Advice Needed V-Box blew up :(

[Waveguide]

If stepper motors were a part of your design the solutions are free if you want to
"roll you own". This is a prototype for a controller for an indexing table (to be used on
a milling machine). The tiny blue rectangle in the upper right-hand corner is a
stepper motor driver that can switch 2 A and at 24 V and on a NEMA23 motor that will
move heavy stuff, geared down it will produce a considerable torque. Next step
(pun intended) will possibly be motorising my dish. Having the teething problems
indoors is easier, 0,3 meters of snow everywhere here.

The Arduino runs on 5 V but the motor supply is whatever you like/need.
And look ...
... no diodes or inductors anywhere.

 

[Llew]

VS - Just a thought. With the toroid, how about serially wiring the 36V and 12V supplies, but with the 12V winding connected in opposition to the 36V to obtain 24V? They're two separate windings, so choose the winding sense to output 24V rather than 48V!

 

[Vipersan]

VS - Just a thought. With the toroid, how about serially wiring the 36V and 12V supplies, but with the 12V winding connected in opposition to the 36V to obtain 24V? They're two separate windings, so choose the winding sense to output 24V rather than 48V!

Not sure if the 12v winding would be adequate current handling wise ...
but a good suggestion nonetheless ..
Id really like to get this unit working with a smpsu if possible ...
So not given up on this idea as yet..

 

[Vipersan]

View attachment 72933 If stepper motors were a part of your design the solutions are free if you want to
"roll you own". This is a prototype for a controller for an indexing table (to be used on
a milling machine). The tiny blue rectangle in the upper right-hand corner is a
stepper motor driver that can switch 2 A and at 24 V and on a NEMA23 motor that will
move heavy stuff, geared down it will produce a considerable torque. Next step
(pun intended) will possibly be motorising my dish. Having the teething problems
indoors is easier, 0,3 meters of snow everywhere here.

The Arduino runs on 5 V but the motor supply is whatever you like/need.
And look ...
... no diodes or inductors anywhere.

Nice idea buddy ..but I'm not that ambitious as to want to redesign the motor drives as well..
Im on a budget here so have to work within practical limits..
nice system you have tho
;)

 

[Llew]

Not sure if the 12v winding would be adequate current handling wise ...
but a good suggestion nonetheless ..
Id really like to get this unit working with a smpsu if possible ...
So not given up on this idea as yet..

In theory it would work, but as you say, it depends on the total current distribution.
Don't know the exact figures, but for use in the Pace, the 36V fuse is rated at 3.15A and the 12V at 630mA.

 

[Vipersan]

Still in persuit of a solution ..
Varistors seem a good idea ..but then do degrade over time as the (oxide grains) are exposed to transients..
apparently ...and I do find great difficulty in understanding the various ratings/attributes.
They do however perform like a non polarised zener diode in many respects..but apparently there are TVS diodes which may also do the job..
These being back to back zeners in one package ...thus not polarised ..??
Never used such a thing ..ever ..
Has anyone reading this ...?
I wonder if I can then use 2 of these thusly ...

From what I can gather reading about them ...
They do still have a bias
Protecting against a larger +ve going transient in one direction ...and a lesser -ve transient in the other ..
A photo of such a diode still shows a stripe at one end as if to indicate the diode has a direction ??
So 2 in parallel one of them up ended ??
Will this work ?
http://www.ebay.co.uk/itm/5-X-STMIC...UK_BOI_CircuitBreakers_RL&hash=item5b0a862122

 

[Vipersan]

Just by way of experiment Llew ...
Just tried running one of the motors (AZ) using just the 12v lower current output of the torroid ..
Off load 15.6 v not 12
...but as soon as is loaded by the motor ...this drops to about 10v ..
motor moves a couple of clicks then Err 2 displayed ..
ie motor has stopped thus no returning control pulses from the reed switch..
The 12v winding simply cannot supply sufficient current..
But since this is a torroid ..
I _could_ easily unwind and remove the 12v winding ...and remove half the 50v winding ..
Thus achieving 25v..
but again ...a last resort.
rgds
VS

 

[Llew]

