Author Topic: EM Drive Developments - related to space flight applications - Thread 4  (Read 927791 times)

Offline Silversheep2011

  • Member
  • Posts: 85
  • Austraila
  • Liked: 40
  • Likes Given: 257
For the diy's
Found this paper
Please have a look for interest sake, and I would make the same recommendations stated.
the need to place 'a gap at the bottom' of a rack to the first electronic amplifier component for increasing the free air movement and whether we could do the samething could apply to test setups involving magnetrons. There are also some do's and don't do that could be useful  as well

I'm of the thought, that targeting "thermal air eddy elimination"  will bring everybody a better and more accurate results in that desire to isolating out the direct EM Drive effect by removing or at least minimising and/or linearizing the thermal noise  component for easier isolation and removal. I like the quote earlier on a few pages back
"A surrogate to noise is the standard deviation or variance."
so its logical to minimise it.

Schlieren Optics is tell us its there... but that we just cannot see it yet...  [thermal air eddies and noise]
much like turbulence you experience in an air plane

One other thermal  management reference's pointed to using rectangular slotted holes as being better than round holes for purposes of easy  air flow that's why you find slots in more common use on computers.
and yet another promotes what is called 'pin finning' instead of normal  finning on Electronic chips

https://en.wikipedia.org/wiki/Thermal_management_of_electronic_devices_and_systems
and how about fin orientation vertical vs horizontal?

another one of those things  to ponder about

« Last Edit: 10/03/2015 01:53 PM by Silversheep2011 »

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2166
  • Liked: 2684
  • Likes Given: 1124
For the diy's
Found this paper
Please have a look for interest sake, and I would make the same recommendations stated.
the need to place 'a gap at the bottom' of a rack to the first electronic amplifier component for increasing the free air movement and whether we could do the samething could apply to test setups involving magnetrons. There are also some do's and don't do that could be useful  as well

I'm of the thought, that targeting "thermal air eddy elimination"  will bring everybody a better and more accurate results in that desire to isolating out the direct EM Drive effect by removing or at least minimising and/or linearizing the thermal noise  component for easier isolation and removal. I like the quote earlier on a few pages back
"A surrogate to noise is the standard deviation or variance."
so its logical to minimise it.

Schlieren Optics is tell us its there... but that we just cannot see it yet...  [thermal air eddies and noise]
much like turbulence you experience in an air plane

One other thermal  management reference's pointed to using rectangular slotted holes as being better than round holes for purposes of easy  air flow that's why you find slots in more common use on computers.
and yet another promotes what is called 'pin finning' instead of normal  finning on Electronic chips

https://en.wikipedia.org/wiki/Thermal_management_of_electronic_devices_and_systems
and how about fin orientation vertical vs horizontal?

another one of those things  to ponder about
Nice info! I have decided to construct a home-built Schlieren Optic system over the next several weeks to analyze the NSF-1701 assembly, under power and not on the balance beam. I will video this.

Online SeeShells

  • Senior Member
  • *****
  • Posts: 2380
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3109
  • Likes Given: 2671
For the diy's
Found this paper
Please have a look for interest sake, and I would make the same recommendations stated.
the need to place 'a gap at the bottom' of a rack to the first electronic amplifier component for increasing the free air movement and whether we could do the samething could apply to test setups involving magnetrons. There are also some do's and don't do that could be useful  as well

I'm of the thought, that targeting "thermal air eddy elimination"  will bring everybody a better and more accurate results in that desire to isolating out the direct EM Drive effect by removing or at least minimising and/or linearizing the thermal noise  component for easier isolation and removal. I like the quote earlier on a few pages back
"A surrogate to noise is the standard deviation or variance."
so its logical to minimise it.

Schlieren Optics is tell us its there... but that we just cannot see it yet...  [thermal air eddies and noise]
much like turbulence you experience in an air plane

One other thermal  management reference's pointed to using rectangular slotted holes as being better than round holes for purposes of easy  air flow that's why you find slots in more common use on computers.
and yet another promotes what is called 'pin finning' instead of normal  finning on Electronic chips

https://en.wikipedia.org/wiki/Thermal_management_of_electronic_devices_and_systems
and how about fin orientation vertical vs horizontal?

another one of those things  to ponder about
Nice info! I have decided to construct a home-built Schlieren Optic system over the next several weeks to analyze the NSF-1701 assembly, under power and not on the balance beam. I will video this.

