DRACO Smart Telescope: 90mm, 300-Second Exposures

Tuesday, September 8, 2026

Written By

Richard Harris is the founder and editor-in-chief of ScopeTrader, with over 40 years of experience in observational astronomy and astrophotography. He serves as director of the Ozark Hills Observatory, where his research and imagery have been featured in scientific textbooks, academic publications, and NASA-affiliated planetariums. Among his theoretical contributions is a cosmological proposition known as The Harris Paradox, which explores deep-field observational symmetry and time-invariant...

DRACO Smart Telescope Is Trying to Fill a Real Gap

There has been a fairly obvious hole developing in the smart telescope market. At one end we have small, portable instruments in the 30 to 50 mm range. They are inexpensive, easy to carry and increasingly capable. At the other end are larger instruments such as the Celestron Origin, which bring substantially more aperture but also more size, weight and cost.

A 90 mm smart telescope weighing a little over 12 pounds lands in an interesting place between them.

Aperture still matters, and I think that sometimes gets lost in discussions about smart telescopes because we spend so much time talking about sensors, AI processing, focal ratios and software. Those things matter too, but software has not repealed optics.

At first glance, a 90 mm aperture has about 3.2 times the collecting area of a 50 mm aperture. Draco, however, is not an unobstructed 90 mm refractor. Its optical design contains a central secondary obstruction, so we need to account for that before treating it as though we have the full unobstructed collecting area of a 90 mm objective.

That does not somehow make the 90 mm aperture meaningless. There is still substantially more aperture here than in a 50 mm smart telescope, and the obstruction may turn out to be a perfectly reasonable engineering tradeoff for achieving this optical design in such a compact package. What I do not want to do is look at the number 90 and pretend Draco is optically equivalent to a 90 mm unobstructed refractor. It is not.

When I have one here, that is something I will be able to evaluate with actual data instead of doing astronomy with a calculator and a product page.

The 340mm Focal Length Is What Draco Actually Has

There is another specification being tossed around that I think deserves clarification.

Draco has a 340 mm optical focal length. DWARFLAB also talks about reaching the equivalent of 1200 mm, and this is where marketing and optics begin getting mixed together.

Draco does not have a 1200 mm focal length. It has a 340 mm focal length.

You can crop an image. You can use smaller pixels. You can digitally magnify it, drizzle it, resample it and use increasingly sophisticated image processing. All of those techniques can be useful. None of them changes the focal length of the telescope.

I can crop an image from my Takahashi and make Jupiter fill my computer monitor. I have not turned the Takahashi into the Hubble Space Telescope.

This matters because people are understandably looking at Draco and wondering whether this is finally a smart telescope for smaller galaxies, planetary nebulae and other objects that benefit from longer focal lengths. The 90 mm aperture helps, and the small pixels may provide useful sampling, but 340 mm remains a relatively short focal length.

That does not make Draco less capable. It simply means we should judge it for what it actually is rather than what a digital equivalent number suggests.

Fast Optics and 90mm of Aperture Make an Interesting Combination

The f/3.8 focal ratio gets my attention because I have spent years imaging with fast systems. My Takahashi FSQ 106 with its reducer gets into roughly this territory, and there is a reason I like using it. Fast optics can collect useful signal efficiently, and under a good sky that means getting more out of the limited number of hours we actually have available.

But focal ratio is not the entire story. Aperture, focal length, pixel size, image scale, seeing and the target itself all become part of the equation.

That is why I am more interested in Draco as a complete optical system than I am in any one number. Ninety millimeters, minus the effect of the central obstruction, operating at 340 mm and f/3.8 with 2.4 micron effective pixels in its binned deep sky mode could be a very productive combination.

We will know considerably more when we have real FITS files to look at.

SHO Is Real SHO, But It Is Still Not Mono

This is probably my biggest disappointment with Draco, although at this price point I also understand the decision.

The main imaging camera is still one-shot color.

Now, I want to be careful here because DWARFLAB has done something genuinely interesting with the SHO Edition. It includes two dual-narrowband filters: one passing Hα + O III, and another passing S II + O III. Draco automatically images through both filters in sequence and uses those datasets to separate the S II, Hα, and O III signals before mapping them into the familiar blue-and-gold Hubble Palette.

