UJT transistor and solar engine

[Dmitri]

Hello everyone!

Hoping the collective intelligence with bring an answer to the question
that's giving me hard time.
Scrambling for parts for a weekend project that involves a 'solar engine':
Small solar panel charges a capacitor, which then is being cyclically
discharged through a small DC motor. The scheme contains a UJT transistor
(2N2646 as per the design). This part is apparently very old and
discontinued everywhere I go. Does anyone know where to get a replacement?
Or, maybe someone knows a similar design that does not include UJT
transistors?
Any link, idea or reference will be greatly appreciated.

There is no sun here in PA at the moment, but maybe I can get it ready
before summer... ;-)

Thanks in advance for all responses!

--
Dmitri Abaimov
http://www.cabling-design.com
Cabling Forum, color codes, pinouts and other useful online resources for
premises wiring users and professionals
http://www.cabling-design.com/resources/documents/residential.html
Residential Cabling Guide 2003

 

[CFoley1064]

Subject: UJT transistor and solar engine

From: "Dmitri Cabling-Design.com " <info@REMOVE_NO_SPAM_cabling-design.com>
Date: 2/6/2004 12:40 PM Central Standard Time
Message-id: <[email protected]>

Hello everyone!

Hoping the collective intelligence with bring an answer to the question
that's giving me hard time.
Scrambling for parts for a weekend project that involves a 'solar engine':
Small solar panel charges a capacitor, which then is being cyclically
discharged through a small DC motor. The scheme contains a UJT transistor
(2N2646 as per the design). This part is apparently very old and
discontinued everywhere I go. Does anyone know where to get a replacement?
Or, maybe someone knows a similar design that does not include UJT
transistors?
Any link, idea or reference will be greatly appreciated.

There is no sun here in PA at the moment, but maybe I can get it ready
before summer... ;-)

Thanks in advance for all responses!

Try the NTE replacement P/N NTE6401

http://www.nteinc.com/

Good luck
Chris

 

[John Popelish]

Hello everyone!

Hoping the collective intelligence with bring an answer to the question
that's giving me hard time.
Scrambling for parts for a weekend project that involves a 'solar engine':
Small solar panel charges a capacitor, which then is being cyclically
discharged through a small DC motor. The scheme contains a UJT transistor
(2N2646 as per the design). This part is apparently very old and
discontinued everywhere I go. Does anyone know where to get a replacement?
Or, maybe someone knows a similar design that does not include UJT
transistors?
Any link, idea or reference will be greatly appreciated.

There is no sun here in PA at the moment, but maybe I can get it ready
before summer... ;-)

Thanks in advance for all responses!

UJTs are not very efficient switches. They have to pass current to
act as a voltage divider to detect the trip voltage and they have a
fairly large resistance when they are on. Give me some idea what
voltage you want to turn on the cap to motor connection, and how much
motor current you expect (or the motor resistance when stationary),
and I will try to come up with a blocking oscillator that uses more
available parts and maybe works better. Also, are you after the
highest possible peak torque or the most rotations under a low torque
load? In other words, what is the point of this circuit?

 

[Dmitri]

Hi John,

What I'm trying to accomplish is to have a small cart with (I suspect not
very efficient) DC motor that is connected through a pulley-belt to a wheel
move in a direct sun. The ratio between the pulley diameter and the wheel's
is about 1:10. The solar powered car has been bought at a craft store just
as a science toy for my son. We did find out right away that the direct
connection of the solar panel (about 1.1V at its peak) does not really move
the car (well, except for an exceptionally smooth declined surface ;-)) So,
the idea was to store the energy in a capacitor and discharge it through the
motor to make it move. I would guess the torque is of the highest importance
because the motor has to be able to start the movement, and the whole cart
with the panel, wheels and the motor is rather heavy.

Thanks for your input!

--
Dmitri Abaimov
http://www.cabling-design.com
Cabling Forum, color codes, pinouts and other useful online resources for
premises wiring users and professionals
http://www.cabling-design.com/resources/documents/residential.html
Residential Cabling Guide 2003

 

[John Popelish]

Hi John,

What I'm trying to accomplish is to have a small cart with (I suspect not
very efficient) DC motor that is connected through a pulley-belt to a wheel
move in a direct sun. The ratio between the pulley diameter and the wheel's
is about 1:10. The solar powered car has been bought at a craft store just
as a science toy for my son. We did find out right away that the direct
connection of the solar panel (about 1.1V at its peak) does not really move
the car (well, except for an exceptionally smooth declined surface ;-)) So,
the idea was to store the energy in a capacitor and discharge it through the
motor to make it move. I would guess the torque is of the highest importance
because the motor has to be able to start the movement, and the whole cart
with the panel, wheels and the motor is rather heavy.

Thanks for your input!

Thanks for those details. But I need to have a good idea of how high
the capacitor voltage will go if left connected to the solar cells
with no load, in order to design the trip circuit and power switch.
And can you power the motor with a couple D cells and measure how much
current it takes or measure its resistance?

 

[Dmitri]

Thanks, John.

I'm not sure about how to estimate the maximum voltage on the capacitor. I
guess, the motor would fry around 9V to 12V. So, the optimal voltage is
somewhere between 3V and 5V.

The resistance is about 1 Ohm, if my multimeter serves me right.

Thanks again!

--
Dmitri Abaimov
http://www.cabling-design.com
Residential cabling guide and other online resources for premises wiring
users and professionals.
http://www.cabling-design.com/resources/documents/residential.html
Residential Cabling Guide 2003

 

[John Popelish]

Thanks, John.

