BMP180 sensor for RPi – It’s not broken

Back in the heady days of being in the Royal Australian Navy, I did a few years as a meteorological observer. You do it for a while and it never goes away. So I still watch the clouds and the forecasts. I have two Raspberry Pi (RPi) computers and I thought it about time I did something with my B3+. Why not make it into a environmental sensor so I can look at what is happening when my astrophotography stuff is outside.

So I wandered down to Jaycar and spent up big getting two sensors: the XC4520 (temp and humidity) and a XC3702 (barometric and temp again) which is BMP180 compatible. The website says:

Features:
– 0.01hPa and 0.1 degree accuracy (altitude accuracy about 1m)

Not saying that I rush in and don’t take must notice of the documentation but wouldn’t you think that it provides that data as part of the output? Well, it doesn’t. You have to know your mean sea level pressure or your altitude to get meaningful data.

I also know nothing much about Python. So when I installed the Adafruit_Python_BMP library and set it up, I got rubbish from the example output. So I took the sensor back and said it was broken. It was replaced and the same rubbish appeared again. I searched the net trying to figure out how the altitude and mean sea level pressure was calculated from whatever reading the sensor provided BY DEFAULT. Some scary maths (for some) and I realised that the end user (me in this case) had to make adjustments to the BMP085.py AND then set up the library again. The two relevant entries are below.

The default mean sea level pressure is 1013.25 hPa.

# def read_altitude(self, sealevel_pa=101325.0):
def read_altitude(self, sealevel_pa=101590.0):
“””Calculates the altitude in meters.”””
# Calculation taken straight from section 3.6 of the datasheet.
pressure = float(self.read_pressure())
altitude = 44330.0 * (1.0 – pow(pressure / sealevel_pa, (1.0/5.255)))
self._logger.debug(‘Altitude {0} m’.format(altitude))
return altitude

The default altitude is zero meters. I know my altitude is 574 metres.

#def read_sealevel_pressure(self, altitude_m=0.0):
def read_sealevel_pressure(self, altitude_m=574.0):
“””Calculates the pressure at sealevel when given a known altitude in
meters. Returns a value in Pascals.”””
pressure = float(self.read_pressure())
p0 = pressure / pow(1.0 – altitude_m/44330.0, 5.255)
self._logger.debug(‘Sealevel pressure {0} Pa’.format(p0))
return p0

The altitude gave me the mean sea level pressure. So I put that into the def file and it gave me my altitude. Crazy. If you were in the middle of the Gobi Desert, you wouldn’t probably know your altitude or your mea sea level pressure and having the local pressure does you no good whatsoever other than watching what it does. Drop fast means batten down the hatches.

After I worked this conundrum out, the data is good. So why not tell the end user that they have to do something to get good data. Grumble..

While looking at the Adafruit_BMP definitions file, I saw other “raw” definitions. My eyes starting rolling when I saw them. Moving on.

Astrophotography – something to keep me occupied

I started taking pictures of the Milky Way about three years ago with my Sony A6000 and a Samyang 12 mm F2 lens all sitting on a flimsy tripod. Each of the pictures were about 30 seconds and with wide field, that was ok. I didn’t know anything about stacking then and generally used one pic edited in Photoshop (PS) and this was when my PS skills were almost non existent.

So in three years, I kept wanting better pics. I bought an 85 mm Samyang lens and then quickly discovered that I needed tracking to take a pic any longer than a few seconds. So I bought a Sky Watcher Star Adventurer (SA). I then found out that the Samyang lens had some serious chromatic aberration issues and I had to wind the F stop down to 4 to stop the horrible magenta coma. It didn’t look good no matter what I did and there were these small star trails on my pictures that should not have been there with tracking. I blamed the SA for my problems. I then purchased a new Sigma F1.4 85 mm lens that was so much better than the Samyang. I could not work out the star trail issue and raised a problem with Sky Watcher. In the meantime I thought I needed a better camera and purchased a Sony A7iii. I used it on my SA and low and behold, the star trails mostly disappeared. This made me look at my older pics in detail and they were there. I believe there was a fault with my A6000 but it was well and truly out of warranty.

One of the important things with using the SA is to achieve the best polar alignment (PA) possible. This means trying to find some really dim stars that sit above the light polluted area of my city. Given that I had to bend over to look for these frustratingly impossible stars, I started getting upset about the total usefulness of the whole setup. I then discovered SharpCap that is software that can help get excellent PA. I needed a guidescope and an astronomy camera. I also needed a licence to run SharpCap. What can I say, all my PA problems went away in an instant.

