Quantcast

HamamatsuHam device adapter 1 ms exposure time restriction

classic Classic list List threaded Threaded
10 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

HamamatsuHam device adapter 1 ms exposure time restriction

Xue Wen Ng
Dear Personnel,

I have recently configured the HamamatsuHam device adapter into MicroManager to run our Hamamatsu ORCA Flash 4.0 V2 sCMOS camera. However, an issue of a 1 ms exposure time restriction on running the camera was encountered where exposure times below 1 ms were not achievable. This is unusual since the specifications of the ORCA Flash 4.0 V2 allows for faster readout. Is it possible that the 1 ms exposure time restriction was imposed by MicroManager? If so, is there a way to remove this restriction? I have also checked with Hamamatsu on this problem and they said that it is not an issue with their device adapter.

Your help will be greatly appreciated.

Thanks and best regards,

Xue Wen  


------------------------------------------------------------------------------

_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: HamamatsuHam device adapter 1 ms exposure time restriction

Nico Stuurman-2

Hi Xue Wen,


On 8/29/16 7:03 PM, Xue Wen Ng wrote:
Dear Personnel,

Not sure who that is, but I am happy to chime in;)


I have recently configured the HamamatsuHam device adapter into MicroManager to run our Hamamatsu ORCA Flash 4.0 V2 sCMOS camera. However, an issue of a 1 ms exposure time restriction on running the camera was encountered where exposure times below 1 ms were not achievable. This is unusual since the specifications of the ORCA Flash 4.0 V2 allows for faster readout. Is it possible that the 1 ms exposure time restriction was imposed by MicroManager? If so, is there a way to remove this restriction? I have also checked with Hamamatsu on this problem and they said that it is not an issue with their device adapter.

I can confirm the 1ms minimum exposure with the Flash 4.0.  However, this limit is not imposed by either the device adapter or Micro-Manager itself since the same version of the software happily sets the exposure time of our C9100-13 to values below 1 ms. So, this limit is somehow imposed by the interplay of the camera and device adapter code.  You may want to go back to your Hamamatsu contact and ask them to solve the issue for you.

Best,

Nico



------------------------------------------------------------------------------

_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: HamamatsuHam device adapter 1 ms exposure time restriction

Xue Wen Ng
Dear Nico,

Thanks a lot. I'll check with the distributor of Hamamatsu on this issue then. 

Best regards,

Xue Wen

On Aug 31, 2016, at 8:52 AM, Nico Stuurman <[hidden email]> wrote:

Hi Xue Wen,


On 8/29/16 7:03 PM, Xue Wen Ng wrote:
Dear Personnel,

Not sure who that is, but I am happy to chime in;)


I have recently configured the HamamatsuHam device adapter into MicroManager to run our Hamamatsu ORCA Flash 4.0 V2 sCMOS camera. However, an issue of a 1 ms exposure time restriction on running the camera was encountered where exposure times below 1 ms were not achievable. This is unusual since the specifications of the ORCA Flash 4.0 V2 allows for faster readout. Is it possible that the 1 ms exposure time restriction was imposed by MicroManager? If so, is there a way to remove this restriction? I have also checked with Hamamatsu on this problem and they said that it is not an issue with their device adapter.

I can confirm the 1ms minimum exposure with the Flash 4.0.  However, this limit is not imposed by either the device adapter or Micro-Manager itself since the same version of the software happily sets the exposure time of our C9100-13 to values below 1 ms. So, this limit is somehow imposed by the interplay of the camera and device adapter code.  You may want to go back to your Hamamatsu contact and ask them to solve the issue for you.