Hook up those dual muffin fans like I suggested, if you wouldn't mind, blowing horizontally off from the top.
Shell

Offline Silversheep2011

  • Member
  • Posts: 85
  • Austraila
  • Liked: 40
  • Likes Given: 257
For the diy's
Found this paper
Please have a look for interest sake, and I would make the same recommendations stated.
the need to place 'a gap at the bottom' of a rack to the first electronic amplifier component for increasing the free air movement and whether we could do the samething could apply to test setups involving magnetrons. There are also some do's and don't do that could be useful  as well

I'm of the thought, that targeting "thermal air eddy elimination"  will bring everybody a better and more accurate results in that desire to isolating out the direct EM Drive effect by removing or at least minimising and/or linearizing the thermal noise  component for easier isolation and removal. I like the quote earlier on a few pages back
"A surrogate to noise is the standard deviation or variance."
so its logical to minimise it.

Schlieren Optics is tell us its there... but that we just cannot see it yet...  [thermal air eddies and noise]
much like turbulence you experience in an air plane

One other thermal  management reference's pointed to using rectangular slotted holes as being better than round holes for purposes of easy  air flow that's why you find slots in more common use on computers.
and yet another promotes what is called 'pin finning' instead of normal  finning on Electronic chips

https://en.wikipedia.org/wiki/Thermal_management_of_electronic_devices_and_systems
and how about fin orientation vertical vs horizontal?

another one of those things  to ponder about
Nice info! I have decided to construct a home-built Schlieren Optic system over the next several weeks to analyze the NSF-1701 assembly, under power and not on the balance beam. I will video this.

Hook up those dual muffin fans like I suggested, if you wouldn't mind, blowing horizontally off from the top.
Shell

I'm glad to see its of interest to both  seashells and rfmguy
 but  I can see possible trouble with that arrangement shells...

while I'm enjoying morning coffee, and looking over sketch's it appears there is the makings of an airplane wing here the same effect as blowing over top of a sheet of paper - could be mild lift but I'm sure its there.
So up for some ideas?
1. start by allowing some air space "under" the magnetron so that it disconnects it from the top plate sheet surface perhaps use some adjustable spacer legs
2. consider using  he balance beam to find a counter lift angle could be found by adding a few degrees say it works out at 1-2 of  down rake via the adjustable legs  to counter  balance out any lifting force being  generated by the cooling fans. looks like it could be a very much likely to be trial and error  approach for a while.
I do give credit though, blowing through sideways seems to be the best mounting direction for fans compared to top and bottom mounting fans.

there is another aspect to think about on this issue: all the heat has go 'somewhere' and that  'somewhere' has to be dissipated heat into air, there is simply no other path.

- its going to mean a lot of hot air going up, and a lot of cold air going down.-

Online SeeShells

  • Senior Member
  • *****
  • Posts: 2380
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3109
  • Likes Given: 2671
For the diy's
Found this paper
Please have a look for interest sake, and I would make the same recommendations stated.
the need to place 'a gap at the bottom' of a rack to the first electronic amplifier component for increasing the free air movement and whether we could do the samething could apply to test setups involving magnetrons. There are also some do's and don't do that could be useful  as well

I'm of the thought, that targeting "thermal air eddy elimination"  will bring everybody a better and more accurate results in that desire to isolating out the direct EM Drive effect by removing or at least minimising and/or linearizing the thermal noise  component for easier isolation and removal. I like the quote earlier on a few pages back
"A surrogate to noise is the standard deviation or variance."
so its logical to minimise it.

Schlieren Optics is tell us its there... but that we just cannot see it yet...  [thermal air eddies and noise]
much like turbulence you experience in an air plane

One other thermal  management reference's pointed to using rectangular slotted holes as being better than round holes for purposes of easy  air flow that's why you find slots in more common use on computers.
and yet another promotes what is called 'pin finning' instead of normal  finning on Electronic chips

https://en.wikipedia.org/wiki/Thermal_management_of_electronic_devices_and_systems
and how about fin orientation vertical vs horizontal?

another one of those things  to ponder about
Nice info! I have decided to construct a home-built Schlieren Optic system over the next several weeks to analyze the NSF-1701 assembly, under power and not on the balance beam. I will video this.