However, my tiny star trails were still there and people suggested that my “setup” was not balanced. My Sigma lens is heavy and I thought that getting a collar for it would help balancing. I found a generic Canon type collar and fitted it on my Sigma with the aid of some electrical tape. This helped balance everything and my pictures started looking much better with little star trails even which is reasonable without guiding.

Guiding is something new to my setup and it can be done to a certain extent with the SA but it is limited. To really use guiding to its fullest extent, you need a better mount. The SA is not good but moving your equipment on it or with any wind, things go pair shaped in a hurry. I thought that I should look at a new mount. I decided to get a “forever” mount. The only reason I say that is because the daughter wanted to get her forever house and that went badly when the marriage broke up. Anyway, I ended up getting a Saxon NEQ6 Pro.

I wanted a better lens and bought a Williams Optics Zenithstar Z61. It also needed a field flattener and a Sony T Mount. I couldn’t use all this equipment for months due to cloud, wind and finally bad smoke that hung around for months. An overseas holiday also added to the time none of the equipment was used.

I discovered rings on my pictures when using the Z61 and found out that there was a fault with the field flattener. This had to go to William Optics in Taiwan for rectification. Apparently it is on the way back to me.

In the meantime I thought that my current guide scope/camera was best used to get PA and that I might try getting an Off Axis Guider and associated equipment to help me guide my telescope. I am still waiting to use this on my Z61

In the meantime, COVID-19 appeared and I had intended on going to Wrangal Island in August. The balance of the holiday was to be paid on April 14 but given the travel restrictions, I doubt very much that we will be allowed to travel in August. I am in Canberra Australia and getting to Anadyr in Russia is an adventure itself. I decided to not pay the balance and as I have some money to burn, I thought get some new Astrophotography equipment, mainly a monochrome astronomy camera. I need an auto focuser for the monochrome camera to focus because when changing filters, the focus changes. Getting an autofocuser for my Z61 isn’t easy as finding a company that officially supports that scope is not easy. ZWO say their autofocuser will work but I have heard some issues about the mount.

I contacted a company here in Australia about their autofocuser and after a day or so, they had convinced me to buy a premium mono camera, that being the QHY 16200A CFW7. I thought I would get a better scope (Sky Rover ULT 115 ED Glass Triplet 115mm APO Refractor Telescope with the Sky Rover APO Triplet Astrophotography 0.8X Reducer Flattener) and I needed some filters for the camera. The camera also has an OAG included and I ofcourse bought an autofocuser.

When will the money stop going out?

Getting all this stuff is easy. Learning how to use it is another story altogether. Everything changes with a monochrome camera. I hope to document the learning process for my reference.

Gimbals for mobile (cell) phones – Samsung Note 10+

I recently became interested in mobile videography when I saw people using gimbals with their mobile phones. I am sure everyone has tried to take a video on their phone or camera and well, it leaves a lot to be desired when moving.

I took the plunge and bought what I thought was a good gimbal (Zhiyun Smooth 4) for my Samsung Galaxy S8. It turned out some pretty good video (I was walking up steps in the first part of this video)

but the panorama feature was terrible (read very buggy and unstable). Nothing to show here because it was rubbish most of the time.

Zhiyun didn’t try to fix the problem even after plenty of complaints by the Android users and seemingly decided to spend less resources on Android support for the Smooth 4. I get the feeling that Apple is their preferred market. So much for me buying something else from them.

I upgraded my phone to a Samsung Galaxy Note 10+ and it was NEVER going to work on the Smooth 4. I watched a Youtube video where some guy used the Note 10+ and a DJI Osmo 3 gimbal. I thought wow, I want that so I raced out to buy the gimbal. I never thought about checking the compatibility list for the Osmo 3 and sure enough, the Note 10+ is NOT supported.

I could not balance the Note 10+ as it seems bottom heavy.

I took another look at the DJI balance and calibration clip on Youtube:

I have come to the conclusion that I was misinterpreting how to balance the phone. I noticed that another vlogger used a S10+ and when at rest, it did the same thing. So I decided to ignore the bottom heavy look and focus on trying to balance it as per the DJI clip. It seems to work but the video was not “flat”. There is a calibration tool in the software and I managed to get the video flat.