Best,

Nico


------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general

------------------------------------------------------------------------------

_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: HamamatsuHam device adapter 1 ms exposure time restriction

gwitz
In reply to this post by Xue Wen Ng
Hi Xue Wen,

Xue Wen Ng wrote
However,
an issue of a 1 ms exposure time restriction on running the camera was
encountered where exposure times below 1 ms were not achievable. This is
unusual since the specifications of the ORCA Flash 4.0 V2 allows for faster
readout.
The technical specifications of the ORCA Flash (https://www.hamamatsu.com/eu/en/community/life_science_camera/product/search/C11440-22CU/index.html) say that the minimal exposure time is 1ms unless you use it in Internal Trigger mode with a subarray. If you just used MM's live mode with the full chip, there's no way you are going to go below 1ms (live mode uses software triggering). The only way I can see to reach exposure times shorter than 1ms would be to do an MDA acquisition in sequence acquisition mode (set the delay shorter than exposure) and reduce acquisition to a subarray. Is this what you've already tried ?

Best,

Guillaume
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: HamamatsuHam device adapter 1 ms exposure time restriction

Nico Stuurman-2
On 9/1/16 12:25 AM, gwitz wrote:

> Xue Wen Ng wrote
>> However,an issue of a 1 ms exposure time restriction on running the camera was
>> encountered where exposure times below 1 ms were not achievable. This is
>> unusual since the specifications of the ORCA Flash 4.0 V2 allows for faster readout.
> The technical specifications of the ORCA Flash
> (https://www.hamamatsu.com/eu/en/community/life_science_camera/product/search/C11440-22CU/index.html)
> say that the minimal exposure time is 1ms unless you use it in Internal
> Trigger mode with a subarray. If you just used MM's live mode with the full
> chip, there's no way you are going to go below 1ms (live mode uses software
> triggering). The only way I can see to reach exposure times shorter than 1ms
> would be to do an MDA acquisition in sequence acquisition mode (set the
> delay shorter than exposure) and reduce acquisition to a subarray. Is this
> what you've already tried ?

Great point.  Just played around with internal trigger mode and
subarrays,  I was able to get it to down to 48 micro-seconds exposure
with a 8x8 subarray.  Pretty close to the listed 30 something
microseconds minimum exposure time.

Best,

Nico


------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: HamamatsuHam device adapter 1 ms exposure time restriction

Xue Wen Ng
Dear Guillaume and Nico,

Thanks a lot for the input. I have just tried out the internal trigger mode along with subarray acquisition (8x8) and it does bring the exposure time down to 49 microseconds as what Nico had pointed out. However, when I tried to acquire 1000 frames with an exposure time of 49 microseconds (8x8), the frame rate I obtained was 1315 fps (1000 frames/0.76 s total measurement time) at best. I think this could be due to the readout rate of the camera which restricted its acquisition time or perhaps the limitation of the PC? Is there a way to resolve this issue?

Best regards,

Xue Wen

On Fri, Sep 2, 2016 at 4:08 AM, Nico Stuurman <[hidden email]> wrote:
On 9/1/16 12:25 AM, gwitz wrote:
> Xue Wen Ng wrote
>> However,an issue of a 1 ms exposure time restriction on running the camera was
>> encountered where exposure times below 1 ms were not achievable. This is
>> unusual since the specifications of the ORCA Flash 4.0 V2 allows for faster readout.
> The technical specifications of the ORCA Flash
> (https://www.hamamatsu.com/eu/en/community/life_science_camera/product/search/C11440-22CU/index.html)
> say that the minimal exposure time is 1ms unless you use it in Internal
> Trigger mode with a subarray. If you just used MM's live mode with the full
> chip, there's no way you are going to go below 1ms (live mode uses software
> triggering). The only way I can see to reach exposure times shorter than 1ms
> would be to do an MDA acquisition in sequence acquisition mode (set the
> delay shorter than exposure) and reduce acquisition to a subarray. Is this
> what you've already tried ?

Great point.  Just played around with internal trigger mode and
subarrays,  I was able to get it to down to 48 micro-seconds exposure
with a 8x8 subarray.  Pretty close to the listed 30 something
microseconds minimum exposure time.

Best,

Nico


------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general


------------------------------------------------------------------------------

_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: HamamatsuHam device adapter 1 ms exposure time restriction

Tim Feinstein
Hi Xue, 

I also use a Flash4 v2 with Micro-Manager.  As you can see from the thread linked below, I am also not satisfied with the speed.  


