Advice Needed V-Box blew up :(

[Channel Hopper]

Now I always prefer to allow for loading (wind etc) and problems encountered ..
So allowing an extra 0.5amp for each motor ...we are back at my original approximation (2 amp per motor)
4 amp in total ..
So at 25v roughly this would hopefully equate to 24 x 4
my estimate is therefore 100W or 100VA
The psu that failed was 200W ..so no excuse for failure imo
but still ...if I went down the torroid route which I'm seriously considering ..
A dual secondary 12v (x2) seems a sensible option ..
Thus I would have 2 speeds
Slow and very slow
..the rating of which would I assume be 100VA
Do I have this right ?

Not quite, the power supply is providing power for an inductive load, and therefore you could be drawing somwhat more owing to the reactive power of the load when driving the dish.

( Did I mention sticking capacitors on the relay contacts ?)

http://electronics.stackexchange.co...y-inductive-loads-producing-input-square-wave

 

[Vipersan]

Hmmm..
Interesting stuff CH ..
So what value caps would you suggest across the contacts ?

From what I've been reading a C/R series combo would be best ..
directly across the contacts ...but how to ascertain the values I have no idea
input welcomed

 

[Vipersan]

Well from what info I have gathered a simple RC snubber would comprise 0.1uf and approx 120 ohm in series accross the contacts ..
Some relays have this built in but in my case I have no way to confirm one way or t'other..short of disassembling one ..
Maybe I could try searching for the relay on the pcb..
rgds

 

[Waveguide]

No filter capacitor after the diode bridge?
Charging/discharging capacitors without current limiting (resistor) is
one good way to fuse relay contacts if the peak currents are high.
Easily tens/hundreds of amps for common voltages and sizes for
electrolytic filter capacitors. Even if the relays don't fuse the contacts are ruined
much faster than for normal loads.

 

[Vipersan]

Well ...
The Gbox schematic is as they made it buddy ..
I havent changed anything other than replace the 36VDC (transformer) source with a 24v (SMPSU) source..
They may have compromised it's design for cheapness..
Anyway ..I have found a pdf for my relays ..and they do not show any built in snubber ..
Time to fix this perhaps ?

The good news is that I can get 10 x of these relays from China for £5..
So easily changed if the contacts have suffered ..but how to mod the cct to prevent future problems..
Do we still think the basic RC snubber will provide a solution ?

 

[battenfan]

Edit: Never mind

 

[Vipersan]

So ..put pen to paper and traced the Gbox motor control schematic ...insofar as it is relevant to back emf and transients.
TBH it looks ok ...but only a single 100nf cap used accross the contacts ...
no series resistor ....(snubber)
but a 100nf cap and varistor across the motor contacts ..M+ and M-
A 470uf cap decoupling the motor supply
There is no series current limiting resistor from what was the bridge rectifier to the relay contacts
Perhaps I should add one ? ...but it will get a bit hot I suspect

Keep in mind also that there is a second Gbox pcb in parallel on the 24v line...
thus you can increase the relay contact cap to 200nf ...and the decoupling electrolytic to 1000 uf +
anyway..
I've done enough staring at the pcb tonight..

suggestions for any improvements are most welcome..

rgds
VS

 

[Channel Hopper]

suggestions for any improvements are most welcome..

rgds
VS

I would add a low resistance (8 - 16 Ohm) resistor in series with the relay drive supply (possibly two, one on each side).

You must have a couple of droppers from analogue TV days in your box of tricks.

 

[Vipersan]

I like your thinking buddy..
I was considering doing just that with a couple of Positors ...
They're about 10 ohms cold ...
use to use 'em in simple crt degause circuits
good idea or not ?

Positive temperature coefficient resistor. Another name that might be
used is PTC varistor

But wait ....I think I have that upside down ..
I need a NTC device ...
back to the drawing board...
lol

 

[Vipersan]

Ok ...
I think the SG26 should do the job...
Negative coefficient resistor ...

Specs are :-
RESISTANCE AT 25 DEG C = 5 OHMS +/- 15%
MAXIMUM CONTINUOUS CURRENT (Imax) = 12 AMPS
RESISTANCE AT Imax = 0.06 OHMS NOMINAL
MAXIMUM VOLTAGE RATING = 265 VAC (RMS)
MAXIMUM SURGE RATING = 100 JOULES
ENCAPSULATION = SILICONE RESIN
MARKING = OPTIONAL
LEADS = RODAN P/N STL-209-18

 

[Channel Hopper]

I think a couple of nice wattage wired resistors will do the job.