So yes, Draco really is collecting S II, Hα, and O III signal. This is more sophisticated than simply taking an ordinary dual-band image and applying an SHO color treatment to it.

But there is still an important distinction.

Draco is collecting those signals through a one-shot-color sensor and its Bayer matrix. A traditional SHO astrophotography system normally uses a monochrome sensor with separate S II, Hα, and O III filters. With a mono camera, every photosite is available to collect the photons transmitted by each selected filter. With a color camera, those photons still encounter the red, green, and blue filters of the Bayer array.

And that matters, particularly when you are chasing faint narrowband signal.

I would have loved to see DWARFLAB take the SHO Edition one step further and use a monochrome sensor. Imagine this same telescope, with its automated filter system and software, but with a cooled mono camera gathering S II, Hα, and O III. That would begin to blur the line between a smart telescope and a conventional automated narrowband astrophotography rig in a very serious way.

That is not what Draco is.

What DWARFLAB has built instead may actually be more interesting than I initially gave it credit for. It is an automated SHO system built around a color sensor, using sequential dual-band filters to obtain the three emission-line signals with essentially no intervention from the user.

The question is no longer whether Draco can produce SHO data. It can.

The question is how much sensitivity and signal quality are sacrificed by collecting that data through a Bayer color sensor rather than a monochrome one.

And that is something I very much want to test rather than assume.

The 300 Second Exposure Claim Is Where Things Get Interesting

DWARFLAB says Draco can take individual exposures as long as 300 seconds, and five minute exposures put us into a very different conversation than the short exposure live stacking most people associate with smart telescopes.

Draco uses a dedicated guide sensor to correct tracking errors and, more unusually, physically rotates the main imaging sensor to compensate for field rotation. I think the physical rotator may turn out to be one of Draco’s more important engineering features.

An alt azimuth telescope normally experiences field rotation as it tracks across the sky. Software can register a stack of short exposures afterward, but it cannot remove rotation that occurred during one long exposure. Draco attempts to solve that problem mechanically by rotating the imaging sensor while the exposure is being collected.

That is clever engineering if it works well.

The words ‘if it works well’ matter because five minutes is plenty of time for tracking errors, guiding errors and mechanical problems to show themselves. I am less interested in whether the app lets me select 300 seconds than I am in what the stars look like when those 300 seconds are finished.

Give me a five minute FITS file and let me zoom into the corners. That will answer the question pretty quickly.

The SHO Edition Is Interesting, Even With the Color Sensor

Despite my disappointment that Draco is not mono, the SHO Edition is still the model I wanted for ScopeTrader. And now that I understand exactly how DWARFLAB is doing it, I actually find the approach more interesting.

Draco automatically alternates between its H alpha plus O III and S II plus O III filters, collects the different emission-line data, separates those signals and handles the SHO mapping and composition automatically. According to DWARFLAB, the result is essentially one-tap Hubble Palette imaging.

Think about what that means for someone who has never done traditional SHO astrophotography. No filter wheel. No separate S II, H alpha and O III filters. No sequencing software. No deciding how many hours to allocate to each channel. No manually combining three monochrome datasets just to get to the point where you can begin processing the image. You point one telescope at the sky and Draco handles the acquisition sequence for you.

I have traditional mono systems with filter wheels and dedicated S II, H alpha and O III filters, so I am probably going to be harder on this feature than someone experiencing SHO imaging for the first time. I know what good mono narrowband data looks like, and I know how much signal a cooled monochrome camera can collect. But that is exactly why I find Draco interesting.

The fair question is not whether a $1,499 Draco can beat one of my traditional imaging systems with a premium telescope, cooled monochrome camera, filter wheel, Chroma filters, mount, guiding system and everything else attached to it. It almost certainly cannot, nor should anyone expect it to.

The question is how much of that experience DWARFLAB can deliver for $1,499 in one box.

If Draco can sit outside unattended, automatically collect legitimate S II, H alpha and O III signal, and hand a beginner a respectable Hubble Palette image without that person ever having to understand filter sequencing or channel combination, that is a pretty remarkable accomplishment regardless of what sensor sits behind the filters.

For the money, that proposition is very hard to ignore.

The Sample Images Look Good, But I Want to See Independent Data

DWARFLAB has published some attractive images from Draco. They certainly demonstrate what the company says the system can produce, and I have no reason at this point to say otherwise.

