I could use just a little guidance

[OverTheTop]

FWIW I have just completed the HEI parts for my impending two-stage flight. I have built a capacitor into the HEI to provide ESD protection. That, and a short-circuit shunt across it until I want to fly it is quite likely to keep me, and anyone nearby, safe.

 

[JimJarvis50]

FWIW I have just completed the HEI parts for my impending two-stage flight. I have built a capacitor into the HEI to provide ESD protection. That, and a short-circuit shunt across it until I want to fly it is quite likely to keep me, and anyone nearby, safe.

Would you mind sharing a bit more information on the capacitor and the basis for it (I'm clueless on ESD protection)? Do you have a link for your flight?

Jim

 

[TonyL]

OverTheTop,

In many years of working with electronic systems and devices, I have never directly encountered capacitance used as an ESD mitigation. That said, I have learned something today, which is nice. I too would like to see your approach and reasoning.

Jim there is a mfr web site here, I also found they have a few selection guides that may be interesting too here.

Removing the source is generally better than reducing the effect when practical. Doing both should improve reliability.

br/

Tony

 

[UhClem]

OverTheTop,

In many years of working with electronic systems and devices, I have never directly encountered capacitance used as an ESD mitigation.

I have seen it in a document on protecting SAW filters but that used a series capacitor to provide a low level of protection. But not suitable here.

A parallel capacitor will slow the voltage rise time but will do nothing to decrease the energy of the ESD pulse.

 

[OverTheTop]

Short answer: A capacitor will couple the two leads of the eMatch together for a short period of time effectively (AC coupling). The capacitance of the aggressor (100pF in the human-body model, machine model is 330pF from memory) holds a charge. When that charge is transferred (static discharge) to a larger capacitor the voltage is correspondingly lower (Q=CV, just equate the charge on the two capacitors) . If the charge from the 100pF is put on say 100nF, then what might be a 16kV static zap ends up at 16V. Seriously short in duration also (not enough to rattle an eMatch I think ). Energy in a capacitor is E = (CV^2)/2. Work out the numbers and I think the energy reduces nicely .

Long answer: Let me get back to you another time with some sketches and broader dialog. I am seriously busy with a couple of university teams and my own flight currently. I'll do a thread on my HEI rather than spamming this one.

Link to my Nike Apache here: https://forum.ausrocketry.com/viewtopic.php?f=6&t=5019

Keep up the good work Jim .

 

[TonyL]

OverTheTop,
I think this is important: static discharging current along the bridge-wire in an igniter is not generally considered the most relevant failure mode for static induced ignition [static does not typically have enough energy to heat a segment of wire to an ignition temperature]. Discharge from the bridge-wire through the pyrogen to a nearby component is considered the most likely failure mode [per WFF ordnance awareness training]. Your capacitor across the leads can't reduce a common mode voltage like the one that will occur when something discharges to the shunted leads. A bleed resistor, ideally one near the bridge-wire and one near/on the shunt, is a conventional approach. Adding a parallel capacitor to the bleed resistor will offer some of the reduction you are intending, but parasitic inductance will reduce the theoretical performance.

I would also caution against thinking the human model is sufficient for estimating static issues for rockets. The ordnance ESD standard would be a better starting point.

br/

Tony

 

[JimJarvis50]

Long answer: Let me get back to you another time with some sketches and broader dialog. I am seriously busy with a couple of university teams and my own flight currently. I'll do a thread on my HEI rather than spamming this one.

Yes, let's discuss your HEI system after your flight. Looking forward to a good flight report!

In the meantime, I have completed my magnetic shunt. I was able to further simplify it, and I think it will work fine. A video with some pics and testing is attached.

Please keep in mind that this is intended to provide just a bit more insurance against the igniter firing. It is a third level of protection in that the wifi switch would have to fail, powering the altimeter, and then the altimeter would have to fail on power-up to fire and the channel. The shunt would come into play only if those two things happen.