Just by way of experiment Llew ...
Just tried running one of the motors (AZ) using just the 12v lower current output of the torroid ..
Off load 15.6 v not 12
...but as soon as is loaded by the motor ...this drops to about 10v ..
motor moves a couple of clicks then Err 2 displayed ..
ie motor has stopped thus no returning control pulses from the reed switch..
The 12v winding simply cannot supply sufficient current..
But since this is a torroid ..
I _could_ easily unwind and remove the 12v winding ...and remove half the 50v winding ..
Thus achieving 25v..
but again ...a last resort.
rgds
VS

Yes, should've thought the 12V wouldn't be man enough.
Certainly could try removing the windings, the secondary is normally the outer one on a toroid. Nothing to lose I suppose - apart from a few volts

 

[Channel Hopper]

Been doing some more research this morning ..
and it appears Varistors may be the answer ...
Specifically I narrowed the search down to a V10E25P ...but may well be wide of the mark..
This part is available at digikey.
I'd appreciate someone else taking a look at this pdf link..
http://www.littelfuse.com/~/media/e..._dc_application_varistor_design_guide.pdf.pdf

These devices are used in telecoms and the automotive industry ...for apps like starter motor transient suppression...
acting as an active spark gap.
But not yet convinced I have the correct component rating..
discussion welcome..

Varistors fail (they are always built to a price, not a standard). Add a fuse or two just to be sure

You could pull a couple out of an existing circuit to try though, if you find a suitable board that has them installed. Usually analogue stuff from the last decade.

 

[Vipersan]

Thanks CH ..
looks like we have covered the various simple solutions..
here is the pdf Im using for reference..
granted it is used in relation to relays ..but one assumes it eqully applies to motors..

Chances are I will employ some overkill in the end ..
Possibly a 10 ohm ww resistor in series for current surge limitation...
and at least 1 x 30v mov plus 1 30v TVS in parallel with the motors supply...
Apparently it is advantageous to use more than 1 MOV to increase lifetime useage...
and of course all protected with a seies 4amp fuse.

 

[battenfan]

If you get a short then you'll have 2.4 A through the 10 ohm resistor and the fuse will never break.

If overkill is the way to go, how about a big fat diode in series with the 24 V supply? At least a positive back-emf will not affect the ps.

 

[Vipersan]

If you get a short then you'll have 2.4 A through the 10 ohm resistor and the fuse will never break.

If overkill is the way to go, how about a big fat diode in series with the 24 V supply? At least a positive back-emf will not affect the ps.

Chances are Ill be using a 10 ohm thermistor instead of a WW resistor buddy..
but point taken..
;)

 

[Vipersan]

Chances are Ill be using a 10 ohm thermistor instead of a WW resistor buddy..
but point taken..
;)

Actually settled on device NTC 5D-15
Negative Temp Coefficient Thermal resistor ..
max current 10amp
5 ohm at room temp..
0.15 ohm at max load/temp
1.5cm diameter
So hopefully a soft start for the psu under motor load..
Already tested the on 10 ohm at room temp I had ..with a 50 volt supply and it handles the motor beautifully ...hot to the touch after 5 seconds but not that hot it would melt solder or burn plastic
..

Temp then remains stable as the resistance reduces and effectively takes itself out of circuit by becoming almost short circuit at 0.2 ohm
So am choosing a 5 ohm version of same..
5 ordered for £2.50 inc postage from ebay
rgds
VS

 

[Vipersan]

Not out of the woods yet it seems ..
The new 5amp switch mode arrived ...and so far appears to be ok handling the motor without failing...
I suspect due to the addition of in line negative temp coefficient resistors in the motor line ..
effectively giving the motor a 'soft start'.
A new issue has now presented itself ...and not so easy to resolve.
I need to understand better just how diseqc commands are received an dealt with in relation to stored positions..
I can send the dish anywhere using the remote control or by channel selection on the receiver...but only within a roughly 35 degree variance...??
To give an example I can send the dish from 0.8W to 34 W by remote or receiver without issue ..but 0.8W to 40W is a no go ..
The motor activates ...and heads off to 40W but overshoots 40W and continues to the limit switch.
The motor doesn't stop moving the dish ...but the selected position/command is ignored or forgotten along the way ?
I can get to the desired position by doing it in 2 stages ..
0.8W to say 20W ...and then 20W to 40W ...
It's as though the micro doesn't hold on to the 'issued' command long enough ?
Now ...if for example this is normal and uses a voltage maintained by an electrolytic then maybe I can find a solution ...
BUT if it is a timer built into the micros software ...I can't fix it without having access and re-writing the software.
I believe this problem has been calculated on typical journey time of a dish moving with an actuator powered at 36V ...and thus reducing this to 24v as it now is slows the travel ..resulting in a time out ..and the command is thrown away ..because the desired movement wasn't completed within a fixed time limit.
I -could- go back to using a higher voltage I suppose ...but although I have to wait longer ..I much prefer the motor/dish moving slower at 24V.
Or get used to making shorter arc hops..
This is indeed a calculated error imo...
If only I had a full schematic and and explanation of how the software works..
Suggestions and comments welcomed.
rgds
VS