Hook up those dual muffin fans like I suggested, if you wouldn't mind, blowing horizontally off from the top.
Shell

I'm glad to see its of interest to both  seashells and rfmguy
 but  I can see possible trouble with that arrangement shells...

while I'm enjoying morning coffee, and looking over sketch's it appears there is the makings of an airplane wing here the same effect as blowing over top of a sheet of paper - could be mild lift but I'm sure its there.
So up for some ideas?
1. start by allowing some air space "under" the magnetron so that it disconnects it from the top plate sheet surface perhaps use some adjustable spacer legs
2. consider using  he balance beam to find a counter lift angle could be found by adding a few degrees say it works out at 1-2 of  down rake via the adjustable legs  to counter  balance out any lifting force being  generated by the cooling fans. looks like it could be a very much likely to be trial and error  approach for a while.
I do give credit though, blowing through sideways seems to be the best mounting direction for fans compared to top and bottom mounting fans.

there is another aspect to think about on this issue: all the heat has go 'somewhere' and that  'somewhere' has to be dissipated heat into air, there is simply no other path.

- its going to mean a lot of hot air going up, and a lot of cold air going down.-

The magnetron is rfmwguy's main issue with hot air eddies rising vertically up from the frustum. this creates a low pressure directly above the frustum and the air below the frustum gets pulled along the sidewalls to fill the void. Chaotic actions occur because of the pressure differentials between bottom and top.

By blowing the thermal heated air to the side it can be allowed to rise without adding a chaotic component to the frustum.

Offline Silversheep2011

  • Member
  • Posts: 85
  • Austraila
  • Liked: 40
  • Likes Given: 257
Quote:"The magnetron is rfmwguy's main issue with hot air eddies rising vertically up from the frustum. this creates a low pressure directly above the frustum and the air below the frustum gets pulled along the sidewalls to fill the void. Chaotic actions occur because of the pressure differentials between bottom and top.

By blowing the thermal heated air to the side it can be allowed to rise without adding a chaotic component to the frustum."


Seashells, Agree with comment- and on further thinking about it.  Wouldn't  it make sense to test  with the magnetron Off, and the cooling fans On first in the actual beam configuration as part of a  mock up in a calibrating pre-run? and  see what the numbers say?
if there was no noticeable movement up or down, or oscillations, with fans On only operation
you would have to say only then that's its onto a "good thing" to do as a standard part off the setup on EMD runs  [run cooling air sideways to magnetron with cooling fans],   if correct. And if not, well it means looking  for other alternatives or ways to mitigate those  thermal and air movement effects?

Online SeeShells

  • Senior Member
  • *****
  • Posts: 2380
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3109
  • Likes Given: 2671
Quote:"The magnetron is rfmwguy's main issue with hot air eddies rising vertically up from the frustum. this creates a low pressure directly above the frustum and the air below the frustum gets pulled along the sidewalls to fill the void. Chaotic actions occur because of the pressure differentials between bottom and top.

By blowing the thermal heated air to the side it can be allowed to rise without adding a chaotic component to the frustum."


Seashells, Agree with comment- and on further thinking about it.  Wouldn't  it make sense to test  with the magnetron Off, and the cooling fans On first in the actual beam configuration as part of a  mock up in a calibrating pre-run? and  see what the numbers say?
if there was no noticeable movement up or down, or oscillations, with fans On only operation
you would have to say only then that's its onto a "good thing" to do as a standard part off the setup on EMD runs  [run cooling air sideways to magnetron with cooling fans],   if correct. And if not, well it means looking  for other alternatives or ways to mitigate those  thermal and air movement effects?

Absolutely! You need a baseline test without the magnetron to characterize the effects of the fans. Then do your powered test and since you have the non-fan powered tests you would have simply more data to digest and hopefully the heat making the vertical chaotic component would be greatly negated by horizontally diverting it to the outside of the frustum. You should get clearer data.

