(New) Contender at 3.33x? 128 f/2.4 HR Linos Inspec.x

JohnDownie · Post by **JohnDownie** » Tue May 21, 2019 1:01 pm

New? - not sure if it has been mentioned here before.

http://www.qioptiq-shop.com/en/Precisio ... l#tab_docs

“The inspec.x HR 2.4/128 3.33x lens was developed to achieve an object resolution up to 300 Lp/mm. This lens is optimized for a magnification factor of 3.33x and supports sensors up to 82 mm length.”

This is 50% higher (object side) resolution than the 105 f4 has.

Would require a double lightsaber tube.

And a fair amount of cash.

chris_ma · Post by **chris_ma** » Tue May 21, 2019 5:36 pm

seems to be new indeed, most announcements I could find were april 2019

couldn't find a tech document, but this page has an MTF comparison to the 105/4:
https://www.invision-news.de/allgemein/ ... ngsfrei/2/

here a copy in case it vanishes:

they mention the 128/2.4 being better in the lower paragraphs, but the MFT seem very similar. I guess the 104/4 has a slight advantage being measured at 3.5x vs 3.33x

EDIT: just realised they mention the graphs are for 72Lp/mm on the inspec.x HR 2.4/128 3.33x vs 50Lp/mm on the inspec.x L 4/105 3.5x prism

Lou Jost · Post by **Lou Jost** » Tue May 21, 2019 6:45 pm

That looks exciting. Note that the resolution figure given is the resolution on the object, not the sensor. The resolution on the sensor would be less than 100 lines/mm.

However, for a such a big image circle it makes sense to also consider the resolution in lines per picture width. That would be 7000-8000, similar to the Schneider blue ring 120mm Phase 1 macro lens

chris_ma · Post by **chris_ma** » Wed May 22, 2019 1:07 am

found the data sheet:
https://www.qioptiq-shop.com/out/Graphi ... 6239_0.pdf

but found no clue about the price, other then a mention that it features a "good price/performance ratio"

chris_ma · Post by **chris_ma** » Wed May 22, 2019 6:27 am

the graphs in that data sheet are quite spectacular - the lens actually performs significantly better at F2.4 then F2.8! (both being essentially diffraction limited)
(btw, the topic of this thread mentions the lens being 2.8, so maybe it's possible to fix that, and to add the HR while at it)

the light fall off (or lack of it) is also incredible.

I hope they make a version at 1x and 2x and keep the price under control - would probably be a contender to the printing nikkors (or quite possibly exceed that)

chris

ps: the datasheet is dated end of feb 2019, and is labeled "Q Internal Use" - so seems to be pretty fresh indeed

ray_parkhurst · Post by **ray_parkhurst** » Wed May 22, 2019 11:36 am

Interesting lens, and great to see the data sheet since it answered my question about LoCA. Seems that the lens is an achromat, unlike the 105/4, which is apochromatic. It does look well-corrected though.

Does anyone know what industry requirement is driving the development of lenses at ~3.5x? Is there some need to resolve ~1um on subject, maybe for FPDs or something else?

This lens looks like almost a drop-in replacement/competition for the Rayfact 3.5x (128mm vs 127mm FL, both f2.4), so there must be a particular requirement driving this spec. I can certainly see the improvement in the Rayfact 3.5x vs the 105IXL3p5, so I assume this 128IXHR3p3 will be similar.

Looks like another lens I'll need to wait for surplus stocks to become available to be able to buy. I now have at least 3 such lenses on my list!

ray_parkhurst · Post by **ray_parkhurst** » Wed May 22, 2019 11:44 am

chris_ma wrote:the graphs in that data sheet are quite spectacular - the lens actually performs significantly better at F2.4 then F2.8

Yes, this is very encouraging. I could not see much improvement at f2.4 vs f2.8 on the Rayfact 3.5x, so this 128IXHR3p3 may outperform it. That would be pretty cool to see!

I'd love to see curves for the Chiopt f1.6 3.5x lens. That one could in theory be a world-beater.

RobertOToole · Post by **RobertOToole** » Sun Jun 02, 2019 7:33 pm

Some additional details from the press release that came out on May 5 or 6, 2019.

The inspec.x HR 2.4 / 128 3.33x achieves an extremely high resolution of up to 300 line pairs per mm in the object. The inspec.x HR offers a significantly higher resolution than the established series inspec.x L. The aperture of 2.4 is significantly larger than the predecessor models with a corresponding magnification. The very bright new HR lens enables short exposure times and very high speeds in fully automated industrial inspection applications.

Due to the high aperture of 2.4, the diffraction limit is shifted upwards, whereby the high resolution and a high light intensity can be realized. The HR lens is the first of the Linos Machine Vision Lenses range that supports image sizes larger than 1 pixel size up to 3.5 microns. Comparable inspec.xL lenses are optimized for 5 micro pixel size and theoretically achieve a maximum imaging performance at the sensor of 50Lp / mm with 38% contrast at a spectral distribution from 420 to 680nm. The inspec.x HR reaches these values even at 72Lp / mm over the entire field of 82mm.

