Welcome everybody! Here's Alex Jasmin. As you can see, I'm setting up a new
workbench for filming these videos. This actually is a computer desk that doubles
as an electronics workbench. The thing is, it should give me a good excuse to clean
up after each project as I sometime want to use the PC here. Now, I need a
soldering setup on that bench. So I went ahead and purchased an FX-8801
soldering iron from Hakko. This is a classic design with a lot of tips available for
it and a lot of third-party manufacturers have cloned it and make
compatible parts. Now, I didn't get the base station that the Hakko hand piece
normally plugs into. The thing is, newer Hakko base stations use push buttons for
setting the temperature and that seems a bit fiddly. So, in this video we're going
to build our own. I think we can make good use of the remaining space
available on that desk and put an analog slider instead of the push buttons
for setting the temperature. So let's see what we're going to work with here.
As you can see, on the side of the bench I've attached a small power strip and a USB
hub. These make it easy to plug in stuff and switch between various tools and
computer accessories. Now, the plan is to fit our soldering station in a small
project box like this one. We can then attach it to the side panel as well. So,
it will stay out of the way and we can just plug in our soldering iron when we
need it.
Now, let's take a look at how the iron itself is assembled. The soldering tip is held in
place by a metal sleeve. In order to change the tip,
we can simply unscrew that sleeve and slide off the tip from the heating element. If
we take the iron apart further, you can see the ceramic heating element has wires
for both the heater itself and a temperature sensor. These are joined to
the iron's lead on a small circuit board. If we want to replace the heating
element, we can undo these solder connections. This board also holds a
spring that connects the metal parts of the iron to earth ground. Finally, at the
end of the iron lead, you have a 6-pin DIN plug with pins for the
temperature sensor, ground and heating element.
Alright, now, let's go through the iron
specifications. The heater is supposed to draw 2.5A from a 26V supply
which would make the resistance 10.4Ω. Though, at room temperature, the
resistance is much lower between 2.5 and 3.5Ω. There, about 3.3
and that would make the cold current draw about 8A. Much higher than when it
heats up. Now, for the temperature sensor which is a thermistor, the resistance should
be between 42 and 58Ω. You can see ours measures about 48, which is good!
And lastly, the ground to tip resistance should be below 2Ω.
There, we have 0.3 which is good as well.
Lets now take a look at this YiHUA iron. Like many others, it is inspired by the Hakko
design and uses the same T18 style tips as the Hakko. But, you can see that many
parts are not compatible. The heater, which I've taken out, is rated at 24V/50W.
While the Hakko was 26V/65W. That may seem close enough but the YiHUA
temperature sensor measures below 1 ohm. This indicate that this is a thermocouple
instead of the thermistor we had on the Hakko. In fact, I've broken off part of the
the heater so you can see that thermocouple junction and heating coil. You
can also see that the plugs on both these iron are not the same. So, even though
I intend to use the genuine Hakko for my project, it shows that if you'd like to
use some third-party parts in your own, you have to do some research beforehand.
This soldering iron will normally be powered from an AC transformer but in
order to save space on the bench, I'd prefer to use a DC power brick instead
I can just attach the brick behind the bench so it won't get in the way.
Our iron is rated at 26V but I couldn't find a power brick that outputs 26V,
so we're going to use this 24V one that can output up to 72W of power.
I did the math and by lowering the voltage the power output goes from 65W down to 55W.
It's not perfect but it should still be usable. What's more concerning though is
that when the iron is cold it has a lower resistance and from this 24V
supply it will draw a humongous 230W. In fact, if I try to power the heater
directly from the supply, you'll see that the supply goes in over
current protection. An AC line transformer is more forgiving as its
output voltage drops on the load. But, this kind of supply just keeps shutting
off and the iron never gets hot.
So, until the iron heats up, we need to limit its
current draw. There are a few ways to do so. One method I considered, is to switch on
the heater in brief pulses. By doing so, the average power is reduced and we wont
overload the supply. As the iron heats up, we can increase the pulse width until
the heater's resistance is large enough that we can just apply a steady 24V to it.
Another method will be to use a switch mode converter. What's great about this is
that the output voltage could be higher than the supply's. That will make it
possible to use the iron at its rated 26V.
