Schematic For Echostar AD3000ip Power Supply

[Llew]

Hi, It's fed from the 21V line out of the PSU. I had to replace the switch some time back, and from the notes I made, the 21V goes to the +12V regulator U901 which feeds the emitter of the switch (Q903).

If no switching occurs, there will be a standing ~12V on the collector of Q903, which is enough to power the vertical polarity of the LNB. Switching is derived from the PowerPC controller chip via Q904.

Not sure how the horizontal voltage is derived from this circuit when switched, I'll check my STB when I have time; meanwhile if you could supply some voltage readings of the transistors concerned I could compare with mine.

Llew

[vlammie]

Hi Llew

My ad-3000ip has given up again recently. Again ticking sounds came out of the cabinet (and flickering display). I have investigated the power supply separately again. Turned out that I could not switch it On and Off with PCS line. Finaly replaced Q2 for a BC237 which solved this issue. Put it back in, but the receiver still had the same problems. Tried to load the outputs of the device as it seems that the PSU is switching On and Off. If I load the 5 V output with 18 ohms resistor it is already breaking down. Anybody any idea how much current can be drawn from 5V digital. Expect that it should be much more then just 300 mA as I am trying now.

[scooby12]

thanks for the diags have got the r3 hotspot problem which with the help of your drawing and info i will calling in at maplins to get a replacement
so fingers crossed it will working soon


Cheers

[Llew]

Hi Llew

My ad-3000ip has given up again recently. Again ticking sounds came out of the cabinet (and flickering display). I have investigated the power supply separately again. Turned out that I could not switch it On and Off with PCS line. Finaly replaced Q2 for a BC237 which solved this issue. Put it back in, but the receiver still had the same problems. Tried to load the outputs of the device as it seems that the PSU is switching On and Off. If I load the 5 V output with 18 ohms resistor it is already breaking down. Anybody any idea how much current can be drawn from 5V digital. Expect that it should be much more then just 300 mA as I am trying now.

The 5VD line draws just over 500mA fully loaded. So you need a 10 ohm resistor with a rating of at least 2.5W or more.

Check that current is actually being drawn on this line - it may be the source of the problem if it's overloaded and shutting down the power supply.

Same with the 5VS line. This takes 1A. To check this, you will need a 5 ohm value resistor rated at 5W or more.

Llew

[vlammie]

Hi Llew

Actually it does not make any difference if I load the 5VS or 5VD line. If i attach a 18 Ohm / 10 W resistor it already goes into hick-up mode. I have swapped several components now (IC3 with IC7, IC2 with IC6) just to see if one of these is bad. Without any succes however. Also measured components in the feed-back loop, like C15. The behaviour of the SMPS is still the same. So most suspect is now IC1 (1H0565R). I have ordered 2 pieces in the US at http://www.bluestar-online.com. I will let you know if this solved my problem after they arrive. This will take a couple of day's i presume. So we have to keep

[Llew]

It does seem to point to a problem in the primary side - check the components ariund IC1 before replacing it; the IC has several protection circuits within (voltage, current, thermal).

Check D5/R9 which provide the main Vcc voltage after turn-on. If this is missing, it could be causing the tripping. Be careful when checking this side of the supply though!

I have a spare receiver that I use for checking queries on the SMPS, if you need any voltage comparison checks, just ask.

Llew

[scooby12]

quick update tried replacing r3 with the value listed 15k 2w and it didn't work lets say it had a short but brilliant life it glowed like a 40w bulb looks like a new sat on the way any ideas out there to replacements ????

[Llew]

So the original resistor wasn't fried, just the board burnt? If this was the case, I have to ask - what was the Maplin's code for the resistor? A quick end to your replacement suggests the wrong value was given.

As for a new STB, not much available now with a built -in positioner, most boxes now use a DiseqC controlled external unit (Vbox) for operating a 36V motor. Plenty to choose from - Technomates are a popular choice.

Llew

[vlammie]

Check D5/R9 which provide the main Vcc voltage after turn-on. If this is missing, it could be causing the tripping. Be careful when checking this side of the supply though!

Llew[/QUOTE]

Hi Llew

I will take care. I have been "shocked" in the past when I was young and also have blown up some probes already

I work now with an oscilloscoop "of ground" which is dangerous also. So don't experiment with this if you don't know what you are doing.

