[sdiy] LTspice - exponential voltage curve

[Justin Owen]

Neil has a point - I could have phrased the question better.

In my defence, the questions I ask are generally about things I don't know or don't fully understand. Sometimes phrasing a question clearly relies on a better knowledge of the problem and it's variables than I actually have.

Now whether it required tutting at me quite so loudly is a different matter - but Neil answers a lot of my questions on this list, I'm not too big to admit that I've learnt from him but I am big enough to ignore it when he talks to me like some grumpy old man berating a naughty little boy. 

;)

x





-----Original Message-----
From: Neil Johnson [neil.johnson71 at gmail.com]
Received: 30.10.2013 09:48:43
To: Tom Wiltshire
Cc: Justin Owen; SDIY List
Subject: Re: [sdiy] LTspice - exponential voltage curve

No, Tom, it's not about LTSpice, you missed the point.

It's about asking questions about the problem to solve, which the
first post didn't, yet the second post did and gave sufficient
information for a back of envelope calculation to result in an answer
that Justin could then use to solve the real problem.  Coming up with
the answer did not require LTSpice at all, just a bit of mathematical
juggling, but that was only possible once Justin had described the
actual problem (fitting a curve through a specific set of points).

Neil

On 30 October 2013 00:10, Tom Wiltshire <tom at electricdruid.net> wrote:
> He did say "Taking my first few steps with LTspice". That includes doing maths with it. Mostly I don't understand what the hell even the interface is doing well enough. Getting the actual models to do what I want is a nice idea, but probably not happening without some pointing in the right the right direction. I think that's what he was after. It's not the world's most helpful software. On the plus side it's free, which is probably why we use it, like Eagle (similarly counter-intuitive and unhelpful).
>
> T.
>
>
> On 29 Oct 2013, at 16:31, Neil Johnson <neil.johnson71 at gmail.com> wrote:
>
>> Justin,
>>
>> In future it would REALLY help if you ask what you actually want.  In
>> this case it's an easy bit of mathematical manipulation to come up
>> with the right exponent formula to plot what you want:
>>
>> V=5*exp(3.466*(time-1))
>>
>> Neil
>>
>> On 29 October 2013 16:15, Justin Owen <juzowen at gmail.com> wrote:
>>> Still not having any luck with this.
>>>
>>> I've tried scaling the arb voltage based on a 5V ramp using V=exp(V(Ramp)/2.8)-1 where (Ramp) is a 5V Pulse that ramps up from 0V to 5V across 1 second.
>>>
>>> I've tried scaling exp() against time V=exp(1.7917*time)-1.0 and a variation on Neil's suggestion of V=exp(time/0.56)-1.
>>>
>>> These all give pretty much the same results - but none of them have the exact curve I'm looking for which is:
>>>
>>> 0s = 0V (OK 0V = 0.15625)
>>> 0.2s = 0.3125V
>>> 0.4s = 0.625V
>>> 0.6s = 1.25V
>>> 0.8s = 2.5V
>>> 1s = 5V
>>>
>>> Any other suggestions? Thanks.
>>>
>



--  
--
http://www.njohnson.co.uk