As with our previous solution, the switch in this circuit is pulse-width
controlled but the pulses are filtered out. That leaves a steady output voltage
a voltage that we will increase as the iron heats up and gains in the resistance
the voltage inverting topology that we have here, works in two steps. First the
switch is closed building current through an inductor. Then the switch opens, the
inductor wants to keep the built up current flowing and goes to the output
capacitor and diode. The cycle repeats continuously, in a controlled manner so
as to get the desired output voltage. A feature of this topology is that the
positive side of the output capacitor is connected to the negative supply that's
why the values in our graph are below zero. I prefer this solution to that
of switching the heater on and off directly. The great thing about it is
that it could be made to work with various heater and power supply voltages.
There is a missing piece to this though, we need a circuit that limits the
current by controlling the switch.
We'll look at this next.
The mc34063 is a switched-mode controller that can be used with various
DC to DC converter topologies. It is one of those jellybean parts that are easy
to obtain and favored by electronic hobbyists. Let's see if we can use it in
our project. A good place to start is with the mc34063 development aid. This is
an online tool that from a few given parameters can provide a circuit built
for that chip. Given that we are going with a switch mode converter, I've
changed my mind about the power brick we're going to use. This 130W Dell
laptop power supply will give us plenty of headroom in term of power. The output
voltage of this power supply is 19.5V, so we can start by
entering this as the converter input voltage in our tool. Next, the output is
-26V. That's because we want the voltage and routing topology
that allows the output voltage to range from 0V to -26V. Next,
for the output current, I put twitter's and 3500mA. One amp more
than what the heater is supposed to the draw. Which should give us some head room
Vripple is the output ripple voltage i.e. how many millivolts it is
acceptable for the output voltage to vary as the converter goes through its cycle.
For a resistive heater, that does not matter much. So, let's put 100mV
for now. Next is the frequency of operation, up to 100kHz.
I tried entering values frequency values here and found that at 55kHz, the
inductor used in the circuit is of 12µH.
I'll go with 55kHz then. Simply because that works with an inductor
that I already have. Now, let's press on calculate. We get a
warning that the peak current of about 17A exceeds the 1.5A limit. That's
because this tool assumes that we'll rely on a small transistor in the IC for
switching the inductor current. We can avoid that limit simply by using an
external power transistor instead. After we get rid of that warning, we get some
numbers and a circuit diagram. I'll print this out so we can have a closer look.
So, here's the circuit we get. You see there are a couple of problems with it though. First
is that the peak current to the switch is much higher than the chip internal
transistor can handle. We already got a warning about that and
we can fix it by using some kind of an external switch instead. Now, an important
thing to note about this circuit is that pin-4 of the chip, that is its negative
supply is tied to Vout that is -26V.
The chip is normally used with a positive output where the switch stops
turning on when the voltage is high enough. Here, we want the opposite. If the output
voltage is higher than expected, say at -5V, we need to switch more
current to the inductor until the output reaches -26V. So, the trick
here is that by tying pin-4 to Vout, the output feedback on pin-5 is seen as
going up as the output voltage goes down. That's how the chip can be used in the
voltage inverting topology. So, we have the chip's supply
connected between Vin and Vout, that is between 19.5V and
-26V which makes its supply voltage 45.5V.
But, if we look at the MC34063 datasheet, you see the chip supply can only go up
to 40V. So we have a problem. In order to work around this, we can
connect pin-4 to the actual circuit ground and use an external comparator to
drive pin-5 high when the output is too low or vice versa.