 

[Captain Jack]

That's rather odd, VS. Can you wire up a temp 36V line to the motor part and see if you get the same issue?

 

[Vipersan]

That's rather odd, VS. Can you wire up a temp 36V line to the motor part and see if you get the same issue?

Yep ..
that should be easy enough ..
I have at least one original transformer I can use ...
Ill give that a try ...
probably tomorrow..
rgds
VS

 

[Vipersan]

So ...2 things to note..
I hooked up one of the original Gbox transformers ...
with a simple bridge rectifier and 470uf 63v electrolytic ..
First checked the voltage OFF LOAD ..
Like with the torroid ...it was greater than expected
50V ..?
So not 36V ...and this would run the cap which fails leaks and bursts in the Gboxes at max ...and borderline tollerances..
No wonder they fail ..
A note to every Gbox owner ..
CHANGE THIS CAP from one fitted rated 50V to one rated at least 63volts..
If not you are just delaying inevitable failure.

Next tried shifting the dish from 1 West to 58 West ..
First by remote control ..
The back from 58W to 1W by selecting TPC on the F15 ..
In both directions there were no errors ..no issues ...no overshoot..
So it looks like I was right ..
Running at 24v (slower travel) causes a timeout after which the command issued by remote control or by diseqc 1.2 from the F15 is dropped ..thus the dish overshoots any arc movement initiated greater than 35 degrees ...
Which equates to approx 60- 70 seconds @24v...
Time taken from 0.8W to 40W is approx 90 seconds @24v
So my guess is a 1 minute timer ..

Can I fix this ..??
In truth I do not know..at this point.
rgds
VS

 

[Vipersan]

So ..as a further test of my theory ..
I used the little preset pot on the psu to tweak the motor voltage from 24v up to 26v ..
Sadly I cannot get much above 26v ..but the increased voltage means the actuator motor turns faster so arrives at the preset destination a tad quicker...
Previously not possible to go from 0.8W to 40W without the command vanishing and the dish overshooting ...but the 2 volt increase made it possible ...
So now I tried 0.8W to 53 West ...and the problem was once again evident..
So looks like my theory of time taken against command retention holds much water..
rgds
VS

 

[Captain Jack]

That sucks. I wonder how the timing is controlled. Is it by the firmware or by circuit components that you can change (caps, resistors)?

 

[Vipersan]

That sucks. I wonder how the timing is controlled. Is it by the firmware or by circuit components that you can change (caps, resistors)?

I rather suspect the dedicated micro for which I cannot find any info is running an internal clock which counts a fixed number of cycles approximating 60 seconds duration...
If parity of referenced command to stored command isn't achieved within this time frame the command is discarded ..
This being the case I see no fix ...other than to run at a higher motor voltage.
I am considering a rather odd and drastic bodge ...but no idea if it is even viable..
That is to run the micro at a lower clock speed..
Currently it uses a 3.58mhz ceramic 3 pin oscillator....presumably internally frequency divided by the software ..
I'm wondering if changing the osc to a lower frequency will allow the micro to still run ...but the clock cycles to reduce..
???

There are some receivers ...for example the AZ box I think ..which can be set to 'repeat' the command several times ...
I often wondered why this was needed ...
but it would make sense insofaras ..this action would refresh the command during the dish movement ...
,,and possibly re-set the vbox command countdown,,
rgds
VS