I just did a quick walk around the house and it all seems ok. Still have a few more tests but so far so good. I don’t know if that bottom heavy problem is going to make the stabilisation motors work overtime but when using my older S8, there was no bottom heaviness (if that is a word).

WO Flat61A back focus for astrophotography with a ZWO OAG.

I have been mucking about with astrophotography for a few years mainly doing some snapshots of the Milky Way. However, when seeing what people with better gear than what I have, I decided it was time to upgrade as I want to get perfect shots of Rho Ophiuchi

with other areas such as the Orion Nebula. So I spent money like the US Quantitative Easing policy. A new camera, new lens, new tripod and a Star Adventurer (SA) made for good pics of Rho Ophiuchi but, like an addict, I wanted better pics.

Let me introduce some lengths before we go any further.

Back Focus: 55mm
Maximum Spacing: 12.9mm
67.7 mm. You may ask what this 67.7 mm is for. Check out the graphic:

You will see a measure for Z61 and A being 67.7 mm. You may also want to figure out WTF it all means. Fail WO for putting measurements and no description of what they are.

While in Malaysia, I bought a William Optics (WO) Zenithstar 61 (Z61). It has a great reputation due to the 360 mm focal length. However, to get pictures without some strange curvature on the fringes, you need to buy a field flattener (Flat61A).

Black Friday and the sales caused me (yeah, not forced) to get a ZWO Off Axis Guider (OAG) because I wanted to do guiding and my SA is just not all that good at doing what you buy it for and that is because of the piece of junk wedge.

So now I have to put the OAG in the imaging train. It is 16.5 mm thick. The Flat 61A says that the back focus between the camera sensor and the bit of glass in the Flat 61A has to be 55 mm. Overall, with everything included such as the Flat 61A, the distance has to be 67.7 mm.

The Sony camera sensor has a flange depth of 18 mm. The WO 48 mm T Mount adaptor for the Sony camera is 38 (could be 37 mm but I don’t have any callipers but they are on the way). I gather that at 37 mm with 18 mm, that means a total of 55 mm. Add that 55 mm and the 12.9 mm the WO says the Flat 61A HAS to be set at, and we have 67.9 mm. Interestingly, that is 0.2 mm over the “absolutely necessary” length of 67.7 mm. Perhaps my T Mount is only 36.8 mm. I really need those callipers.

Lets assume we need 67.7 mm and then do the maths for the ZWO OAG at 16.5 mm thick.

18 mm (flange) + 16.5 mm (OAG) and 36.8 mm (T Mount), we have 71.3 mm. I have to remove 3.3 mm out of the imaging train to get that 67.7 mm. The only feasible method of doing that is to find a replacement T Mount of say 33.5 mm. This is very difficult. There are lots of 42 mm x 1 but finding 42 mm x 0.75 is frustrating and so far, only one place seems to have them and they are in the US. Looks like more money is going to be spent.

Update: I found some adapters at https://www.telescopeadapters.com. I have ordered a 10 mm thick T Mount and a set of expansion rings with sizes of 20mm, 15mm, 10mm and 5mm. I should be able to achieve 20 mm thinkness which adds up to 54.5 mm or just under the 55 mm required. Whether the Flat 61A adjustment will allow that 0.5 mm remains to be determined. I might be forced to use the 10 + 15 mm spacers and adjust the Flat 61A to be 8.2 mm. Wish me luck.

In the meantime I have other new stuff ordered and it will be like Christmas (well, it is Christmas in a few days) when it gets here. I have a new very sturdy tripod and mount and a CLS filter for my Sony on the way.

Labor is sending the wrong message

Annually, around $50 billion goes towards paying the aged pension. Based on DSS Demographics – September 2018 data, there are about 1.5 million people who receive the full aged pension with about 941,000 people receiving the part aged pension rate.

On average, each person on the aged pension costs the government, and vicariously the tax payers, about $20k per annum. This is purely pensioning costs and no other services such as health, etc are included. Labor has stated that an unknown amount of those pensioners with an SMSF will still be able to receive franking credits tax rebates.

Self-funded retirees (not receiving any aged pension) in an SMSF don’t cost the government $20k per year in aged pensions. Currently, based on the median SMSF of two people with total assets of around $700k, those two people receive around $5k combined ($2,500 each) in franking credit cash refunds. That cohort also does not get all the benefits of aged care pensioners who have concession cards.