Offline cl33250

  • Member
  • Posts: 3
  • Liked: 4
  • Likes Given: 0
Please don't waste time increasing airflow across the magnetron.  Pressure/lift is related to velocity squared, and the fans blowing any air around the magnetron will create chaotic air currents with orders of magnitude more force than relatively slow thermal air currents.

Online SeeShells

  • Senior Member
  • *****
  • Posts: 2380
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3109
  • Likes Given: 2671
Please don't waste time increasing airflow across the magnetron.  Pressure/lift is related to velocity squared, and the fans blowing any air around the magnetron will create chaotic air currents with orders of magnitude more force than relatively slow thermal air currents.
Thank you for your post. This is an open discussion and any thoughts on this are very welcome.

So what you're saying is the the hot vertical air rising from the magnetron cannot be diverted to the sides of the frustum changing chaotic air movements from a vertical component to a horizontal one that doesn't effect the vertical measurements as much? The will effect the vertical measurements now be more then just letting it rise as a convection current?

I also think there might be another factor, because the top of rfmwguy's frustum has sides and could also be trapping hot air generated from the frustum.


Shell

Edit: speeling
« Last Edit: 10/04/2015 03:21 PM by SeeShells »

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2166
  • Liked: 2684
  • Likes Given: 1124
Please don't waste time increasing airflow across the magnetron.  Pressure/lift is related to velocity squared, and the fans blowing any air around the magnetron will create chaotic air currents with orders of magnitude more force than relatively slow thermal air currents.
I would agree the horizontal fans may increase the chimney effect and create unwanted horizontal oscillations. Thermal plumes can only be controlled by ambient air temp equaling device temp...or vacuum testing.

This thought experiment has kept me up at nights...finally thought it best to not add anything and characterize lift, using dsoftware to "extract" displacement changes during mag ON state. Following an ON state, the lift rises a a fairly predictanbe rate with some minor variations as the heat plumes exit the frustum somewhat uniformly.

However, am curious enough to use Schleiren photography to see this for myself. Though my test stand is temporarily disassembled, can do a static power test and shoot some video later this year...if I get some donation help. The proper mirrors and light sources are not cheap.


Offline cl33250

  • Member
  • Posts: 3
  • Liked: 4
  • Likes Given: 0
In reply to shell:  Assymetric horizontal airflow creates lift.  Since the magnetron is not a symmetric airfoil, it will have chaotic, turbulent airflow around it at best.  Characterizing the resulting forces accurately enough to discern an underlying mN signal under it would be difficult.

Considering that, work to streamline the magnetron may help regulate the lift and drag from the convection currents. 

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2166
  • Liked: 2684
  • Likes Given: 1124
Disappointed to notice wallofwolfstreet has left the building here and on reddit apparently. Anyone have the details on this? I enjoyed his posts, even tho he was not a firm believer...

Online SeeShells

  • Senior Member
  • *****
  • Posts: 2380
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3109
  • Likes Given: 2671
In reply to shell:  Assymetric horizontal airflow creates lift.  Since the magnetron is not a symmetric airfoil, it will have chaotic, turbulent airflow around it at best.  Characterizing the resulting forces accurately enough to discern an underlying mN signal under it would be difficult.

Considering that, work to streamline the magnetron may help regulate the lift and drag from the convection currents.

Thanks, that helps me understand.

Offline graybeardsyseng

  • Full Member
  • *
  • Posts: 134
  • Texas, USA
  • Liked: 199
  • Likes Given: 773
Please don't waste time increasing airflow across the magnetron.  Pressure/lift is related to velocity squared, and the fans blowing any air around the magnetron will create chaotic air currents with orders of magnitude more force than relatively slow thermal air currents.
I would agree the horizontal fans may increase the chimney effect and create unwanted horizontal oscillations. Thermal plumes can only be controlled by ambient air temp equaling device temp...or vacuum testing.

This thought experiment has kept me up at nights...finally thought it best to not add anything and characterize lift, using dsoftware to "extract" displacement changes during mag ON state. Following an ON state, the lift rises a a fairly predictanbe rate with some minor variations as the heat plumes exit the frustum somewhat uniformly.

However, am curious enough to use Schleiren photography to see this for myself. Though my test stand is temporarily disassembled, can do a static power test and shoot some video later this year...if I get some donation help. The proper mirrors and light sources are not cheap.

