了,然后呢?
选了frequency, 然后挑个l & C? 具体是个啥步骤,你们做完后,就可以丢给designer
了? |
|
ET 发帖数: 10701 | 35 ltspice,probe vout,直接phase bode plot也出来了。 |
|
ET 发帖数: 10701 | 36 啥是跨导非线性?啥是输出导纳非线性?
我觉得这个问题,理想情况下,
Vout/Vin = -1 这是理想线性关系。
= gm1*1/gm2 (2 是cascode transistor, 1是input transistor)
但实际是gm1/(gm2+gmb)
gm1= w/l*kncox(vgs - vth1)
gm2 = w/lkncox(vg-vs-vth2)
对于transistor 1.. vs = vb = 0, normally so vth 不和vsb相关
对于transistor 2, vb = 0, vs 是第一级的output, 就是在bias point, vth2 是f(vs
b), 这就带来了所谓的sqrt(vsb)的非线性
假设输入信号是sin.. 即使transistor biasing在gm最平的地方, 第一级的输出也是个
放大的sin..第2级的vth 就是第一级输出的函数。
所以我认为第2级,casscode transitor带来的影响大。 |
|
ET 发帖数: 10701 | 37 不行,太笨了。
如果以附件中的m4的gate作为输入,图中的vout作为输出,
by any chance, dc gain可以表达为gm4*[(ro5+r07)/(1+gm7*ro7)]
这里的transistor尺寸都一样。。
我再怎么化,也至少是gm4*(gm7*ro7*ro1/(1+gm7*ro7)
thanks,, |
|
b********n 发帖数: 493 | 38 我认为算M4 gate端miller cap 的gain应该是从M4 gate 端到M4的drain 端,而不是到
M7的drain 端(即整体电路的Vout) |
|
ET 发帖数: 10701 | 39 还是不明白。
我知道反馈回来用来比较的是电压,即使sense的是inductor上的current, 通过一个i-
v转换,将v做为error amplifier的输入,与verr比较,verr是vout和vref在voltage m
ode下产生的。
verr & current*R, 再和saw tooth waveform, 输入comparator, 从而控制switching
transistor? 这弄出的是啥?
还是verr & current*R, 先比,弄出个vcontrol, 不过这个是啥意义呢?如果不是个c
omparator, 剩下的vcontrol就和voltage mode一样,用sawtooth waveform去modulate
d pulse width.
current mode有几种方式,似乎。 peak current是一种.
控制duty cycle和频率。 |
|
g******u 发帖数: 3060 | 40 我关心的问题是
(1)ic(t), ic(t)怎么就能变成peak current的值 - 还就是control的nature
try to find out later.
(2) comparator - 一般的comparator,
v(+) - (v-)> 0 , 输出1
v(+) - (v-)< 0 , 输出0
这里看来要求的是ic(t) - is(t) = 0, comparator输出change status.
It senses input current, if zero, means switch open, able to turn on
switch if necessary. Then if Vref>Vout, comparator provides grounded output,
when the SR latch sees a pulse from clock, it turns on the switch as the
load drops.
Quite simple.
(3) SR -latch这块。 s接pulse, r接 |
|
ET 发帖数: 10701 | 41 compensator是个OTA我明白,但从我贴的示意图上来看,它输出一个ic(t), 这个ic(t)
怎么就成了inductor上的peak current呢?这里面一定有个定量关系吧。
对buck来说,inductor上的current,在充电过程中,有slope = (v0-vg)/L, 在放电过
程中,有slope = v0/L,
同时,output voltage上也有ripple, 用linearwise piece来近似的话,应该是vout=v
dc + vac, 这个vac应该控制着ic(t) - 经过compensated的OTA.
我这个推导中,漏了2个地方,一个是inductor charge cap.得到vac,
如果这个vac的式子有了,我乘以Gm,也就能得到ic(t)了。
anything
. |
|
ET 发帖数: 10701 | 42
alittle
form
on
switch
膙oltage ->current 这我明白,i->v转化下也行。
compensator不是ota这块我不明白。那能叫做compensator的ota吗?
你说这compensator的目的是driver the error为zero - error = vout - vref吧?
为zero 意味着ic(t) = 0? - 那这岂非就不是peak current mode control了?
本来我还觉得我明白点啥了,这下更不明白了。
还有d>0.5时,我知道会有current offset, 你也提到了,但是是啥cause这个offset呢
? 因为imperfection of switches, inductor esr , etc. |
|
S****e 发帖数: 1063 | 43
offset呢
I don't know why you keep refering to OTA, and I run out of words to explain
it. however, you can refer to this app note for how a voltage error
amplifer (compensator) works. Vout-Vref=0 does not mean the output Ic(t)=0.
http://www.intersil.com/data/tb/tb417.pdf
If you take a look at the pdf i linked previously, it explains your question
regarding D>0.5 alot better than I can do in a post. |
|
g****t 发帖数: 31659 | 44 例如google: buck nonlinear
几十年来,实验报告非常多.仿真和理论就更多了.
所以我的建议是做个仿真看看问题出在哪.
常规的DC/DC分析,
必须假设电路各参数都变化不大.
不然稳态分析,小信号分析,假设状态法之类的
东西都会失效.
Vout=Vin*(1-D)之类的公式的推导都是假设参数
变化不大.
|
|
l***g 发帖数: 1035 | 45 pe guys.. have u controlled a boost with dsp? since the boost duty cycle is
not linear with the Vout, how do you close the loop? linearization or brutal
force pi? |
|
l***g 发帖数: 1035 | 46 thanks
right it's non linear so the dynamic response would be non linear. I have on
e implemented with PI only it is stable but I feel it is a bit slow. For thi
s controller I have a changing input (fast change) and a constant Vout. I wo
nder if linearation may help. I have a linearized model i'm going to try nex
t if time is allowed. I have to be very careful not to over voltage the outp
ut (clamping the duty) so my mosfet won't blow up... |
|
b**********g 发帖数: 141 | 47 比如说一个damped RC integrator,
R1 is between Vin and the negative input terminal of opamp.
R2 and C in paralled, between Vout and the negative input terminal of opamp.
请问R2的noise的contribution怎么算呢?
谢谢! |
|
a****l 发帖数: 8211 | 48 如果电压在中间(2.5)的时候两个管子都是截断的,那么中间vout不就是和两边断开来了
吗?怎么还会有电压?
PMOS and NMOS are in cut off mode.
still don't understand it, run the simulation by yourself, with NO load.
Then tell me what you get. |
|
r********r 发帖数: 677 | 49
本来上面的运放就已经是衡压放大器, Vout = Vin(1+R3/R4), 输出的电压基本不受负
载影响,你这样加了下面的OPA只起反效果。 如果电流不够的话应该加buffer。 |
|
h*****n 发帖数: 23 | 50 公司给了个小东西,其中要把传感器输出的0-5v DC相对于2.5v做翻转,即输出是Vout =
5v - Vin.
想用运放做个减法电路,但给的电源是单48v直流电源,经理也说最好不要加电路做成双
电源运放的减法器.
但是单电源减法器似乎在 vin跟5v很接近的范围内,输出的线性很差
请教一下,低成本的这样一个减法器有什么样的解决思路么?
电路新手真心请教,万分感谢 |
|