TS3002 Simulation Model

R

rickman

Guest
I'm looking at using this device as a low power VCO at 240 kHz. I would
like to model it along with some of the associated circuitry, but the
company doesn't provide a spice model. I was wondering if anyone knows
how this thing works internally enough that a model could be created.

http://touchstonesemi.com/products/ts3002

It's an interesting little gadget.

--

Rick
 
"rickman" <gnuarm@gmail.com> wrote in message
news:kfuf76$nm0$1@dont-email.me...
I'm looking at using this device as a low power VCO at 240 kHz. I
would like to model it along with some of the associated circuitry,
but the company doesn't provide a spice model. I was wondering if
anyone knows how this thing works internally enough that a model could
be created.

http://touchstonesemi.com/products/ts3002

It's an interesting little gadget.

--

Rick
Click to expand...
I recall Linear has something similar. But their models are custom
behavioral thingies.

Cheers
 
On 2/18/2013 9:18 PM, Martin Riddle wrote:
"rickman"<gnuarm@gmail.com> wrote in message
news:kfuf76$nm0$1@dont-email.me...
I'm looking at using this device as a low power VCO at 240 kHz. I
would like to model it along with some of the associated circuitry,
but the company doesn't provide a spice model. I was wondering if
anyone knows how this thing works internally enough that a model could
be created.

http://touchstonesemi.com/products/ts3002

It's an interesting little gadget.

--

Rick

I recall Linear has something similar. But their models are custom
behavioral thingies.
Click to expand...
Don't know about LT. I think this part may be a second source for a
part by Maxim or someone similar. I haven't found that part yet.

I am working on my own model, but LTspice isn't always easy to use.
Like many tools it is not easy to pick up again after you don't use it
for a year or two. I can't seem to get a hierarchical model to work
properly.

I'm just constructing a functional model which will charge and discharge
a cap from a pair of constant current sources as controlled by a FF
triggered from the voltage trip points. That sounds like what they are
doing based on the description in the data sheet. I'm not a proper
analog designer so I don't know how their current source works. It
seems to pass a current determined by the set resistor. I expect they
are using current mirrors to scale a current in the resistor to the
current to the cap. I don't think I need to simulate all the details.
I just want to get a warm fuzzy that the control loop will work ok. The
part I really have no idea about is how the resistors for the VCO input
interact with the current source. This may be important. We'll see...

I may have to "simulate" it on a PCB.

--

Rick
 
On 2/20/2013 6:01 PM, rickman wrote:
On 2/18/2013 9:18 PM, Martin Riddle wrote:
"rickman"<gnuarm@gmail.com> wrote in message
news:kfuf76$nm0$1@dont-email.me...
I'm looking at using this device as a low power VCO at 240 kHz. I
would like to model it along with some of the associated circuitry,
but the company doesn't provide a spice model. I was wondering if
anyone knows how this thing works internally enough that a model could
be created.

http://touchstonesemi.com/products/ts3002

It's an interesting little gadget.

--

Rick

I recall Linear has something similar. But their models are custom
behavioral thingies.

Don't know about LT. I think this part may be a second source for a part
by Maxim or someone similar. I haven't found that part yet.

I am working on my own model, but LTspice isn't always easy to use. Like
many tools it is not easy to pick up again after you don't use it for a
year or two. I can't seem to get a hierarchical model to work properly.

I'm just constructing a functional model which will charge and discharge
a cap from a pair of constant current sources as controlled by a FF
triggered from the voltage trip points. That sounds like what they are
doing based on the description in the data sheet. I'm not a proper
analog designer so I don't know how their current source works. It seems
to pass a current determined by the set resistor. I expect they are
using current mirrors to scale a current in the resistor to the current
to the cap. I don't think I need to simulate all the details. I just
want to get a warm fuzzy that the control loop will work ok. The part I
really have no idea about is how the resistors for the VCO input
interact with the current source. This may be important. We'll see...

I may have to "simulate" it on a PCB.
Click to expand...
I was kinda hoping some of the chip gurus here would offer some opinions
of how they thought this circuit might work. The company FAE has been
very helpful in running some tests for me to make the chip run at my
frequency and show that the VCO control will work. He hasn't been able
to answer any of my detailed questions on how to get a fine control
rather than a coarse adjustment. I also don't get any sense from him of
how the chip works on the RSET pin. He just doesn't answer those
questions and I don't want to bug him further. Its not like I'm going
to be buying a million of these things.

In reality I'd prefer to make my own VCXO, but I can't find a 240 kHz
crystal. Whatever, I think I can get this to work even if I have to
play with it on the workbench. I prefer to simulate and get it right
the first time as my eyes make it hard to solder the small stuff these
days.

--

Rick
 
On Thu, 21 Feb 2013 20:33:31 -0500, rickman <gnuarm@gmail.com> wrote:

I'm just constructing a functional model which will charge and discharge
a cap from a pair of constant current sources as controlled by a FF
triggered from the voltage trip points. That sounds like what they are
doing based on the description in the data sheet. I'm not a proper
analog designer so I don't know how their current source works. It seems
to pass a current determined by the set resistor. I expect they are
using current mirrors to scale a current in the resistor to the current
to the cap. I don't think I need to simulate all the details. I just
want to get a warm fuzzy that the control loop will work ok. The part I
really have no idea about is how the resistors for the VCO input
interact with the current source. This may be important. We'll see...

I may have to "simulate" it on a PCB.

I was kinda hoping some of the chip gurus here would offer some opinions
of how they thought this circuit might work. The company FAE has been
very helpful in running some tests for me to make the chip run at my
frequency and show that the VCO control will work. He hasn't been able
to answer any of my detailed questions on how to get a fine control
rather than a coarse adjustment. I also don't get any sense from him of
how the chip works on the RSET pin. He just doesn't answer those
questions and I don't want to bug him further. Its not like I'm going
to be buying a million of these things.

In reality I'd prefer to make my own VCXO, but I can't find a 240 kHz
crystal. Whatever, I think I can get this to work even if I have to
play with it on the workbench. I prefer to simulate and get it right
the first time as my eyes make it hard to solder the small stuff these
days.
Click to expand...
OK i found them here:

http://www.af4k.com/hc6ucrystals20000-21999khz.htm

and here:

http://www.golledge.com/pdf/products/xtl_sm/cc1v.pdf

Without even trying hard

?-)
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top