[ リスト | 詳細 ]



[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]

[ 次のページ ]

This desulfator makes pulses shown below and this sure can revive
sulfated batteries.

イメージ 1
p { margin-bottom: 0.25cm; line-height: 120%; }a:link { }
               Currentwaveform to a battery 40B. Y=2A/div, X=1us/div

イメージ 2
                         X: 5μsec/div. Cycle is about 50μsec with some jitter

イメージ 3
                            Run in part. negative peak -4.5A. X: 100nsec/div

イメージ 4
                               Current peak 6.6A  X: 200nsec/div
イメージ 5
  Modified AC adapter for a laptop

I happened have informed about a desulfator which de-sulfate
a car battery by a friend of mine. His purpose was to use aged
batteries that weren't good enough to use for automobiles any
more due to the increased internal ESR caused by the sulfation.
He has a home made solar system designed to work at 24V and
he uses 2 car batteries in series. His desulfator is almost the
same with this one explained here below.

The power source of above is the battery itself and a battery
charger is needed separately when to use because the battery
is discharged while desulfating and becomes empty as the
desulfation goes on.

I happened to find a patent below when searching the Net
and it explains about the actual frequencies and the current.

Patent: Pulse generation device for charging a valve-
regulated lead-acid battery


The important figures are claimed as below.

4. A pulse generation device as claimed in claim 2,
wherein said first frequency is in the range of 50 kHz
to150 kHz and said second frequency is in the range
of 10 Hz to 500 Hz.

5. A pulse generation device as claimed in claim 1,
wherein said pulses are square wave pulses.

6. A pulse generation device as claimed in claim 2,
wherein the first frequency pulses have an on time
in the range of 1 to 10 μs and an average current
in the range of 20mA to 60 mA.

When I was reading the contents, I got an idea to make a unique
desulfator which can charge and desulfate the battery at the
same time. My idea is to modify an AC adapter for a laptop PC.

Many of those AC adapter has a out put of DC 19-20V range
with a PFC circuit to meet the regulation of the power factor.
Most of those stop the PFC function when the load is small
and the deactivation is useful to get the second frequency
explained at the number 4.

Below is the result of 4 days after the desulfation was started.
I also took a photo just when started but I can't find it now.
I remember almost all the part was white then.

イメージ 6
                                             Day 4.

イメージ 7
                                Day 14.

 You can see how the sulfation was decreased. Actually the
battery became quite OK to start the engine easily.
Before the desulfation was done, no matter how I charges the
battery using a regular charger,it was impossible to start the

開く コメント(0)

開く トラックバック(0)

Mirage ECU problem

The photo below shows a part of failed ECU for a 96 Mirage 115000km.
It didn't communicate with a scanner.The car had an engine fluctuating
problem too.

イメージ 1

A local dealership manager asked me for a help sending me a
message below.

"We are having a hard time fixing the car. The ECU and the scanner
didn't communicate at all and we couldn't read the error code.
The engine fluctuated and we've tried our best, replacing a
PCV valve, cleaning the throttle body and changing the PCV valve.
But the problem persists and now we have no idea to go further.”

I'm just a DIYer and I can't make a big deal, but for some reason,
I understand some of the electronics.

When I examined the board,the ripple voltage at the Vcc lines, there
were spikes and ripples and it was natural that it couldn't communicate
with the scanner connected.

Looking at the mounted board,Nichicon's electrolytic capacitors
P series were used.  Although no leak could be visually confirmed,
when ESR was examined, one was completely dead and the other
showed high values. I removed the dead one and found the leaked
liquid was solidified on the board as shown in the first picture, and it
showed about 1kΩ/1mm distance by an ohm meter as below.

イメージ 2

The car now runs fine, but the problem of fluctuating idling has not
been solved. I had already done an experiment to find out the
cause of it using a home made smoke machine. It was caused by
a worn O ring used at the throttle shaft. The air was absorbed there.

開く コメント(0)

開く トラックバック(0)

L275S ミラの盗難警報が誤動作してお困りの方が多く、近くのショップ



The owner of the car below was upset due to the fact that the problem
was happened at a middle of night. It happened all of the sudden when
no one was there.