Jim

 

[UhClem]

[static does not typically have enough energy to heat a segment of wire to an ignition temperature].

That is true for things that meet the 1 Watt for 5 minutes test. A quick calculation shows that a 8kV 100pF zap has more than the all fire energy for a N28B.

 

[TonyL]

Jim,
I do appreciate your thoroughness and testing. If you were to quantify your margin against the no-fire current [e.g. data log current through the match or watch it on a scope], then you would have some objective confidence.

David,
just looking at the no-fire energy and all-fire energy values, they don't correspond directly with the delivered energy in the firing current specs/times, so it is unclear how to interpret them. The no-fire current for the all-fire time [40ms] is 3.04mJ, so still not an obvious answer other than perhaps 'it depends...' Here is the datasheet for other's light reading. It would be interesting to get a Daveyfire application engineer on the phone and ask about these things. MJG might be another candidate to ask.

If someone wants to scuff their shoes across a carpet [I don't have carpet] and zap a match, that would be something interesting to try.

br/

Tony

 

[Steve Shannon]

https://www.emcconsultinginc.com/docs/ematch1.imm.pdf

A study on ESD and electric matches.

 

[TonyL]

Thanks!

 

[JimJarvis50]

Jim,
I do appreciate your thoroughness and testing. If you were to quantify your margin against the no-fire current [e.g. data log current through the match or watch it on a scope], then you would have some objective confidence.

I don't have the ability to data log current or watch it on a scope. I can measure the resistance of the shunt circuit, which appears to be essentially zero (on a meter capable of reading tenths). Even if I say it is 0.2 ohms (which it isn't), and the match resistance is 1.3 ohms (it would be higher in a flight configuration), then the maximum current would be 150 ma against a 300 ma maximum no fire. It is probably not higher than 100 ma and may be well below that. Not sure about objective confidence, but it passes the smile test and it meets the objectives I set out to accomplish. If you can propose a test that can be done with a VOM, I'm game.

Jim

 

[BradMilkomeda]

Jim,

Thank you so much for sharing your setups' and your process in developing your multi-staging system. They are a great reference and guide as I ponder my first HPR two-stage configuration and safe procedures.

I have been wondering how you access your screw switches in a vertical line through your airframe. Now you have a magnet which needs access too! How do you access them then the coupler is installed in the air frame? Do you also have the typical "band" of vent holes for altitude sampling?

Thanks!

 

[JimJarvis50]

Jim,

Thank you so much for sharing your setups' and your process in developing your multi-staging system. They are a great reference and guide as I ponder my first HPR two-stage configuration and safe procedures.

I have been wondering how you access your screw switches in a vertical line through your airframe. Now you have a magnet which needs access too! How do you access them then the coupler is installed in the air frame? Do you also have the typical "band" of vent holes for altitude sampling?

Thanks!

Thanks Brad. I get more than I give, trust me.

I'm not quite sure I understand your question, so if I miss, try again. In my designs, all of the switches are on the skid itself. I assemble the bay, and the whole thing is self contained. Then, I slide the bay into position in the air frame, usually at a point where there is no tube break (I don't like unnecessary tube breaks). Access to the switches is just through matching holes in the bay and in the air frame.

With the holes for the switches, I don't add any additional "vent" holes. I have examined the data from many flights and I have never seen evidence of an issue with this approach for the altimeters that I have used. On the current project, the existing holes for the magnet switch were not previously used, and were covered in prior flights. They had to be enlarged to allow the magnet to fit. I will probably add something to the skid itself to restrict air flow through those holes. I have done that on other bays.