We need a comparator to control the temperature anyways and a typical IC
like the LM339 will have four in one package. So it shouldn't be much of an
issue. But, I would have preferred the website to warn us that we are exceeding the
chip supply voltage in any case. Here's the plug and the power supply we're
going to use. If I turned on the light, you can see there's a nice pilot light
on it. There are three contacts on the end: the outside of the barrel is the
negative, the inside of the barrel is the positive and the center pin we don't
have to use (it connects to an ID chip inside the power supply) I've got the
matching connector on eBay. If I plug that in, you can see that the negative
comes out on the pins on the side
and the positive is on the back. The center pin on the back again is for
charger identification and we won't have to use it will now look at the IQ and
Pat portion of the circuit between the power plug and disorder in Aaron
this was built on a piece of copper clad board using this carving technique the
board is divided into insulated pad and which the component leads are soldered I
first marked these with a sharpie then the copper layer scored using a cab I
tipped sky
you can use a multimeter to ensure there's no continuity between the pad
the first thing I placed on the board is the current sense with the store when
the mc34063 trees is a 300 millivolt drop across that resistor it starts
limiting the current we need this feature when the heater is cold and
draws more current that resistor was calculated to be 18 mediums
unfortunately I don't have any resistors in the order of mediums so I'd like to
make one of the ferries starts wire I'm still during school terminals to the
board 12 attach that wire I've taken these from a terminal sweep that I had
this is the kind of thing you use to attach wires on the back of a device
I then schooled in a length of night nam wire to make our 18 million resistor
with the 18 gauge nichrome wire that i have the length is a bit short and it
was difficult to get right
the second component I've installed is the injector
as we've seen this circuit calls for an injector of 12 micro henries I add an
injector of that value but we must also make sure that the saturation current
for the injector here 19 amps is higher than the peak current in or secrete
there 17 amps when the current to one injector exceeds the saturation current
it start behaving like a low value resistor and if that were to happen the
current to the MOSFET would beat Huai and it risked the measuring it another
parameter is the injector rated current I believe that at more than twice the
output current it should be sufficient next i insulting MOSFET were using as a
switch
you may have noticed I had to move some of these components after first ordering
them that's because I didn't leave enough space to fit these eight things I
believe I may also I've changed the value of his star and don't need the
MOSFET at that time that is a star's only day to ensure the MOSFET is off
should that gate wire be left unconnected I chose one of the p-channel
MOSFET that I have and channel fats may be more efficient but we'd need a higher
voltage rail to drive one if we look at some of the main points in the datasheet
the description says suitable for switch mode power supplies which is a good sign
we can put twenty seven amps weight of nineteen amps at 100 degrees C our power
supply can only output six point seven and so we clearly won't be exceeding
that the post current can go up to 108 amps which is also much higher than the
17 amps peak we have to deal with the MOSFET also is rated at sixty volts we
have 45 volts between the input and output of our converter the MOSFET may
have slightly higher voltages across it even the energy spikes that are absorbed
in the output capacitors but I take a 60 watt MOSFET should be sufficient a bunch
of water parameters we can look at gate charge and CRSs give an idea of the
switching performance for instance but overall I think we can use this part
next I put this diode
it's a forward Schottky diode we really want to be using Schottky diodes in
switched-mode applications because they work at higher frequencies than your
regular silicon diode this one is a bit bigger than we need at 10 amps 100 volt
but I have a bunch of them so we're going to use it anyways
you may have noticed that these to eat sinks are most touching at this point
this is not really an issue though if you look at the diode you can see that
the mounting time that is attached to the eight things is electrically
connected to the diode cathode and for the MOSFET the tab is connected to the
drain if we look at the schematic you can see that the drain and the diode
cathode already connected to the PCB so the it sector chain really isn't an
issue
as for the capacitors you may be wondering why doubt three on each side
of the schematic but two pairs on the circuit board unless Lee I'm not too
sure myself I obviously got confused somewhere along the way I wish Alan
Turing various parameters in the design tool at one point it may have called for
two or three thousand microfarads of output capacitance but in the current
design we should have more than three thousands all things being equal the
output capacitance affects the output ripple voltage so I try entering various
Whipple values and now the corresponding capacitance we can see that at two
thousand microfarads we should have 169 millivolt of
peak-to-peak ripples of course the iron eater normally runs on easy and it will
work just the same with ripples on it supply that said the poles can cause as
you interference so it may be desirable to minimize them I'm not sure where I'll
go with this I think I'll leave it as is for now and wait to see how it performs
on the bench the mg34 63 is now on the breadboard all the circuit board is