And......... if they are wound in the opposite way to the motor, they might just counter the inductive load...........................

 

[battenfan]

Sorry, I'm not following your reasoning. Why talk about inductive load of a dc motor? Sure, a motor consumes lots of current just before it starts moving but that resembles more like a capacitive load to me, no? An old trick for cheap current limiting is having light bulbs in series. And back emf when the dish stops is dealt with diodes, right?

 

[Waveguide]

Only a diode will not work if you want to run the motor in both directions. Diodes
in series with zener diodes (two pairs anti-parallell) will limit the back emf.
Light-bulbs are nice, low resistance cold and higher resistance at full voltage.

Another question to VS. The voltage rating of the relays is 24 V?
Running relays on voltages lower than nominal makes them much slower.
When the voltage is marginal you can hear a relay "dance the two-step".
The first part of the movement is aided by the spring action of the contact
it's leaving, the contact with the normally open part takes longer, that
contact is resisting the movement.

Reversing a motor at full speed will create a lot of back emf. Can a SMPS handle
power dumped into the output of the PS?

 

[Archive-8]

 

[Vipersan]

Only a diode will not work if you want to run the motor in both directions. Diodes
in series with zener diodes (two pairs anti-parallell) will limit the back emf.
Light-bulbs are nice, low resistance cold and higher resistance at full voltage.

Another question to VS. The voltage rating of the relays is 24 V?
Running relays on voltages lower than nominal makes them much slower.
When the voltage is marginal you can hear a relay "dance the two-step".
The first part of the movement is aided by the spring action of the contact
it's leaving, the contact with the normally open part takes longer, that
contact is resisting the movement.

Reversing a motor at full speed will create a lot of back emf. Can a SMPS handle
power dumped into the output of the PS?

The actual relay control (coil) voltage is actually 12v ....SRD-S-1.12
So I have the right control voltage in place ...
I do however take your point about reversing the motor when in motion...and this I believe MAY have been the reason for smpsu failure..and why transformer supplies are used..
This is my gut feeling anyway..
So before I fit ..try...and destroy another smpsu..
Suggestions for mods to the schematic above are imo what need most attention.
Anyone have any useful ideas ..?
Yes we can limit the surge through the relays/motor either with a WW resistor ...NTC thermistor ..or crudely with a lightbulb..
Any better ideas ?
Back EMF being the killer I think..

I really dont want to go back to a transformer ..but unless a viable solution can be found ...I may have to..
Can't go changing psus every week..

rgds
VS

 

[Channel Hopper]

So before I fit ..try...and destroy another smpsu..
Suggestions for mods to the schematic above are imo what need most attention.
Anyone have any useful ideas ..?
Yes we can limit the surge through the relays/motor either with a WW resistor ...NTC thermistor ..or crudely with a lightbulb..
Any better ideas ?
Back EMF being the killer I think..

A couple of chokes (coils to the younger generation) in the supply rails to the relays will halt the back EMF.
Back to the dropper resistor idea (some sites call them bleeders......), and shove a bit of ferrite down the middle.

 

[Vipersan]

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..

 

[Archived-1]

This might be an easy mod VS if all else fails,

http://www.ebay.co.uk/itm/like/3811...1=ICEP3.0.0-L&ff12=67&ff13=80&ff14=108&ff19=0

 

[Vipersan]

This might be an easy mod VS if all else fails,

http://www.ebay.co.uk/itm/like/3811...1=ICEP3.0.0-L&ff12=67&ff13=80&ff14=108&ff19=0

Seems a bit like giving up buddy...
I've no doubt it will shift the dish AZ/EL ...but doubt it would accommodate stored positions and definately not diseqc control..
I'm not giving up what I've built so far ..
A torroidal 24v 100va transformer being my last resort ...and relatively inexpensive.

 

[Archived-1]

I know what your saying Evan, you've put a lot of hard work into this, it would be a shame not to be able to finish it, good luck pal and sorry I cant help you I wish I could.