The problem was the malfunction of the security system and it
was not duplicated after that for months. But it happened again just
once recently and it was impossible to duplicate the malfunction.

イメージ 1

I was requested to help by a local car shop owner to solve the problem.
He has been trying to duplicate the problem and find the reason why
the malfunction was caused. He has needed to fix it finding the root
cause of it since the complain from his customer was very serious.
But his efforts so far have been not successful. Whatever he tried,
the result was just a vain. The occurrence was very seldom and could
not be duplicated. Then he decided to call me.

As he and I checked the Internet, there were so many posts about
the problem but we couldn't find any definitive answers or solutions.
You can see how there are so many malfunction problems in the fields.
Some owners were quite angry with the problem. Judging from the
announcement by the manufacturer listed at the bottom with *,
they too are struggling. Every post below is in Japanese since the
car is only sold in Japan. Use the Google Translation if you need to
understand the contents.



L275S ミラのセキュリティアラーム


純正セキュリティー 誤作動 L275S ミラ




* キーフリーシステム装着車をご使用のお客様へ

As I read above and hear from the shop owner, the malfunction tends to
happens when it's badly raining of after that. The rain water may be related
and doing something, I imagined.

There are 5 switches for the detection. Each door/gate has it's own switch.
One or more of them may be malfunctioning or the loom to those may be
leaking or the MPX body computer may be malfunctioning.

As I checked the internal circuits of the MPX body computer where the
security circuit was built in, I found that the sensing line B5 is pulled up
to 5.0V DC line via a diode with a 680 ohms series resistor. This pin is
the input pin for 5 switches. All those 5 are paralleled and connected
to here. I must say the line is rather high impedanced. High here means
that the current draw at the line is sensitive. A slight leakage may cause
the problem. Actually pouring small amount of water to the rear gate
switch causes the malfunction and  the problem is duplicated easily.

The switch is covered with a cushion rubber as is shown in the photos
below. It absorbs the water and the insulation between the switch
contact plate and the body became poor. In addition to that, the
material used for the contact is the beryllium copper and this makes
a battery with the base material zinc plated steel when the cushion
rubber is wet. As I check the voltage, it shows around 800mV.

Come to think of it, the beryllium copper's galvanic voltage is around
-0.2V and the zinc plated steel's -1.0V. It is natural these 2 metals
make the voltage difference of 800mV. The voltage affects the
sensitivity too. You can find galvanic voltages here.

イメージ 2

After the black cushion rubber is removed.
イメージ 3
The black one absorbs water very well and this makes a current bridge
between the contact and the zinc plated steel frame.

I don't say this is a design fault but it is not a good idea to use a
cushion rubber just above the switch for the rear gate.
The designer's intention must be to protect the switch contacts
against dusts and other foreign objects falling from above but
he should have to think about the side effects such as a moisture.
It sure causes the current leaking problem between the contact
and the body easily when the rubber is wet. In addition to that,
there makes a 800mV battery due to the Galvanic series and
this accelerates the problem.  I was easily duplicated problem
just poring water there. If I were him, I wouldn't design that way.
I'd use a moisture free switch such as a lead switch there to
prevent the malfunction of the security system especially when
rained or the humidity is high enough to have water condensations.
p { margin-bottom: 0.25cm; line-height: 120%; }a:link { }

開く コメント(0)

開く トラックバック(0)

Fixing an old tachometer

My local car shop owner asked me for a help. He was having a
problem with a non working tachometer used for a 93 Celica GT.
He has tried to replace the tachometer but it was totally
impossible to obtain any new or used one.

The car is 25 years old and the parts availability is now quite
low. But the car itself looks very new both inside and out.
The car was designed and made at the golden era of Toyota. 
The engine runs well too and he really wanted to fix the
tachometer expecting me to do it.

He quickly removed the cluster unit from the dashboard and
showed it to me although he was busy doing another project
when I visited.