Also some images of the competition from a Korean website:

Translation from Korean:

An exclusive product developed through a two-year consultation and development period
: High resolution lens for TFT pattern inspection (72LP/mm 35% or more)
: 82mm image Circle corresponding to 16k 5um sensor
: Color applies Apo-chromatic design for inspection
: BS type lens with coaxial lighting

[/i]

ray_parkhurst · Post by **ray_parkhurst** » Sun Jun 02, 2019 7:57 pm

The press release states the lens is apochromatic, but the CA curve shows that is is actually an achromat, or at least that it only has 2 focal crossings in the visible spectrum. Maybe it crosses again in IR, but if so I'd think they would say it was corrected there.

Has anyone researched the FEM or Zeiss offerings for line scan apps?

RobertOToole · Post by **RobertOToole** » Sun Jun 02, 2019 8:43 pm

ray_parkhurst wrote:The press release states the lens is apochromatic, but the CA curve shows that is is actually an achromat, or at least that it only has 2 focal crossings in the visible spectrum. Maybe it crosses again in IR, but if so I'd think they would say it was corrected there.

They do show LoCA graphs for the lenses and the Zeiss is much better corrected, the Linos is not as good.

Has anyone researched the FEM or Zeiss offerings for line scan apps?

The FEM is from Mejiro Genossen. (Mejiro was a Japanese company but was bought by a Chinese Corp)

http://cgi3.genossen.co.jp/

They dont list the 16K lens, but the Korean ad does mention:

"An exclusive product developed through a two-year consultation and development period"

So maybe it is an exclusive. There is zero info on the Zeiss lens so I am sure that is also the case maybe?

RobertOToole · Post by **RobertOToole** » Sun Jun 02, 2019 8:48 pm

For you Ray:

This is with the beamsplitter in place I assume since its labeled as such.

I assume the Linos error is in mm?

The Zeiss is really well corrected!

RobertOToole · Post by **RobertOToole** » Sun Jun 02, 2019 8:58 pm

For the record I did request a price for 1 pcs when the lens was released at the beginning of May and they never bothered to reply after two attempts.

I believe the sales force is on commission. I should change the email to

" 100 pcs "

I would probably get a response in minutes, even on a weekend.

(in my previous pre-photography life I was a sales manager for a few years so I'm a bit of an expert on how commission sales work)

Robert

ray_parkhurst · Post by **ray_parkhurst** » Mon Jun 03, 2019 6:38 am

RobertOToole wrote:
ray_parkhurst wrote:The press release states the lens is apochromatic, but the CA curve shows that is is actually an achromat, or at least that it only has 2 focal crossings in the visible spectrum. Maybe it crosses again in IR, but if so I'd think they would say it was corrected there.
They do show LoCA graphs for the lenses and the Zeiss is much better corrected, the Linos is not as good.

Has anyone researched the FEM or Zeiss offerings for line scan apps?
The FEM is from Mejiro Genossen. (Mejiro was a Japanese company but was bought by a Chinese Corp)

http://cgi3.genossen.co.jp/

They dont list the 16K lens, but the Korean ad does mention:

"An exclusive product developed through a two-year consultation and development period"

So maybe it is an exclusive. There is zero info on the Zeiss lens so I am sure that is also the case maybe?

Sounds like these are going to be unobtainium, at least for a long while until some TFT facilities start shutting down.

The Linos and Zeiss are not really comparable since they are not intended for the same magnification, and the aperture on the Linos is so much bigger. The correction on the Linos seems pretty good given the broad coverage and high mag.

ray_parkhurst · Post by **ray_parkhurst** » Mon Jun 03, 2019 8:30 am

RobertOToole wrote:For you Ray:

...

This is with the beamsplitter in place I assume since its labeled as such.

I assume the Linos error is in mm?

The Zeiss is really well corrected!

I didn't see this before my last reply. Thanks for sharing the LoCA graphs! I saw this for the Linos before and was not terribly impressed. 250um is a big focus error, though for most of the visible spectrum it's pretty good. The Zeiss looks great, a true apochromat and very tightly controlled.

RobertOToole · Post by **RobertOToole** » Wed Oct 09, 2019 10:56 am

RobertOToole wrote:For the record I did request a price for 1 pcs when the lens was released at the beginning of May and they never bothered to reply after two attempts.

I believe the sales force is on commission. I should change the email to

" 100 pcs "

I would probably get a response in minutes, even on a weekend.

(in my previous pre-photography life I was a sales manager for a few years so I'm a bit of an expert on how commission sales work)

Robert

Excelitas got back to me with a price on the Inspec.x HR 2.4/128. Delivery TBA.

Anyone want to take a guess first?

Robert