I've been thinking about this quite a bit as well.   I agree that the use of active (fan) cooling is likely to introduce very chaotic flow and -at the level of signal we are looking for here - more noise than they eliminate.

Vacuum is one way of course - likely the best and as is typically for the best - the most expensive route.   One of the reasons I am strongly considering a rotary stage setup for my DIY configuration is that then the force vector being measured will be perpendicular to thermal effects (if designed carefully). 

Another approach which we touched on a few pages back was the second cavity arranged in such a way as to cancel (or nearly so ) thermal effects while adding "emdrive" effects.   While I was advocating the need for making the cavities as near identical as possible - in reality I was wrong and as was pointed out that really isn't necessary - just getting "close" will improve the S/N.   

Just random thoughts on a Sunday afternoon.
EMdrive - finally - microwaves are good for something other than heating ramen noodles and leftover pizza ;-)

Offline graybeardsyseng

  • Full Member
  • *
  • Posts: 134
  • Texas, USA
  • Liked: 199
  • Likes Given: 773
Disappointed to notice wallofwolfstreet has left the building here and on reddit apparently. Anyone have the details on this? I enjoyed his posts, even tho he was not a firm believer...

I have also been noticing a drop in post count here and reddit too - particularly from some of the more prolific contributors.   If I were a conspiracy oriented person - which I am not - I would wonder if it has anything to do with the recent reports with  more and better quality data saying SOMETHING is happening that needs looked at.

  Surely if the data were unequivocally indicating "nothing to see here, move along, we've all been wasting our time" some previous purveyors of preserving the status quo would be shouting that from the rooftops but instead I find myself hearing crickets from them. 

But likewise some of those who were either supportive of the concept or at least supportive of impartially investigation and welcoming of data of all sorts seemed to have dropped of in volume and scope of their posts, not everyone by any means - but some real luminaries none the less.

If it were August I would suspect that lots of folks were just taking a late summer vacation or if September I would suspect many are just getting down to the new semester at their institutions.   

If this was a Hollywood conspiracy theory movie I would be suspecting the hero was going to find these missing voices were part of a vast group of commercial/government/academic investigators who had found something really significant and had been told to be quiet. And of course he/she would have to expose them and the secret and get the girl/boy and live happy ever after (perhaps on Pluto heh heh).

But of course that's just crazy talk - so I suspect they have all been off waiting in line to see "The Martian" in 3D.   (planning to myself RSN). 

Just some weirder thoughts on a Sunday afternoon - after an adult beverage or two.

Herman
EMdrive - finally - microwaves are good for something other than heating ramen noodles and leftover pizza ;-)

Online SeeShells

  • Senior Member
  • *****
  • Posts: 2380
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3109
  • Likes Given: 2671
Please don't waste time increasing airflow across the magnetron.  Pressure/lift is related to velocity squared, and the fans blowing any air around the magnetron will create chaotic air currents with orders of magnitude more force than relatively slow thermal air currents.
I would agree the horizontal fans may increase the chimney effect and create unwanted horizontal oscillations. Thermal plumes can only be controlled by ambient air temp equaling device temp...or vacuum testing.

This thought experiment has kept me up at nights...finally thought it best to not add anything and characterize lift, using dsoftware to "extract" displacement changes during mag ON state. Following an ON state, the lift rises a a fairly predictanbe rate with some minor variations as the heat plumes exit the frustum somewhat uniformly.

However, am curious enough to use Schleiren photography to see this for myself. Though my test stand is temporarily disassembled, can do a static power test and shoot some video later this year...if I get some donation help. The proper mirrors and light sources are not cheap.

I've been thinking about this quite a bit as well.   I agree that the use of active (fan) cooling is likely to introduce very chaotic flow and -at the level of signal we are looking for here - more noise than they eliminate.

Vacuum is one way of course - likely the best and as is typically for the best - the most expensive route.   One of the reasons I am strongly considering a rotary stage setup for my DIY configuration is that then the force vector being measured will be perpendicular to thermal effects (if designed carefully). 