Below is the rear view of it. The tachometer is held by 3
screws as is shown. Those 3 are also used for the terminals
to supply +12V, to connect to the ground and the RPM in signal
judging from traces and other components surrounded.
So removing those 3 screws makes it possible to remove
the tachometer unit from the cluster and check it at a bench.

イメージ 1

I connected the DC12V power supply to the cluster and applied
the 60Hz or 120Hz signals from a simple home made signal
generator. The reason why it is simple is that the signal
source is not generated inside but is getting the 60Hz energy
from the power company. It is consisting of a modified AC
adapter and a circuit to make square waves of 60Hz and
120Hz. A 60Hz square wave signal can drive the tachometer
to indicate 1800rpm and 120Hz, 3600rpm in case of a 4 cylinder
engine. Probably you may wonder why a power company's
60Hz can be used for the check of a tachometer. I will be
explaining it later.

As I checked the board behind the tachometer panel, there
were 2 electrolytic capacitors 10uF/25V. Those 2 had a
symbol mark of Matsushita. These days I have been
experiencing problems caused by aged Matsushita's
electrolytic capacitors. They were commonly dried up
badly. Of course electrolytic capacitors will dry up some
day in accordance with the Arrhenius equation, but
Matsushita's ones dry up quicker than other major
Japanese brands due to the reason that the rubber
becomes more brittle than others according to my
experience. Actually, I recently had a door's lock/unlock
problem on a 18 years old Subaru Pleo and it was caused
by old Matsushita's electrolytic capacitors. With these
my own experience, I dare removed those 2 capacitors
and replaced with new ones manufactured by Nippon
Chemion. Speaking of electrolytic capacitors, QAS capacitors
fabricated by Nichicon also end the life quicker than others
and auto mobile industries had some impact. One example is
the Celsior / LS400's ECU witten here below by a friend of mine.

We have to be careful for aged electrolytic capacitors especially
the one with the QAS liquid inside and some company's one with
a rubber which becomes very hard like a plastic and brittle as
are like those 2 used for a Celica's tachometer.

イメージ 2

As I checked those removed two below using an ESR meter,
one showed more than 600ohms and the other infinity.
These were surely dead or almost so.

イメージ 3

After replacing capacitors, I checked the tachometer
applying 60Hz and 120Hz attaching the signal generator.

イメージ 4

イメージ 5

After installing the cluster unit to the dashboard, the shop
owner said, “It's working perfectly. Thanks for the help and
I'll take you to a dinner appreciating your efforts”.
I am counting on what he'll serve me.

イメージ 6

イメージ 7

Regarding the home made signal generator, originally

it was a AC adapter of a 6VDC. I added 2 rectifying

diodes inside to get the half-wave rectified waveform

and the full-wave rectified waveform. Using a Schmitt

trigger circuit, those are converted to square waves of

60Hzand 120Hz. A  4cycles of 4 cylinders engine's

revolution 1800rpm means 30rps and the engine ignites

2 times at each revolution. This means that there are

60 ignitions in 1 second. This is the reason why a power

company's 60Hz energy can drive the tachometer

1800rpm and the twice frequency 120Hz, 3600rpm.

This simple method is enough for us to check the

tachometer without buying an expensive signal

generator. This certainly is a poor man's method

but quite effective to check a tachometer as well

as a speedometer since 60Hz drives the speedometer

85km/H and 120Hz, 170km/H.

p { margin-bottom: 0.25cm; line-height: 120%; }p { margin-bottom: 0.25cm; line-height: 120%; }

開く コメント(0)

開く トラックバック(0)

This time, I was requested to fix a NEC PC-LL750F26B. It has a
CPU Core i7-2670QM and the memory 8GB. It's not very new
but it's too early to discard. As you see the photo below, it says
Core i7.
イメージ 1
                                             The label says, "CORE i7".

I have once experienced a short circuited problem of a ceramic
capacitor used for a NEC laptop at here.

As I have written above, there were many similar problems among
NEC PCs in the market. When one of ceramic capacitors used
for the bypass purpose is short circuited internally, it is not easy
to find out which one is failed since there are many other ceramic
capacitors in parallel with it.