As far as I can tell streaming mode simply does not work with the Flash4 v2.  There is no way to get the maximum theoretical frame rates with MM, or at least I have not heard about it.  I have discussed this with the engineers at Hamamatsu and I recommend that you contact them as well.  They have told me that the problem has to do with how Micro-Manager handles acquisition timing.  I like Hamamatsu and I like the Flash4 quite a lot, but I know people who chose the Andor sCMOS (no commercial interest) because they could not get their demo Flash to record fast enough.  

In the meantime you can still acquire at full speed using the software provided with your camera, HCImageLive.  You have to turn off Micro-Manager or at least use a configuration file without the Flash4, so that may or may not be a viable option.  

All the best, 


Tim

Timothy Feinstein, Ph.D. 
Research Scientist
University of Pittsburgh Department of Developmental Biology


From: Xue Wen Ng <[hidden email]>
Reply-To: Micro-Manager General <[hidden email]>
Date: Thursday, September 1, 2016 at 9:49 PM
To: Micro-Manager General <[hidden email]>
Subject: Re: [micro-manager-general] HamamatsuHam device adapter 1 ms exposure time restriction

Dear Guillaume and Nico,

Thanks a lot for the input. I have just tried out the internal trigger mode along with subarray acquisition (8x8) and it does bring the exposure time down to 49 microseconds as what Nico had pointed out. However, when I tried to acquire 1000 frames with an exposure time of 49 microseconds (8x8), the frame rate I obtained was 1315 fps (1000 frames/0.76 s total measurement time) at best. I think this could be due to the readout rate of the camera which restricted its acquisition time or perhaps the limitation of the PC? Is there a way to resolve this issue?

Best regards,

Xue Wen

On Fri, Sep 2, 2016 at 4:08 AM, Nico Stuurman <[hidden email]> wrote:
On 9/1/16 12:25 AM, gwitz wrote:
> Xue Wen Ng wrote
>> However,an issue of a 1 ms exposure time restriction on running the camera was
>> encountered where exposure times below 1 ms were not achievable. This is
>> unusual since the specifications of the ORCA Flash 4.0 V2 allows for faster readout.
> The technical specifications of the ORCA Flash
> (https://www.hamamatsu.com/eu/en/community/life_science_camera/product/search/C11440-22CU/index.html)
> say that the minimal exposure time is 1ms unless you use it in Internal
> Trigger mode with a subarray. If you just used MM's live mode with the full
> chip, there's no way you are going to go below 1ms (live mode uses software
> triggering). The only way I can see to reach exposure times shorter than 1ms
> would be to do an MDA acquisition in sequence acquisition mode (set the
> delay shorter than exposure) and reduce acquisition to a subarray. Is this
> what you've already tried ?

Great point.  Just played around with internal trigger mode and
subarrays,  I was able to get it to down to 48 micro-seconds exposure
with a 8x8 subarray.  Pretty close to the listed 30 something
microseconds minimum exposure time.

Best,

Nico


------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general


------------------------------------------------------------------------------

_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: HamamatsuHam device adapter 1 ms exposure time restriction

Nico Stuurman-2
In reply to this post by Xue Wen Ng
Hi Xue Wen,


On 9/1/2016 6:49 PM, Xue Wen Ng wrote:
> Thanks a lot for the input. I have just tried out the internal trigger
> mode along with subarray acquisition (8x8) and it does bring the
> exposure time down to 49 microseconds as what Nico had pointed out.
> However, when I tried to acquire 1000 frames with an exposure time of
> 49 microseconds (8x8), the frame rate I obtained was 1315 fps (1000
> frames/0.76 s total measurement time) at best. I think this could be
> due to the readout rate of the camera which restricted its acquisition
> time or perhaps the limitation of the PC? Is there a way to resolve
> this issue?

Those Micro-Manager time stamps may not be an issue. MM time-stamps the
images when they arrive in its memory space, not when they are actually
taken.  When using internal triggering, the internal clock of the camera
is used, which is very precise.  Hence, if no images were lost you can
be quite certain that the desired frame rate was achieved (however,
since MM is lagging, you can not do so forever, since one of the buffers
will give out and you will start loosing images).