Labor is sending the wrong message by discouraging retirees to be self-funded especially when a couple on a part aged pension can have assets of $853,000 (where deeming works in their favour and allows them to earn up to approximately $27,500) before losing the aged pension and associated benefits.

An average of $2,500 in cash refunds per person through franking credits to a SMSF self-funded retiree doesn’t seem a lot to pay to keep people off the aged pension. These are the average figures and the ridiculous outliers that Labor suggests should be ignored.

Mortgage offsets funded by trust accounts

I wrote the following as a letter to the editor and so far, it has not been published so I will include it here:

Four years ago, Daryl Dixon, one of Canberra’s financial planning stalwarts, suggested a scheme to allow non mortgagors to be able to contribute funds towards a mortgage offset account.

For example, let’s say the parents of one mortgagor had $100,000 that they would like to invest with the certainty of a bank deposit but with the current interest rates of around two percent, that money is not going to have a high return of investment. Over five years, it would return just over $11,000 without tax considerations.

If banks could provide a trust account associated with the mortgage offset account, this might encourage people to deposit money because of the financial safety of a trust. Only the people who have money in the trust could access the funds, where neither the mortgagee nor mortgagor could access the funds.

Given that the average mortgage is around $388,100 which is taken out over 30 years with a current interest rate of 4.5%, the current monthly repayments would be $1,966.45. With a $100,000 offset, the monthly repayments would be $1,459.76.

After a five year term with the $100k withdrawn, the amount owed would be $353,783.88 with no offset and $328,603.98 with the offset. It has saved approximately $25,180 off the mortgage but the $100,000 loan would have saved interest of $25,179 which is a superior use of money versus the $11,000 interest.

While the person loaning the money would not benefit financially from this scheme, the mortgagor (probably a child of the person) would. This would make housing more affordable while maintaining the integrity of a person’s finances. It is time for the banks to step up to the plate.

There is a 250 word limit to letters so I could not expand on my thoughts because of the word limit. Basically, mortgage offsets are usually funded by the mortgagor and as you have a mortgage, you probably would not have large sums of money to use in the offset account.

However, as a parent with spare cash that you would like to help your kids with, you might balk at putting money into the mortgage offset especially if you didn’t trust one or either of the mortgagors. Your son-in-law might be of dubious character and he might abscond with any extra funds. But if there was some sort of trust account attached to the mortgage offset account, parents could put money into the trust with the knowledge that neither the mortgagor nor the mortgagee could access any of the money.

In the example above, the parents would lose around $11,000 in interest but the mortgagor(s) would save much more than that in interest payments and also lower the mortgage.

You might be thinking that currently, parents can be guarantors for their kids mortgages but it has a lot of risk associated with it. That arrangement can also destroy family trust so anything that maintains that trust has to be much better in the longer term.

There might be some interesting taxation issues with that sort of arrangement but so far, I don’t think this system is available in Australia. There are also limits with gifting in Australia especially if the giftor is trying to lower their assets for pension purposes. This arrangement would not be a gift of the $100,000 but at the end of the term, the loss of interest income would be considered a gift. In Australia, a gift is a transfer for less than market value.

I do have my doubts that the banks would be interested in such an arrangement as they would lose interest payments. But given the recent inquiry into the bank’s financial services, this might be a good way to make themselves look good in the eyes of the public.

There is some discussion online about doing something similar but the funds owner and the mortgagor draws up a contract totally excluding the banks. When I did Law at university, we were told to never offer advice to family members and never draw up contracts between them.

Now have an imaging camera

After some discussion on a chat forum, I decided to buy an imaging camera to help me with polar alignment. Interestingly, last night, the sky was very clear even though the moon was Waning Gibbous with illumination of 76%. The moon was only a distant glow on the horizon so it hadn’t caused any light problems about 8 PM last night. The cold front that passed through on the weekend really cleared up the skies. I ventured out with my Star Adventurer and tested finding the four little stars that are used to find the south celestial pole and only after about 15 minutes, I found it. I rushed inside to grab my camera and when I put it on, I realised that I had not done up the screw that locks the Star Adventurer onto the bit of metal that is used to put it into the mount. I then gave up for the night as I would have had to do it all again. At least with the cooler nights, the mosquitoes are not so active.