doing is to pass the voltage from the boys supply to the chip all that shows
here is the power connector and to decoupling caps we can ignore all the
rest for now on the breadboard we have the chip its current and voltage
feedback input I'll try to supply rate since this pin is tied to VCC
we don't have any current limit and the oscillator is running likewise the
internal switched is open for the maximum amount of time as the output
voltage appear to stays at zero
at atropos to the timing capacitor which is at this point in the circuit
you can see us a tooth wave as the oscillator is running i can assume when
you pass a gun trace this one shows the state of the antenna switch in
transistor when the transistor turns on it pulls down the voltage at this point
so you can see the switch turns on by de volta I got the timing type a star is
ramping up and it turns off why it's ramping down now we need to derive this
external MOSFET from the chip at a light switch we must take care not to turn on
the MOSFET for too long because past a certain point the injector is hard
living like a small value visitor and will get excessive current flowing
through I put another MOSFET on the breadboard it is the same type as this
one but connected to to power his stores if this MOSFET stays on for too long
we'll just dissipate more heat to this resistor and we won't be breaking
anything and this coke the magenta trace shows the voltage at the MOSFETs gate
and the cyan trace shows the voltage at the MOSFET drain this wire ties the
MOSFET to the positive supply rail keeping the MOSFET turned off if I
remove the wire you can see the MOSFET will be Tony on that's because the
MOSFET gate just shard to the probe which provide a path to ground if I
disconnect this Pro you can see I can turn the MOSFET off or on and why it's
floating the gate gives the charge gate capacitance can be an issue when we try
to drive a MOSFET real fast but I connect the gate of the MOSFET to the
collector the internal switch you can see that this register don't
provide enough current for the gate to discharge completely
and the MOSFET always stays on I put a voltage divider here so the gate doesn't
have to swing the all the way if I connect the gate at this point you can
see that we are not really better off so I also have a voltage buffer using to be
gt's if I connect the gate to the emitter here you can see finally that
the MOSFET is able to turn off there are some voltage spikes at the MOSFET drain
because these with the Stars have some inject into them so even this is a DC DC
converter of sort anyways the important point is that we are able to dry the
MOSFET with the circuit so we'll be able to use the real thing later on now let's
look at how we deal with the voltage feedback portion of the circuit as you
may recall the IC will enable the output when the voltage until 5 is too low now
we want the rivers because our supply as a negative output in order to achieve
that we are using an external comparator I'm using a 7805 here as I have voltage
reference and to power the comparator I see it may not be strictly necessary as
this would happily work at 19 point 5 volts but I plan to add additional Isis
that would require a 5 volt great anyways so I went with that
so again when the voltage feedback in the IC and 5 goes low the output will at
Ron I set up this voltage divider so that between five and minus twenty six
volt so with 34 volt across it this point is at 2.5 volts this way if the
output voltage which normally will be taken from here goes below minus 26 volt
this cause I and the switch won't but if we're above - 26 vote this which
would be turning on in order to test the circuit I'm feeding a senior from a
senior generator since my signal generator can't output a
DC offset of minus 26 world I've ever pulled my lab power supply which is
outputting about - Toni's exalt right now honey do put up the power supply is
enabled and it is below - 26 volts so you can see the output of the comparator
is I and the switch doesn't turn on if I move a voltage above 26 watts the output
of the comparator is low and this which is 20 on and off if I set that at 26 and
enabled signal generator now you can see that we get a square wave at the output
of the comparator if I show the signal we're feeding in this a sine wave with a
DC offset when the voltage goes below minus 26 world
the output is I when it is above the output is low now I also brought up the
oscillator away from if I sing a shot on this you can see that when the voltage
goes above - 26 volt during the ramp up portion of the cycle should i stir the
switches Toni on and it always turns off during the runtime portion of the cycle
there's a latching circuit that touches on the switch you can see it this
matching on here and are joined the other cycle it is undoing the old ramp
up portion because the voltage feedback output is low at the very start of that
portion you can see some other examples here
laughing under laughing under in reality the duty cycle will be fairly concerned
given a constant load but at least the show that this feedback circuit seems to
work perfectly now I completed the circuit by
connecting the MOSFET the current feedback and the voltage feedback all
the way they should go if I turn on the supply you can see the output quickly
goes to minus 26 volts on the scope you can see the switch turning on
sporadically to maintain the output voltage the yellow trace shows the
output of the comparator you can see that before it turns on there's a lot of
oscillation at the output I'm not too happy with that I've
experimented with putting a resistor between the output and the non-inverting
input of the comparator to add some hysteresis let me put back one leg
resistor you can see that even with the one leg resistor which is a relatively
