last day (17 days later) » 

13:08
-2
A: How to achieve analog zero-drift sample and hold for hours?

Johny RadioSolution 1 If you know the leakage slope of your cap, you could repeatedly "top off" the cap at timed intervals, to compensate for droop. However, the slope is likely nonlinear, so the top-off amount would be nonlinear. It's possible the amount could be a simple percentage of the cap's charge l...

it would be helpful to know the reason for the downvote on this answer. This will be a more useful learning resource for future readers, if the impracticality of this answer is explained.
How would you go for recharging the capacitor? You're better served by using a uC then
is it acceptable on SE to combine multiple answers into one? How can people vote if multiple answers are combined? We won't know which solution for which they're voting.
@clabacchio looking forward to your explanation. Plz point me to the SE policy that says different solutions should be combined into a single answer. Thx
@JohnyRadio I'm sorry you got it this way, but the whole question/answer is becoming noisy and unuseful
the reason I've grouped our answers was because some of them were just abstract ideas and not really worth to be considered as answers
besides, some of them go against the constraints you put when you received the first answers from other users
I think that attitude won't get you much help and appreciation from other users
"attitude"?
13:19
"attitude"
which attitude?
"i appreciate the info in this answer, but i'm confused by all the upvotes-- it does not seem to contain an analog solution"
"This answer seems to get most votes for the fun photo. There are practical solutions getting pushed down by this answer due to a fun photo"
"
Hi Trevor, thx for info. But, the SE "what to ask" and "what not to ask" pages emphasize practicality. This answer does not seem to offer an actionable analog solution. "
"you don't offer any math or estimates on how long, or how much loss, relative to the size of the cap, so that's not so helpful."
But i should be ok when people attacked my question?
You disagree with those observations about those questions?
i mean those answer. My comments are accurate critiques of those answers.
I think that you asked a fair question, and it wasn't nice that you got 'bashed' regarding using a microcontroller
I understand it can be frustrating, and I've dealt often with this kind of reaction
Thx for that. But isn't critiquing answers acceptable here? Isn't that how we make them better answers?
13:24
but understand that answers are not just for you, and are meant to be useful to other people as well
if you add unstated constraints or 'undermine' otherwise valid answers because they don't fit your mindset, you discourage people from helping
Of course, the whole question answer page is not just for me, it's to be useful to others. Not saying it's about "me", i'm saying that a stronger answer is better for everyone. i noticed that, since my critiques, we've gotten better answers.
critiquing is acceptable of course
i added constraints to improve the specificity of my question, but i can see how that could seem unfair to answers already posted
and encouraged
but there are better ways and worse ways
to me, the message of the question as stated would be that the easiest solution goes through digital storage
point taken, i should improve my tone.
as stated currently?
13:28
sure it clarifies the reason why you don't want a uC
Trevor's mcu-free digital technique is top contender for my chosen answer.
Or Peter K's method of making a analog S&H work
anyways, if you talk about economically disadvantaged and no coding, why would you propose creating an ASIC?
You're correct, and so those answers can be downvoted. It's not against the rules to post bad answers-- voting system is for that.
Main reason i posted several methods was to get people's juicing flowing.
To show that, with some creativity, a MCU is not only option in universe
i think it worked-- answers started coming :)
@JohnyRadio you're inferring :)
perhaps :)
13:34
anyways, I followed Andy's suggestion and merged the answers, it's true that one can post multiple answers, and that you can answer your own question
but to do both with 5 answers is a bit over the top in my opinion
it's not bad per se indeed, but try to use some moderation
understood.
in the end, i'm glad we now have some viable answers. Peter K's answer is better because it show's how it can be done in analog, which is great from theoretical standpoint. Trevor's answer is better cuz simpler, more stable, more practical. Tough choice.
I'm glad you got a solution, I'd encourage you to post a new answer if/when you try to implement one
that would really stand well as an independent answer
New answer that replicates answers already given?
with practical evidence
yes I think it would be great
anyways, I've just read the 250ppm bit
it's 0.25%, it seems a little too much for a didactical example
not arguing, but consider it
you mean hard to do with analog S&H?
seems easy with ADC
13:40
I mean as analog S&H
and also as experiment
that accuracy is necessary only if you need it
otherwise it's just 'because'
for my use-case, i do need that precision
again, that is my opinion but could help you understand the criticism
k back to work
cheers
13:42
if you want we can talk again later
many thx

  last day (17 days later) »