Today, my imaging camera turned up. It is the bottom of the range ZWO ASI120MC-S and I didn’t buy a finder scope as I just wanted to see what I could do with the ZWO. I also purchased a professional licence of SharpCap that allows me to do polar alignment with the ZWO. So tonight hopefully will be the night. The weather looks good, the moon will be later and I have the equipment but the clouds look like they are increasing. That being a laptop with a USB 3 port, my ZWO, SharpCap Pro and well, the other stuff, camera, tripod, intervalometer (with pretty average instructions) and Star Adventurer and a small table to put stuff on. I don’t expect a lot but you have to start somewhere.

When I got interested in astrophotography, I purchased a dummy battery for my camera. It can either run off four AA batteries or mains power. I have not used it yet for a number of reasons but given that my laptop will be an essential bit of gear for taking photographs, I might look at some sort of huge powerbank to run it and the other various things. Some of the pictures I see of mainly telescope setups would not be regarded as being portable. My wife asked me if we were taking two cars for our upcoming road trip to darker skies. Lucky all my gear packs away easily for transportation.

Astrophotography – numbers and plenty of them

My last post about my adventure with Astrophotography suggested that I might be able to get my first pic in the last week but that never happened. The moon in all its glory and a thunderstorm stopped that. I have to be able to find the South Celestial Pole (CSP) as it is used to do polar alignment and, well, frankly, that is hard to do. I should say almost impossible with the naked eye using the Star Adventurer’s eye piece when there is any light around. Some people have offered suggestions and the best way of doing polar alignment was mentioned in the last post where you use a camera, a guide scope and SharpCap. Easy, eh? Well, there are about as many options for picking the right camera and guide scope as there are days in the year and if you want quality, you pay for it big time.

The simplest method (perhaps I should say the best value) costs about $400 AUD. I looked at a ZWO camera called ASI120MC-S for about $250. The maths then come into the equation. What would I use this camera for? A guidescope to help me find the CSP. Then what? Take pictures? What of? That last question is the tricky one. I want to have a go at Deep Sky Imaging (DSI). It seems that this particular camera is good for planetary imaging and quite reasonable for solar/lunar imaging. But the camera is very average when it comes to DSI. The important thing is that DSI requires large sensors and interestingly, large pixels. My choice has neither. It has a sensor diagonal size of 6.09 mm and a pixel size of 3.75 microns. The Sensor Dimensions (pixels) are 1280×960. My choice seems to work well for imaging planets and small deep-sky objects.

So what should I buy? Should I start small and grow later? How long is a piece of string? Let’s say money was no object and I chose the ASI128MC-PRO. The name is not all that different to my ASI120MC-S right? It has a sensor diagonal size of 43.3 (versus 6.09 mm) and a pixel size of 5.97 mm. The Sensor Dimensions
(pixels) are 6032×4032. There would be no change for $6400 AUD.

It would be a shame if I didn’t like it right? I have seen a second hand camera (but still in the box) ZWO ASI224MC for $300 AUD plus postage (retail is $379 and no postage). It has the same fundamentals as the ASI120MC-S but it has a slightly bigger Sensor Dimensions (pixels) of 1304×976. It also has a Sony IMX224 sensor. Some forum contributors have not been all that kind to this camera but there are some nasty ill tempered people out there who just like to disagree with everyone. I think that if I realise its limitations as an entry level camera, I shouldn’t be too upset if it has less than stellar (pun intended) performance.

Some one suggested I look at the QHYCCD cameras especially the QHY5II-C. It’s specs are pretty much the same as the ASI224MC. Pixel size is 3.75um, resolution is 1280*960. The optical format is 6 mm which I expect is the sensor diagonal size mentioned with the ZWO cameras. It gets a bit more complex as it talks about quantum efficiency and it is around 74. The ZWO 224 has a TBD in its measurement of quantum efficiency. High Quantum Efficiency means that the sensor is more efficient at turning incoming light into an electrical signal = % of photons converted to light information. If you believe all the scientific marketing information, the ZWO would be in the 75% or higher range. Again, that same marketing suggests the ZWO 224 is very suitable for planetary astronomy and also for small DSO imaging.

There is no doubt that a quantum efficiency of 100% provides a very clear image and 60%, well, its pretty bad. So settling for 75 to 80% offers an image that could be better but the price might be ok. Again, its all about the money.

ZWO push the fact that these cameras probably work better with lots of smaller exposure subs than longer exposure mainly to do with the low noise they seem capable of capturing at low exposure times.

Let’s see where this leads us. My advice to myself is to start small and then see what happens. I can easily spend heaps and not be excited with the results.