large value the oscillation has completely disappeared so I'm happy with
that so there we have it it will be interesting to see if this oscillating
comparator has something to do with this myths about board layout and what are we
shall see that oscillation problem when we solder all of this up something else
to look at is the current limiting circuit
if I freeze the display you can see that the blue trace which is the oscillator
waveform goes a bit higher and voltage sometime after the switch turns on
that's because of the current limiting feature at the mc34063 from what I
understand of the application note when the voltage across essence goes above
0.3 volts the tiny capacitor is short at our voltage and it normally is that
caused a switch to turn off and to stay off for longer as the voltage has to go
down from that earlier point compared to there for instance I crossed a sense
with star when the switch turns off you can see an injective spike I don't know
if it's good to have that pin go much higher than VCC
I'll try adding a diode to clamp the voltage to VCC and we should see that
spike go down a bit and it does so with a Schottky diode on
the schematic I'm not sure we already need it but it won't hurt anything to
have it done if we look at the scope again I should
say didn't expect to see current limiting when there is no load though
that much maybe normal yeah I run this code right now if I connect it briefly
can see the current limiting really kicks in let's pause on this and
disconnect our iron you can see the current limiting actually seems to be
more aggressive as the drop across essence goes up there's an interesting
pattern here with a succession of small and large Peaks on the oscillator wave
form I'm not very sure why that is and there's a bit of a solution but you can
see that the equivalent here and arel about the same and that nan pose gives
the time for the Q on to the injector to die out and it can start flowing
normally again I always tried feeding power to the I
run of Chimaera since there's no temperature control I was using this
large copper clad board as I eat sink even so as you can see from the
discoloration I'm afraid it may be getting a bit too what what I'm going to
do here is to turn on the power supply just long enough for the voltage to
reach twenty six volts as you may recall the old point of having a current
limited supply was the that iron resistance increase with temperature
when the iron gets hot enough the current limiting should stop and we'll
see a twenty six volt output this tip thermometer will measure the temperature
at the tip of the iron you can see the voltage on the scope as well as on
Jasmin's meter and the current and that means the meter so let's get started
as if more than 400 degrees C 26
if we look back at the point where the voltage when Alyssa told at twenty five
point nine volt the tip temperature was already close to five hundred degrees C
and the ether must be the nutter cell that shows the current limit is clearly
too aggressive five hundred degrees above the normal cell during temperature
and the power output at this point is only forty seven point five watts for a
65 watt iron I addressed tell the equivalent sense with the star to be
closer to 16 video why it was eighteen before also we placed the output
capacitors by two 2,300 megawatts want so the lower resistance shall give us
the full twenty six output at a lower temperature and with the IR output
capacitance we shall see a bit less noise at the output it's on that
26 volts there you go
this time among the voltage picked up 279 degrees and the power output is of
60 watts not quite at the 65 watt this is
supposed to run up but I'm really happy with that I think we finally have a
decent power supply our soldering station well I think we finally have
something usable I shall see that setting that children since resistance
just right wasn't easy well don't just throw a few millimeters
of nichrome wire and start to bend this just the right way I will have to find
something better in the future also regarding the filter capacitors I looked
back that screencaps of the scope and it actually looks like I'm getting more
noise with the larger capacitors I got these caps at my local electronics store
and I don't know their specs but larger electrolytic caps are known not to work
very well at high frequencies that's why I generally better to use several
smaller caps in parallel to get lower yes I ran yourself so I may have to
experiment with that a bit I also forgot to mention that I lowered the value of
that resist or before we had to 1.5 8 kilo ohms when forming a voltage divider
but I found that we don't really need much resistance here I draw a red 2013
and that shouldn't change the voltage and add the MOSFETs gate which will in
turn get 30 beta at the SN by the way most of the resistance in the circuit oh
just what I happened to pick from an e 96 set and have a user and adult
critical in any cave why I'm sure that circuit could use more tweaking I can
only leave it on for so long or the soldering iron would just burn out so I
think we can leave the twist hole later this video is getting ready longer with
Z so I think this will become a video series next time we can look at
temperature control and then passion to use that little guy
here if you remember to set the soldering iron temperature thanks for
watching I hope to see you next time as we continue this project since this is
my first video I'd like to see your comments and suggestion and the youtube
comment section also please like this video and subscribe
For more infomation >> [SMB Game Builder] Build a Super Mario Bros Level in 10 Minutes! - Duration: 7:18. 
For more infomation >> France and Germany Plan to Build a Sixth Generation Super Fighter - Duration: 5:00. 
Không có nhận xét nào:
Đăng nhận xét