The last time when I fixed the lap top NEC PC-LS150BS6W, I only
needed to use a multimeter which can read down to 0.01Ω to
find out the exact short circuited one since it showed the
resistance 0.01Ω smaller than others and rather easy to distinguish.

This time the symptom was completely no power. Nothing has
happened when the power button was pushed on. No LED nor no
screen lit as if the AC adapter / battery was dead. As I checked
the inside using a multimeter, I found the DC line for the primary
side of the CPU power block and others showed only 0.3 Ω or so.

The photo below shows the resistance at the power supply line.
It is close to the connector where the external AC adapter is
connected. It showed 0.30Ω to 0.33 Ω depending on the
pressure to apply to the capacitor's terminals. When the
photo below was taken, I had to hold the camera by my right
hand and my left hand was barely holing the 2 test leads.
I was not able to give terminals enough pressure and the
read out showed 0.33Ω.
イメージ 2
                                Showed 0.33Ω when leads barely touched

The lowest resistance
0.27Ω was found at the primary side of
CPU's power circuit. There are 4 ceramic capacitors very close
to each other and all of them showed 0.27Ω.
The photo below
shows the location where the lowest resistance was confirmed
and I connected wires to supply the current using a low voltage
DC power supply unit. As you see below there are 4 capacitors near
the red and white leads are soldered. All these are in parallel.
イメージ 3

Connecting a low voltage power supply unit, I used an infra-red
thermometer which area was warmer than others. 
イメージ 8

イメージ 9

I found that the primary side of the CPU power supply was warmer
than others and this explained why the resistance was lowest there.
I also determined which one out of 4 is failed. The voltage is 0.332V
and this does not go through semiconductors but can go through
resistors. So we can find the semi short-circuited capacitor easily.
There still existed 0.27Ω and this should heat up the capacitor.
Suppose the current is 1A, theoretically the power consumed
by the capacitor is given as follows
 P=I X I X 0.27
For an example, when I is 1A,
 P=1A X 1A X 0.27Ω

The actual low voltage DC power unit is as below.
イメージ 4

イメージ 5

The voltage of the unit is 1.256V when the nob is set minimum
and is still high for semiconductors. So I added 2 resistors in
parallel and a schottkey diode to limit the current and the
voltage. The resistor I used was 1.35Ω(2 X 2.7Ω parallel) and
this  limits the current up to 0.93A (1.256V/1.35Ω=0.93A).
The schottkey diode is EC31QS03L(Io 3A, Vrrm 30V, Vf 0.45V max)
to limit the voltage. In my application, the actual clamped
voltage is around 0.33V or so as you can see in the photo above.

When the output is open, the voltage is limited up to 0.33V and
the maximum current is limited 0.948A theoretically by resistors.
The capacitor had remaining resistance of 0.27Ω and the
maximum current should be limited 1.256V/1.35Ω + 0.27Ω ≒ 0.78A.

Applying the current, I touched the second capacitor from the
right side by my index finger, I felt some extra heat and confirmed
it was the culprit. Replacing it fixed the problem. The removed
capacitor  showed the resistance 0.27Ω. I had no idea to know
the original capacitance but judging from the application,
I dare used a 1uF ceramic capacitor manufactured by a Japanese
company Murata which I trust a lot.
イメージ 6

イメージ 7

Now the PC powers up and works well as it used to be.

I think I need to explain the low voltage power unit a bit more.
In addition to the DC power supply, I used external resistors
1.35Ω(2 X 2.7Ω parallel) to limit the current. The schottkey
diode was EC31QS03L(Io 3A, Vrrm 30V, Max Vf 0.45V) aiming
to limit the voltage. Adding these parts allowed me to supply
the DC to the main board without damaging semiconductors.
This is based on the theory "Fermi level".

開く コメント(0)

開く トラックバック(0)


[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]

[ 次のページ ]


プライバシー -  利用規約 -  メディアステートメント -  ガイドライン -  順守事項 -  ご意見・ご要望 -  ヘルプ・お問い合わせ

Copyright (C) 2019 Yahoo Japan Corporation. All Rights Reserved.