Astrophotography: Not nerdy enough

A couple of years ago, I decided to try my hand at doing some astrophotography. This started when I tried to capture the Aurora Borealis near the top of Norway in a place called Kirkenes. I expected to see the lights and I was excited to look up in the sky when we were walking back to our hotel and I saw a green fuzzy look. This turned out to be the Aurora. We raced (as quick as you can on frozen streets when not wearing crampons) back to the hotel and dragged out our tripods and cameras. I didn’t have an DSLR but my wife did and between both of us, we got some fuzzy pics of the lights. Over the next few nights, our pics got better.

We wanted to get better pics so we decided to go to Northern Alaska (Fairbanks, Coldfoot and Wiseman) in winter to try again. This time I came prepared with a camera and lens that could take better pics.

I then tried to get some star pics when I got back to Australia. I have a cousin who lives in the middle of nowhere and has very dark skies. I managed to get some good shots of the Milky Way. I bought more lenses and got better details. I then decided that I needed to get the perfect shot of Rho Ophiuchi.

That goal has completely changed everything as far as astrophotography goes. I needed to buy an equatorial mount that allows me to expose my pictures for longer time periods thus getting more details. I decided to get a Sky-Watcher Star Adventurer. Nice bit of kit but my tripod was not exactly suitable for it. So I got a new tripod. Then I needed a better ball-head to attach my camera to the device. I thought, well, that is all I need for the moment.

How wrong can a person be? The polar alignment capability for the Star Adventurer is frustratingly difficult especially in Canberra as I live in the north western part of the city and the south celestial pole basically sits over the area with the strongest light pollution. It took me nearly an hour of swearing and scratching the mosquito bites before I managed to get somewhere close to the ideal alignment. Even then, I couldn’t tell if the two outer stars were in the proper position because you have to move your eye to look at them and they are very dim with the light pollution.

Apparently, this gets easier when you have dark skies. BUT if you want to spend money (which seems to be the common thing with astrophotography), you can get help finding the pole. You need some software, a guide-scope and a camera. You use this to find the pole and then you end up with somewhat perfect polar alignment.

This is where I am. I haven’t attempted to take a picture yet but with some luck, tonight might be my first test. I don’t expect a lot but you have to start somewhere.

I have a computer science degree and I am quite familiar with most things related to IT but I feel I have met my match with some of the “stuff” that avid astrophotographers use to get that perfect pic. I have discovered a whole new “world” of systems to help including lots of Linux applications that run on various platforms including Raspberry Pi systems. I have two so I will play with one of them just to see what they can do. I have a number of laptops under my bed that would make great Linux systems but I just don’t have the time to play with them. UNTIL now.

Labor’s franking credits misinformation

This is not about what franking credits are but you might want to look them up in detail if you are Australian and invest in Australian shares.

This blog is about the Australian Labor Party (Labor) plan to not let Self Managed Super Funds (SMSFs) claim franking credit “cash payments” for some but not all SMSF pension accounts. This policy is directed towards people who are not getting welfare from various governments in Australia.

You can be a pensioner with an SMSF and still be able to get the franking credit cash payment. You can have an SMSF accumulation account that receives the franking credit cash payment as these accounts do pay 15% tax. What is the difference between an SMSF pension account and a SMSF accumulation account? The former does not pay any tax and that was introduced by the government of the day to try to get more people off the aged pension but that is a macroeconomic story for another day. Governments have struggled with this concept and as a result, they keep changing the goal posts with superannuation.

Labor is going to the next election with four policies that are directly targeting what they consider the wealthier part of town. Labor is so confident of winning the next election, they are offering policies that would not be considered in a normal election. The current federal government is so dysfunctional in its ranks that many people are disillusioned by them and are looking for better leadership. Interestingly, Labor went through a similar time just before they lost power.

Chris Bowen is currently the shadow treasurer and will become treasurer if Labor wins. Last year, he said: “… some SMSFs receive cash refunds of more than $2.5 million a year (presumably in the context of pension accounts)”. Let’s consider how a SMSF pension could get a $2.5 million cash refund. The ASX 20 pays about 4.71% in dividends. An investment of $123,850,000 (yes, nearly $124 million) would get $2.5 million in franking credits.

Stats were rolled out from the Parliamentary Budget Office that supported his claim. 71.85 of SMSFs had balances over $1,974,000. This cohort claimed $1,366.2 million in franking credit cash refunds. Most interesting is that these stats are from 2014 to 2016. A lot has happened since then.