I'm not sure about how to estimate the maximum voltage on the capacitor. I
guess, the motor would fry around 9V to 12V. So, the optimal voltage is
somewhere between 3V and 5V.

Okay, I understand this is the motor voltage you would like to have.
But you can connect your multimeter to the solar cells, with no other
load and measure how high their voltage goes under no load. This
would be helpful information.

 

[Dmitri]

Hi John,
Thanks for being persistent with me!
The no-load voltage on the solar panel is 1.1V.

I the mean time I have purchased what I think is much better battery. I
think it was All Electronics. They had a 3.5V battery for $3.50, so I got
two. Don't know their short current, but no load voltage for those measured
4.2V in the full sun this weekend.

So, now I have a choice of solar panels available to me

I have yet to connect that better battery to the cart to see how it works,
but, even if it would drive the cart directly, I still want to go that
store-discharge route just for the fun of building it.

Thanks again!

--
--
Dmitri Abaimov
http://www.cabling-design.com
Cabling Forum, color codes, pinouts and other useful online resources for
premises wiring users and professionals
http://www.cabling-design.com/resources/documents/residential.html
Residential Cabling Guide 2003

 

[John Popelish]

Hi John,
Thanks for being persistent with me!
The no-load voltage on the solar panel is 1.1V.

I the mean time I have purchased what I think is much better battery. I
think it was All Electronics. They had a 3.5V battery for $3.50, so I got
two. Don't know their short current, but no load voltage for those measured
4.2V in the full sun this weekend.

So, now I have a choice of solar panels available to me

I have yet to connect that better battery to the cart to see how it works,
but, even if it would drive the cart directly, I still want to go that
store-discharge route just for the fun of building it.

Thanks again!

Now we are getting something to work with. If you connect the two 4.2
volt panels in series and load them down to about half voltage, you
are at about the highest energy (product of volts times amperes)
output they will deliver. Since this is still higher voltage than you
need to make the motor run, the best solution may be a little
switching converter than steps the voltage down while stepping the
current up. This may achieve continuous motion. I'll think a bit
about how this sort of thing might easily be done, and also how to go
about making a charge and dump circuit with your new cells. Give me a
day or two.

In the mean time, try connecting your two panels in parallel (for
higher current output ) ands see how they get along with the motor.

 

[John Popelish]

Hi John,
Thanks for being persistent with me!
The no-load voltage on the solar panel is 1.1V.

I the mean time I have purchased what I think is much better battery. I
think it was All Electronics. They had a 3.5V battery for $3.50, so I got
two. Don't know their short current, but no load voltage for those measured
4.2V in the full sun this weekend.

So, now I have a choice of solar panels available to me

I have yet to connect that better battery to the cart to see how it works,
but, even if it would drive the cart directly, I still want to go that
store-discharge route just for the fun of building it.

I am posting a charge and dump circuit on
alt.binaries.schematics.electronic.

If you don't have access to that group, let me know and I will email
it to you.

 

[John Popelish]

I am posting a charge and dump circuit on
alt.binaries.schematics.electronic.

If you don't have access to that group, let me know and I will email
it to you.

Here is the data sheet for the LM10 I used:
http://cache.national.com/ds/LM/LM10.pdf

 

[Dmitri]

Hi John,
Thanks for the design!
Pardon my ignorance, but what is the part that connects between ground and
the plus input of the left LM10? It does look like a battery, and it has
200mV next to it. What is it?

Thanks again!

--
Dmitri Abaimov
http://www.cabling-design.com
Cabling Forum, color codes, pinouts and other useful online resources for
premises wiring users and professionals
http://www.cabling-design.com/resources/documents/residential.html
Residential Cabling Guide 2003

 

[Robert C Monsen]

Hi John,
Thanks for the design!
Pardon my ignorance, but what is the part that connects between ground and
the plus input of the left LM10? It does look like a battery, and it has
200mV next to it. What is it?

Look at the datasheet. Its a built in 200mV reference.

 

[John Popelish]

Hi John,
Thanks for the design!
Pardon my ignorance, but what is the part that connects between ground and
the plus input of the left LM10? It does look like a battery, and it has
200mV next to it. What is it?

Thanks again!

The LM10 contains two opamps, with one internally connected to a 200
millivolt reference. That battery symbol represents the active
reference.

Take a look at the data sheet:
http://cache.national.com/ds/LM/LM10.pdf

By the way, I discovered a goof in the schematic. There needs to be a
resistor added between the opamp output and the PNP base, so that
there is more voltage swing at the output to drive the positive
feedback resistor. Something around 47 ohms might work. I had this
resistor in, then realized that the op amp current limit would keep
the base current about right and eliminated it (after designing the
feedback network). Sorry about that.

The opamp voltage swing is what makes the turn on and turn off
voltages different. The pot sets the turn on voltage, but the two
resistors that make up a voltage divider between the reference voltage
and the output of the opamp determine the difference between turn on
and turn off. Alter this divider to change how far down the cap
voltage goes before turn off.

I am thinking about another variation that is a switching regulator
that holds the solar cell array voltage at its maximum power point
while converting all available to drive the motor at whatever voltage
causes it to consume the available power. With your 9 volt array
loaded to about 4.5 volts, this may allow more continuous motion with
motor voltage varying between zero and 4.5 volts as needed to load the
array at its maximum power condition.