To make sure this interpretation is correct, you should hook up a logic
analyze to the trigger output of the camera.  That should show you the
duration and number of the exposures as well as the time in between.  If
it ends up to be 1000 plus a few, within ~ 50 ms, then things look
good.  I like the logic analyzers from Saleae (https://www.saleae.com/ ,
no commercial interest, just a happy user), although you may want to
make sure that you use a model that is fast enough for the 48
microsecond exposures).

Best,

Nico



------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: HamamatsuHam device adapter 1 ms exposure time restriction

Michael Halter
This post has NOT been accepted by the mailing list yet.
Hello MicroManager Listserv-ers,

We are attempting to acquire images from from a Hamamatsu Flash 4.0 v.2 camera (each frame = 8 pixel x 8 pixel sub-array) at ~20,000 frames per second using MM (1.4.22). We appear to have a similar problem to that discussed on this thread. We know that the camera triggers at ~20,000 frames per second by monitoring camera trigger output. And, it is connected to a computer via a USB 3.0 connection so bandwidth should not cause a restriction. However, we know that the fastest frame rate we can acquire is ~1000 frames per second based on an FCS analysis I describe briefly below.
 
We are looking for any ideas/suggestions that might help us troubleshoot and reach an acquisition rate of 20,000 frames per second?  It seems possible. Or have we reached some limit?  Thank you in advance for any help/suggestions you can provide.

Here are some details about some data we have collected and the settings of the system:

We set the Exposure time = 0.1 milliseconds in the main MM GUI. And, the ROI size is set to 8 pixels x 8 pixels (16 bit depth).

In the Multi-Dimensional Acquisition window, Time points is activated with Interval = 0 and Number = 20,000 (Total memory = 2 MB for 8x8 pixel ROI). Multiple positions, Z-stacks, Autofocus, and Channels are deactivated. Save images is activated and Saving format is set to Image Stack file. This should allow for High Speed Acquisition, right?

We wanted to confirm that the camera is triggering a 20,000 frames per second so we set up an oscilloscope to measure an output trigger from the Flash 4.0. Indeed, the camera is triggering at ~20,000 frames per second:

Oscillosope image

However, the acquisition runs for ~12 seconds and collects 20,000 frames. No evidence of dropped frames, but the acquisition runs much longer than the expected 1 second to collect 20,000 frames. We were wondering whether the recorded frames were at 20,000 frames per second, or some slower rate.  
So, we performed an FCS experiment with fluorescent beads (0.21 micrometer) at 4 different acquisition times (0.1169 ms, 0.2046 ms, 1.0037 ms, and 2.046 ms). We then plotted the time dependent correlation curves based on 1) the exposure time interval and 2) the elapsed time reported in the MM metadata file.  These correlation curves suggest that the elapsed time stamp accurately reflects the time at which the frames was captured, because using the elapsed time yields consistent correlation curves:

FCS analysis of beads

The reported elapsed time suggests a max acquisition rate of ~1000 frames per second.
The computer we are using has the following properties:
CPU: Intel(R) Xeon(R) CPU X5650 @ 2.67GHz
Memory: DDR3 (666 MHz) 12GB
USB3.1 card (StarTech, PCIe x4)

We are stuck. Any idea about what we could change that might help us reach an acquisition rate of 20,000 frames per second?  Or have we reached some limit?  Any and all responses welcome!
 
Michael
N.I.S.T.
100 Bureau Drive, Stop 8313
Gaithersburg, MD 20899-8313


Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: HamamatsuHam device adapter 1 ms exposure time restriction

libisthanks_7
I think usb 3.0 might not fast enough for your application. You could consider the use of camlink card, it will definitely perform faster than 3.0. Also check the DCAM API driver to see if you installed the right one. I used 2.8 and 4.0 at the same time period, and miss matched the software driver and it caused very slow recording speed.
Loading...