The latest ATO SMSF statistics (18 Sep 2018) can provide an estimate that 68.5% of SMSFs (roughly 750,000 members) have balances at or over $1.6 million. That is not a real change since 2016 but a lower figure is used. This appears to support claims that SMSFs are mostly “rich” people.

But why is $1.6 million mentioned in the second stats?

That figure is the balance transfer cap which means that SMSF pension accounts (tax free) can have a maximum balance of $1.6 million. Excess funds must be taken as a lump sum or moved into an accumulation fund which pays 15% tax. In other words, there are no SMSF pension accounts that can claim anything like $2.5 million in franking credit cash refunds.

Based on 2016 figures, around 50% of the SMSF assets were in pension mode which equates to around 375,000 members with $1.6 million. That figure is bound to change based on the balance transfer cap introduced in 2017 which forced major changes in asset allocations so account based pension numbers would be smaller.

Based on SMSF statistics, listed shares account for 30.8% of SMSF assets. That means an SMSF account based pension with $1.6 million could have $492,800 worth of shares. Dividends at 4.71% would pay $23,211. The franking credit cash refund to an SMSF currently would be $9,947.52.

In the unlikely event that all the $1.6 million was invested in the ASX 20, the franking credit cash refund would be $32,300.

Bowen claims stopping SMSF pension accounts from getting the cash refund will save the budget $11.4 billion over the forward estimates from 2018-19.

If 375,000 ($1.6 million in pension account, with 50% of all SMSF funds in pension account, investing average 30.8% in shares) accounts get $9,947.52, that equates to $3.73 billion.

If all 375,000 accounts invested all their funds, that would equate to $12.11 billion or close to the suggestion of $11 billion).

I am not an accountant so I don’t have any authority with what I am suggesting.

If a SMSF member has $4.8 million (three times $1.6 million) in funds, I don’t think that the SMSF can segregate certain assets into certain parts of the SMSF. Eg, I doubt that $1.6 million cash assets could go into the pension account and the rest ($1.6 million cash and $1.6 million shares) could go into the accumulation fund.

In my circumstance, my accumulation account has a set amount of the fund in it. It is purely an accounting figure. My accumulation fund is made up of cash and shares as a percentage of my total SMSFs assets. In other words, as an example, it does not contain x amount of RIO shares when my pension account holds no RIO shares.

So if I had one account based pension with $1.6 million and one accumulation account with $3.2 million, the account based pension would have one third of the total SMSF assets. If the SMSF assets were made up of $1.6 million shares and $3.2 million in cash, the account based pension would have $1.067 million in cash and $533,333 in shares with $10,765 in franking credits. The other assets would be in the accumulation account getting taxed at 15%.

Labor would not be keeping $32,295 in franking credits as one might think by what they say.

I know the numbers become a blur after a while but it just goes to show that there is something wrong with Labor’s forward estimates of not paying out $11 billion which remains in the government coffers.

The 2016-17 median amount of SMSFs is only $693,265. The mean is $1,223,460. Lets use the median amount as it has a better representation of the real numbers and it not skewed by the the richer end of the spectrum. If half of those numbers (the remaining being in accumulation funds) invested 30.8% of that median amount, the median amount of franking credit cash refund is only $4,310. The total amount of franking credit cash refund is only $4.8 billion.

At the bottom of all this money is that 363,741 SMSF members have less than $1.6 million. These are the people who likely have under $700k in accounts because of the skewed data from the wealthy end so perhaps $450,000 might be more reasonable for this cohort. With 50% being in accumulation mode, that leaves around 182,000 members in pension account mode. These people would lose about $510 million ($2,800 each) in cash payments as they can’t fiddle with the accounts like the richer people.

All these numbers do not take into account that pensioners with an SMSF won’t be losing their franking credit cash refunds. They will be keeping a very close eye on how much they have as the government will be applying deeming calculations to the amount the pensioners have in financial assets. Deeming rates are 3.25% which is lower than the stock markets average dividends so they can have more franking credit cash refunds than a normal tax payer.

Given that the balance transfer cap in now part of the superannuation landscape, I cannot see how Labor thinks that it can get $11 billion a year out of franking credit cash refunds. If anything, the smaller end of town will be the SMSF members hit the hardest. Labor will be taking money from the average citizen to give to the poor in wider welfare handouts while the rich can still diddle their finances.