Date: Wed, 10 May 2023 19:11:07 -0400
From: "Stephen C. Menasian (menasian AT ptd DOT net) [via geda-user AT delorie DOT com]" <geda-user AT delorie DOT com>
To: geda-user AT delorie DOT com
Subject: Re: [geda-user] [off-topic] Large resistance value SMD resistors?
Message-ID: <20230510191107.190254b3@queeg.menasians.com>
In-Reply-To: <2c966cbc-4511-7b77-2a10-25e71cfd1f25@linetec.nl>
References: <3bfcfaef-a5f3-5e87-21f4-48a2f620a2a6 AT k4gvo DOT com>
	<6104f4a8-ad58-82a0-03eb-4e6c8da2ca82 AT k4gvo DOT com>
	<b32ca9a0-fe18-9041-6338-c3127bb313a7 AT linetec DOT nl>
	<CAOuGh89nK=94FKpG+V1QZCpuXNkH4rB+4qzA+NDtRtf1ToBJdw AT mail DOT gmail DOT com>
	<2c966cbc-4511-7b77-2a10-25e71cfd1f25 AT linetec DOT nl>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Reply-To: geda-user AT delorie DOT com
Errors-To: nobody AT delorie DOT com
Precedence: bulk

I don't know exactly what your circuit looks like; but you might consider a
capacitance multiplier circuit. A factor of 10, or even 100 might be
easy. If you have a spare FET input op amp, it could be, essentially,
free.

On Wed, 10 May 2023 23:07:44 +0200
"Richard Rasker (rasker AT linetec DOT nl) [via geda-user AT delorie DOT com]"
<geda-user AT delorie DOT com> wrote:

> Hi Bob, and Frank and Gabriel,
> 
> Thank you for your insights.
> 
> This part of the application is very simple: it's a linear audio 
> compression filter with a long release time constant (some 45 seconds), 
> with space constraints making smallish timing capacitors desirable.
> Foil capacitors are used for reasons of low leakage current and good 
> linearity, but these quickly become quite bulky with capacitance 
> exceeding 1 uF. Most MLCCs with X-type dielectric have serious 
> capacitance derating with increased voltage (15 volts maximum), so
> these were avoided.
> 
> However, from your answers I gather that the chosen solution with 22 M 
> resistors is also not optimal, and that potting or coating does not 
> guarantee trouble-free long-term operation.
> 
> Maybe the best solution here is to use high-capacitance MLCCs after all 
> with a maximum 1 M timing resistor, but only use a tiny charging
> voltage (e.g. 0.1 volts), and amplify this voltage with an extra opamp.
> 
> Best regards,
> 
> Richard
> 
> 
> Op 10-05-2023 om 21:17 schreef Bob Paddock (graceindustries AT gmail DOT com) 
> [via geda-user AT delorie DOT com]:
> > On Wed, May 10, 2023 at 7:45 AM Richard Rasker (rasker AT linetec DOT nl)
> > [via geda-user AT delorie DOT com] <geda-user AT delorie DOT com> wrote:
> >  
> >> ... 22 Meg ...
> >> Would it be better to use a larger resistor (e.g. 1206), or forego
> >> SMD altogether? I can of course also specify a moisture-resistant PCB
> >> coating, but that is a rather expensive extra production step.  
> > Coatings are not hermetic seals.
> >
> > They are good at keeping out particle contamination.
> > However moisture will still seep through over time.
> >
> > As it is now devoid of contamination it will never be noticeable in a
> > digital circuit, due to its high impedance;
> > perfectly pure water is an isolator.
> >
> > However in high impedances, or RF tuning circuits, the dielectric
> > changes. Perfectly tuned transmitters are returned in six months for
> > being off frequency. Been there, done that. :-(  Water ran off the
> > boards in removing the coating.
> >
> > This is why some high impedance related parts. such as Electrostatic
> > Grade Op-Amps,  say "expose to air" in their data sheets.
> >
> > As Frank pointed out, flux is an issue.
> >
> > We really try hard to avoid QFNs and anything that is hard to clean
> > under in our low current (uAs) standby products,
> > Any residual flux will build up contamination over time, which could
> > be measured in months or years.
> > A 10 uA poorly cleaned board built and measured today, might measure
> > 100 uA in six months.
> > Been there, done that too.
> >
> > If there are no other through-hole parts consider putting two 10 Megs
> > in series. To avoid the labor costs of the stuffing charges.
> >
> > Knowing more about the overall circuit may lead to better/different
> > answers. 
> Met vriendelijke groeten,
> 
> Richard Rasker
> -- 
> Linetec
> Akkerstafhof 15
> 7544SP  Enschede
> 053-4350834
> http://www.linetec.nl/
> 
>