Jim

 

[TonyL]

Jim,
Unfortunately most meters update at around 3Hz, which is too slow. I have a DATAQ that runs at 240 Hz which is better but not fast enough, really you want >1kHz sampling [more like >10kHz] which is well within literally almost any cheap oscilloscope's ability. If you were looking to rationalize a piece of test equipment, this might be a candidate. A single trigger DSO is what you would like. Respectable MHz class USB ones are $100-$200, I've used the PICO Scopes before and they work. Ebay has 200kHz DSO kits for about $30 pre-assembled w/probe. I just bought one to see how it is.

br/

Tony

 

[OverTheTop]

Your capacitor across the leads can't reduce a common mode voltage like the one that will occur when something discharges to the shunted leads.

It does when one of the leads is the casing.

As to using a tougher model for the ESD, I totally agree. I was just using the numbers I could remember off-hand.

 

[jderimig]

For esd protection why not a tvs diode? Or better yet why not a high current igniter like an smd ceramic cap that you overvoltage?

 

[OverTheTop]

Many ways to skin a cat. Many work.

 

[JimJarvis50]

So, the last thing to do with the magnetic shunt was to close up the large holes to limit the amount of air coming into the bay. That's completed (a few pics attached).

This completes the construction/conversion/retrofit for this rocket and everything that is left is preparation for flight. It should be a fun flight. About 23K at a maximum speed of 1,000 ft/s with stabilization engaged for the entire flight. Looking forward to it.

Jim

 

[sean_journot]

Use the roll output as an input to controls system code from yaw damper device to non linearly adjust the pitch. This way the pitch and roll will no longer cause oscillations after extensive tweaking if the system is successfully developed. Well the system iteratively makes altered commands based on what it senses to remove oscillations. I don't think there's a way to totally remove oscillations entirely but it can nullify most of it in time. It's a method that is complicated to understand and many aspects are beyond my grasp. If you had 2+ degrees of roll at a velocity and you need 0.6 degrees pitch. Then at the next velocity or time higher you need 0.7 degrees roll and like 0.05 degrees pitch then next time stamp 0.1 degree roll and 0.75 degrees pitch. If it keep varying then it's not linear to avoid oscillations between pitch and roll so you want to dampen it ideally.

The methods to do it are there on Matlab website tutorial. The actual specifics of doing it practically and implementing it take skill and knowledge beyond my abilities. I could care less about the 747, the idea of using aileron (roll) control surface output in a controls system to input into another axial direction (pitch for rocket) as control surface to avoid oscillations are what is at heart here in a non linear self dampening math model.

Anyways I got a controls exam and it's gonna kick my rear. I'm still scratching my head.

This is an older post but whatever. A control system like this can be simplified by making a few assumptions and initially ignoring canard angle. If the rocket has sufficiently low roll pitch and yaw then these three systems can be assumed to be independent of each other and tackled separately. After figuring out the force (where each canard is modeled as only a force) diagram for each system, generating equations of motion, and converting them into frequency domain a block diagram can then be drawn and modeled in matlab. At that point you would then find appropriate PID values which critically damp each system. Once you have the control system already designed you would then assume that the force generated by canards are independent of each other and the rockets angle of attack. Adding the force outputs for each control system and interpolating through a multidimensional array which contains canard angles for a given velocity, air density, and total force would give you the correct output. The hard/ time consuming part of all this is the CFD modeling of the canards since you would need data points that span from sub to super sonic flow. Compressible flow data points could be simplified further if you assume immediate transitions (perfect oblique shocks) through turn angles around your canards and your canards have a double wedge design, these data points could then be derived from formulas instead of CFD simulations. Granted this is a long, time consuming process but it's definitely doable on a low budget and as long as your not chasing down a fighter jet with your rocket commercial servos should be more than fast enough to keep up with the controller and a system like this is absolutely linear-izable even though from a pure physics point of view it's not (as is the case for just about every mechanical system).

 

[JimJarvis50]

Granted this is a long, time consuming process but it's definitely doable on a low budget....

Sean, all of that sounds reasonable, but it's not something I have the capability to do. I'm just going to have to fly rockets instead. Darn...

Jim

 

[sean_journot]

Sean, all of that sounds reasonable, but it's not something I have the capability to do. I'm just going to have to fly rockets instead. Darn...