Another approach which we touched on a few pages back was the second cavity arranged in such a way as to cancel (or nearly so ) thermal effects while adding "emdrive" effects.   While I was advocating the need for making the cavities as near identical as possible - in reality I was wrong and as was pointed out that really isn't necessary - just getting "close" will improve the S/N.   

Just random thoughts on a Sunday afternoon.

You know another effect you could see is when you turn one 180 from the other and run.

Shell

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2166
  • Liked: 2684
  • Likes Given: 1124
NSF-1701 Paper Update. With thanks to many, I am releasing my paper a day early. I look forward to your commentary.

All the best,
Dave

Offline jstepp590

  • Member
  • Posts: 13
  • Albany NY
  • Liked: 16
  • Likes Given: 35
NSF-1701 Paper Update. With thanks to many, I am releasing my paper a day early. I look forward to your commentary.

All the best,
Dave

Long time lurker here. I have been following the forum avidly for months, and I have donated to Shells experimental setup. I am so far out of my depth it is ridiculous, but I have learned so much and thank you. The reason why I am writing has to do with rfmwguy's funding. I noticed that you posted your results on YouTube and they're great. However, I believe what could help with your next project and also Shells is that, when posting onto YouTube to make a couple of changes.
The first is to list that these are tests of the EmDrive in the title, and also to add advertising like most of the other video posters. The reason to add EmDrive to the title of your YouTube posts is that very few people are going to look for NSF-1701 and Electromagnetic Engine Experiment as they have no idea what that is. People look for keywords and they will be looking for EmDrive so you are probably missing a lot of results and searches. With more search hits you will make more money from advertising, which will help fund your good work. Just what I hope is a helpful suggestion.  :)

Back to lurking...

Offline X_RaY

  • Full Member
  • ****
  • Posts: 832
  • Germany
  • Liked: 1100
  • Likes Given: 2367
NSF-1701 Paper Update. With thanks to many, I am releasing my paper a day early. I look forward to your commentary.

All the best,
Dave
Can't agree (or at least I am not sure about) your Q definition is really helpful, I think it leads to some confusion about it, S11 and S12 measurements are well defined IMHO.
Nevertheless congratulations to your revealing test report. I like your professional work.
And i wish you good luck for your next project in 2016! :)
« Last Edit: 10/04/2015 08:27 PM by X_RaY »

Online tchernik

  • Full Member
  • **
  • Posts: 273
  • Liked: 315
  • Likes Given: 638
Disappointed to notice wallofwolfstreet has left the building here and on reddit apparently. Anyone have the details on this? I enjoyed his posts, even tho he was not a firm believer...

I have also been noticing a drop in post count here and reddit too - particularly from some of the more prolific contributors.   If I were a conspiracy oriented person - which I am not - I would wonder if it has anything to do with the recent reports with  more and better quality data saying SOMETHING is happening that needs looked at.

  Surely if the data were unequivocally indicating "nothing to see here, move along, we've all been wasting our time" some previous purveyors of preserving the status quo would be shouting that from the rooftops but instead I find myself hearing crickets from them. 

But likewise some of those who were either supportive of the concept or at least supportive of impartially investigation and welcoming of data of all sorts seemed to have dropped of in volume and scope of their posts, not everyone by any means - but some real luminaries none the less.

If it were August I would suspect that lots of folks were just taking a late summer vacation or if September I would suspect many are just getting down to the new semester at their institutions.   

If this was a Hollywood conspiracy theory movie I would be suspecting the hero was going to find these missing voices were part of a vast group of commercial/government/academic investigators who had found something really significant and had been told to be quiet. And of course he/she would have to expose them and the secret and get the girl/boy and live happy ever after (perhaps on Pluto heh heh).

But of course that's just crazy talk - so I suspect they have all been off waiting in line to see "The Martian" in 3D.   (planning to myself RSN). 

Just some weirder thoughts on a Sunday afternoon - after an adult beverage or two.

Herman

I'm really wishing you are right and this is the calm before the storm, and for EagleWorks or some other important lab or agency to come with a big news splasher around the holidays (for making them even better).

But I guess people are simply taking some time for themselves after the frenzy of the latter months. Some may be back if we do get some big news or not, but some others may simply be moving on.
« Last Edit: 10/04/2015 08:17 PM by tchernik »

Tags: