Jump to content

NLCbanner2024.jpg.2478be509670e60c2d6efd04834b8b47.jpg

EOS 1000D extreme modding

yesyes

By yesyes
in DIY Astronomer

Recommended Posts

  • Replies 150
  • Created
  • Last Reply
yesyes

yesyes

Posted
  • Author
Posted

Another question has come up....

I'm going to have to supply the modded camera with a beefy 12V power supply for the peltier and its control circuitry.

I was wondering if it would make sense to include a voltage regulator for the camera power supply inside the new camera housing and have that cooled by the same cold finger/peltier or by the heatsink on the hot side of the peltier. Or is that a heat source that I would rather want to have external?

Link to comment
Share on other sites
RAC

RAC

Posted
Posted

The camera itself doesn't take alot to run. You can just use an LM7808 voltage reg. I just ripped the guts out of the stock battery and ran two wires into the contacts then put it back together. My reg is external but it would be nice to have less cables as you say.

The 8 volt reg won't need a large heatsink.

Link to comment
Share on other sites
Islander

Islander

Posted
Posted

How much heat a regulator produces depends on how large the voltage drop is across the output stages, how much current is drawn and whether it's a linear regulator or a switched mode type. A switched mode regulator will produce less heat but it comes with it's own problems - mainly electrical noise so it'll need to be well screened. Your peltier is 60W so that'll be around 5A at 12V plus whatever other equipment is using 12V. You've already got 60W to dissipate from the peltier so it depends on whether you think the extra from a regulator is likely to make a significant difference overall. On th other hand, if we assume say, 2.5A extra for the mount (conservative figure - if you're using dew heaters etc, then they'll have to be factored in as well) then you're talking about cable with a CSA of 2.5mm2 minimum and preferably 4mm2 for a 5m run at 12v - that's sized for around 5% voltage drop. 2.5mm2 is fine up to 7.5A, 4mm2 is fine up to about 15A.

Them's the choices as I see it. :icon_salut:

Link to comment
Share on other sites
yesyes

yesyes

Posted
  • Author
Posted

Thanks Islander.

This is only about the camera and its cooling. The mount and other accessories will be powered separately (that is, from the same 12V 20A PSU but with separate cables).

So that would be about 5A for the peltier and about 2A for the camera (probably less than that but good to have some margin), so 7A when the peltier is full on.

Then another max 500mA (probably less) for the Arduino and a few tiny 7-segment displays that will show current and target sensor temperatures, ambient temp and dew point (4x 4 digits)

Link to comment
Share on other sites
Islander

Islander

Posted
Posted

To control the peltier, you'll need a variable output power supply anyway - I'd probably go with a PWM unit to minimise dissipation and the Arduino can control that. So you'll have to run 12V to that unit. You may as well build the additional regulation for the Arduino and the EOS into that - it's not going to amount to much compared to the heat dissipated by the cooling unit. :icon_salut:

Link to comment
Share on other sites
Gina

Gina

Posted
Posted

I found my 1100D takes about a third of an amp. So an analogue regulator would dissipate 4/3 = 1.3W, hardly a significant amount :icon_salut: I will be using an LM317T of which I have several for this sort of job. I'll mount it on a small heat sink, air cooled, or possibly on/in the camera, maybe on the hot side of the cooler.

I haven't yet decided which Peltier TEC I shall use but I'm thinking of the 20W 5v one running straight off the 5v o/p of a PC PSU. OTOH the 3.3v supply may be sufficient. I might go for switch mode control of the Peltier.

Link to comment
Share on other sites
yesyes

yesyes

Posted
  • Author
Posted
To control the peltier, you'll need a variable output power supply anyway - I'd probably go with a PWM unit to minimise dissipation and the Arduino can control that. So you'll have to run 12V to that unit. You may as well build the additional regulation for the Arduino and the EOS into that - it's not going to amount to much compared to the heat dissipated by the cooling unit. :icon_salut:

That was the idea. PWM output from one Arduino PWM capable pin into a logic level MOSFET. I'll post the schematic once I've completed it. Maybe someone could have a look at it as I've not worked with MOSFETs before.

The Arduino will then measure the ambient temp and humidity to calculate the dew point. It will also measure the sensor temp (somewhere on the sensor end of the cold finger) and compare that to the target temp. Then, depending on the difference between target temp and sensor temp it will adjust the PWM (the less difference the lower the PWM duty cycle). It will also have an option to not go lower than dew point temp which can be overridden.

Well, that's the plan anyway...

Link to comment
Share on other sites
yesyes

yesyes

Posted
  • Author
Posted

That might be a good idea, although the 12V 20A PSU already has short circuit protection... But it would be nice if just the peltier supply voltage was switched off, not the whole setup.. ;-)

Any suggestions as to how to implement that current limiter?

Link to comment
Share on other sites
Gina

Gina

Posted
Posted

I suppose I ought to think about putting the modified sensor assembly complete with cold finger in my working camera body and see what's what :icon_salut: So far the nights have been quite cold and cooling hasn't been necessary. Mind you, it would enable me to go for even longer exposures and get better data. And the extra red response would drag in a lot more detail. This will be a lot more relevant now that I have a guiding setup.

Link to comment
Share on other sites
Islander

Islander

Posted
Posted

Simplest and most effective way is a low value resistor in the gate/source circuit that will produce a PD of 0.66v when your desired limit current is reached. You use the PD to bias a normal silicon transistor (NPN or PNP depending on the MOSFET type) on which in turn pulls the gate voltage back to the appropriate supply rail to bias it off.

MOSFETs fail faster than fuses do :icon_salut:

Link to comment
Share on other sites
Islander

Islander

Posted
Posted

I don't see why not. However, I hadn't really thought about this in detail (I'm at work) and now that I have, there may be a problem in that your limiting current is going to be quite high say, 8A or so and this is going to mean a miniscule R_sense value at a fairly high power (8A would mean a 75 milliOhm 5W resistor) which isn't generally available. Humph!

This doesn't mean that the idea is a washout though - parallel low value resistors can be used to achieve the value required and as the dissipation will be shared, they can be a lower power value.

Link to comment
Share on other sites
yesyes

yesyes

Posted
  • Author
Posted

I was just calculating this as well...

I take it I need to know the turn-on voltage of Q2. This is the voltage that needs to drop over Rsense when the max current is reached.

I need to check that in the data sheet but let's assume it's 0.7V. The current limit would be 6A. So Rsense would be 0.7 / 6 = 0.116 Ohm. Power over the resistor would then be 0.7 * 6 = 4.2W

I found a 0.1 Ohm 5W resistor on ebay. That should do the trick I guess (cut-off will be slightly above 6A then.

eBay - The UK's Online Marketplace

R1 and R2 are just normal current limiting resistors in the range of about 2 - 5 kOhm, right? Q2 can be a low power TO92 NPN transistor I take it?

Would you agree with that so far?

Link to comment
Share on other sites
Islander

Islander

Posted
Posted

Yes, that checks out. An R_limit value of 0.1 Ohms gives you a current limit of around 6.5A - a 4W resistor would be the minimum power required here so the 5W gives you a nice margin. R1 is the load resistor for Q2 - as a MOSFET is voltage rather than current biased, a highish value here of around 100K would be fine. R1 is to limit the base current of Q2 so yes, 2.2K would be fine here. Q2 can be any low power NPN transistor BC237/337 or similar would be fine here. Q1 would be replaced with a suitable N channel MOSFET.

Link to comment
Share on other sites
Gina

Gina

Posted
Posted

When I made a high current battery charger for our tractor I used an op amp in the current comparator (current limiter) circuit both to make it more accurate and to reduce the voltage drop needed in the sense resistor.

Link to comment
Share on other sites
Islander

Islander

Posted
Posted
When I made a high current battery charger for our tractor I used an op amp in the current comparator (current limiter) circuit both to make it more accurate and to reduce the voltage drop needed in the sense resistor.

That had occured to me, Gina but I thought that the simple approach might be easier overall. I've used this method with a number of power amps and I incorporated it in the output stage of my own dew heater controller. It doesn't need to be super accurate as long as it reduces the drive to keep it within the SOAR of the output device :icon_salut:

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.