Jim

Thanks to your videos I am going to try and make my own system, bought a servo motor from hobby king (robostar sbrs-5314) that I am going to tinker with. Hopefully I can figure out how to mount a set on bearings and cram them inside of a 4"/4.5" tube. Would rather use dynamixel servos but there way to expensive. I forgot half of what I learned in control systems so it's going to take me awhile to design a functional controller.

 

[JimJarvis50]

Thanks to your videos I am going to try and make my own system, bought a servo motor from hobby king (robostar sbrs-5314) that I am going to tinker with. Hopefully I can figure out how to mount a set on bearings and cram them inside of a 4"/4.5" tube. Would rather use dynamixel servos but there way to expensive. I forgot half of what I learned in control systems so it's going to take me awhile to design a functional controller.

Good luck on your project. I have had a lot of fun doing mine. I don't plan to fly mine again until LDRS, but I'm looking forward to that.

Jim

 

[JimJarvis50]

Well crap. I got the project off the ground today at LDRS, but the sustainer didn't light. Don't know why just yet (the igniter didn't burn). Oh well, I will have data to share at some point.

This is the project with the stabilization section at the top of the rocket and the spincan on the sustainer. Would have really liked to see this go.

Jim

 

[VernK]

This is the project with the stabilization section at the top of the rocket and the spincan on the sustainer. Would have really liked to see this go.

From where I was watching the flight, it looked to me like the initial boost was at a pretty good angle but then the sustainer corrected that after separation and was perfectly vertical. Too bad it didn’t light.

 

[SteveNeill]

Think it's beautiful. Good job!

 

[JimJarvis50]

So I've looked over most of the data. At some point, I will put together a video, but I don't have time just now.

One thing that happened is that the rocket coned or oscillated on the boost. I think that got started due to a wind gust (it was pretty windy when I launched). From the ground video, you can see that the rocket weathercocks about 1 second into the flight and then gets hit again about 2 seconds into the flight, and that starts the oscillation. The rocket was at a bit of an angle on the boost - but only about 7 degrees on average according to GPS. After separation, the sustainer went vertical after about 9 seconds of coast (the control canards had a low gain in anticipation of the higher sustainer velocity). Although the rocket was vertical, it still drifted down wind during the coast period (that's how it works with vertical stabilization). The effects of the wind and the return to vertical are evident on the tilt graph.

Due to some factor that I don't completely understand, the sustainer didn't light. According to the simulation, the rocket should have been at 5,860 feet at 12.5 seconds into the flight. The actual altitude at that point in the flight was only 4,480 feet, according to the Raven. I know the Raven altitude is filtered and might be a little behind the actual altitude, but the rocket didn't even get close to the expected altitude, and therefore failed the altitude check inhibition. I have checked the motor file used in the simulation (it's OK) and I have checked the weights of the booster and sustainer, without and with motors, and everything is as it should be. Previous simulations of this rocket have been pretty good.
I think the motor performance was OK, so I have to conclude that the oscillations took more altitude off of the flight than I mght have expected.

At some point, I will likely try this flight again. I have another motor for the booster, and I will likely switch out the Raven for an Easymega so that I can use tilt for inhibition and relax the altitude check.

Jim

 

[OverTheTop]

Great outcome Jim .

 

[Nytrunner]

Plus/minus 10% motor performance? Can happen to research loads too?

Oh well, glad therecovery and guidance worked again! Balls attempt?

 

[JimJarvis50]

Plus/minus 10% motor performance? Can happen to research loads too?

Oh well, glad therecovery and guidance worked again! Balls attempt?

It's hard to conclude much about the motor performance. The Raven I used has the issue with the mis-calibrated accelerometer, so the simulation and flight data can't easily be compared. To a first approximation, the motor performance looks OK. I can't recycle this particular flight in time for Balls, so it will be done at a later date.

Jim