But manufacturer images are manufacturer images.

I feel the same way about telescope manufacturers, camera manufacturers and automobile companies. I do not judge a new car by the photograph in the brochure, and I am not going to judge an astrophotography system entirely by the images selected by the company selling it.

What I find unusual about the Draco launch is how much interest has developed while there remains so little independent hands on information. There are already enormous discussions taking place among smart telescope users, yet most of us are still discussing specifications, marketing material and manufacturer supplied images rather than data from production instruments.

That is why I think the jury is still out on Draco.

Not because I think something is wrong with it. I have no evidence of that.

It is simply too early to know.

Preorder Means Preorder

This is also worth keeping in mind if you are thinking about ordering one immediately.

Draco is currently a preorder product. DWARFLAB says September orders are expected to ship within roughly two months, and retailers are giving estimates that put actual customer deliveries later this year.

The lack of readily available testing units gives me some pause. Several people I know who would normally be candidates for early evaluation hardware have asked about Draco and have essentially been told there are no testing units available.

Maybe DWARFLAB simply wants every available telescope going toward production. Maybe review units will appear tomorrow. Maybe the final production hardware is still being refined. I do not know, and I do not think anyone outside DWARFLAB should pretend they do.

What I do know is that I would feel considerably more comfortable making conclusions about Draco after seeing a dozen production telescopes operating under real skies than I do looking at a product page.

That is not being negative. That is how I evaluate astronomy equipment.

For $1,499, Draco Could Still Be a Remarkable Value

After all those qualifications, I keep coming back to the price.

The Standard Draco is $1,399 with introductory pricing of $1,299, while the SHO version is $1,499. Consider what DWARFLAB is attempting to put into that price: a 90 mm f/3.8 optical system, cooled 50 megapixel sensor, dedicated guide camera, physical sensor rotator, motorized filter system, second wide field camera, automated calibration, 128 GB of storage, battery, computer, software, tripod and the mechanical system necessary to make all of it work together.

Try building a conventional astrophotography rig from individual components for $1,499 that checks all those boxes.

You are going to have an interesting afternoon.

That is why I can be skeptical about some of the marketing while simultaneously thinking Draco may be a terrific value. Those positions are not contradictory.

The specifications are impressive for the price. The concept makes sense. The sample images look promising. I like the engineering behind the sensor rotator and guiding system, and I am particularly interested in seeing what the SHO version can produce.

But Draco is not a mono system. Its 90 mm aperture includes a central obstruction. Its actual focal length is 340 mm, not 1200 mm. The telescope is currently being sold on preorder, and independent production hardware has not yet been widely tested.

Those are not reasons to dismiss Draco. They are reasons to wait for evidence before declaring victory.

What I Want to Find Out When Draco Arrives

When the ScopeTrader Draco arrives, I am not particularly interested in reproducing DWARFLAB’s marketing material. We already have that.

I want to see five minute subs. I want to inspect the stars at 100 percent and look at the corners. I want to see what the central obstruction does to the actual image and how the optical system handles stars. I want to know how well the guiding and mechanical sensor rotation work together over several hours. I want to measure the cooling performance, look at rejected frames and process the FITS files myself.

I especially want to take the SHO data apart and see what is actually there. I want to compare the separated channels with the kind of data I am accustomed to collecting through a cooled monochrome camera and dedicated narrowband filters. I already know Draco will not collect that data the same way. What I do not know is whether the result will be close enough that most people simply will not care.

And perhaps that is the most interesting question surrounding Draco.

I have astrophotography systems that cost many times what this telescope costs. I have large telescopes, cooled monochrome cameras, premium filters, filter wheels, guiding systems and serious mounts. I enjoy all of it, and Draco is not going to make the laws of optics disappear.

But if a $1,499 box can sit outside, find an object, guide itself, rotate its own sensor, cool its camera, collect five minute exposures, change narrowband filters, calibrate and stack the data and eventually hand me usable FITS files, then I am willing to pay attention.

For now, that is where I think Draco belongs. Somewhere between genuine technical promise and a very effective marketing campaign.

The specifications say DWARFLAB may have built something unusually capable for the money. The sample photographs suggest there is real potential here. What we do not have yet is enough independent evidence to know where the marketing ends and the telescope begins.

That is exactly what I intend to find out.