Thanks for your tips. I had tried these all already, but did not find any defect components in the power supply of IC1. I measure a steady voltage of 17.1 Vdc at pin 3 of IC1 after starting up. What I also see is that the SMPS has already problems starting up unloaded. It makes already a number of hicks (about 10) before it is stable. Hicks have an interval of around 100 ms.

[Llew]

I always use an isolation transformer with the 'scope when checking the primary side!

Have you checked all the secondary outputs for shorted diodes (D8,9,10) and their smoothing electrolytics for shorts? Maybe one of them is damping the windings of the transformer and preventing IC1 from powering up.

Llew

[scooby12]

llews

the supply code was d1k5 2w res 1k5 they also have a chart that works out the value if the colours are orange , yellow ,black , yellow , yellow the value works out at 3400000 hope this mean anything to anyone

cheers

[Llew]

No. 1k5 is 1500 ohms. You needed a 15k - 15000 ohms. Hence the quick death.

You may now need to check Diode D1 for possible damage.

[scooby12]

tried again d1 looks ok and put in 22k resistor could get a 15k still get intermittent power ups the dot keep going round and round and sometimes works any pointers

cheers

[Llew]

First culprits to check are the electrolytics. Plenty to choose from, but the ones across the 5V supply (C16,17) work hard. But if any show signs of failing - tops bulging or generally look in a poor state, replace them.

Llew

[patdiamant]

Can't see a Q6 (?)

Hi Llew,
On my Praxis Digimaster 9800CI receiver (same as AD2000) I have Q6.It is a PNP transistor (A1266) between Q2 and Q5 on your diagram. Then Q5 is a power NPN transistor (D2058). </p>Thanks for all your tips about this smps. </p>

[Llew]

Nor could I - it was never used (at least on my supply - I did confirm this in an earlier post)

Llew

[vlammie]

Hi Llew

I finally have my AD3000 power supply working again. The problem was indead not on the primary side. I replaced IC1, but this did not solve the problem. I decided to solder out all the elco's on the secondary side and as you already suggested, some were bad. C23 was halve the original value (22 microF) and C16 had only 1600 microF left. Measured with a standard multimeter. So I can not predict there behaviour for the AC pulses they see when working in the power supply. I ordered replacements and finally the device works correct again. I tested the 5V digital output till 6A load without any problems. Voltage with this load was still about 4.5 V. Regarding the active device list i have found some differences. Q2 was 2SC3198, Q3,4 was 2SC2328 and Q5 was 2SA1273. IC2,6 and 8 were L9946. I also found another hot spot on my printed circuit bord. This was caused by R17 (1k, 1/8 W). This one seems to be a little small for this place. Replaced it with a 1/4 W type. R3 is 2 W in your list. I suggest to use a 3 W type on this spot. On your diagram I think C44 and C46 are swapped. I also could buy a second hand power supply from somebody in the Netherlands, who got it directly from Echostar in the Netherlands as a replacement unit. It has exactly the same hot spots around R3 and R17. Hopefully it won't break again soon. But now at least I have a spare part.

Thanks again for your support, tips and suggestions. If I can be of any help, please let me know.

Let it "burn" again forever

[Llew]

Hi vlammie

Glad the supply is doing its job again.

Better to use an ESR meter on PS caps, as even when showing good capacitance its series resistance can be too high, losing its DC smoothing efficiency.

Yes, manufacturers sometimes use alternative transistors etc. when the original type becomes unavailable. Q5 is indeed a 2SA1273, my mis-type. Will correct the list.

Thanks for your suggestions on R17 and R3, true R3 could benefit a larger wattage, just that space constraints on the board limits size, although with mine, I extended the legs to raise it above the board.

C44 and C46 safety resistors have the same value and do the same job, but well spotted! but I'll leave the diagram as it is unless it becomes necessary to change anything there any time later.

Enjoy your revitalised Echostar!

Llew

[vlammie]

Thanks for your suggestions on R17 and R3, true R3 could benefit a larger wattage, just that space constraints on the board limits size, although with mine, I extended the legs to raise it above the board.

Llew

Hi Llew

Thanks for your tips for measuring elco's. ESR meters do not seem to be very common. I see a lot of home made devices in the internet. The values you get seem to be a little bit arbitrary. In fact you should now the manufacturers specification to be able to judge wetter or not a reading is acceptable.

Talking about R3. I have used 3W devices, which are a little bit larger as the original 2W types, but mounted them in a different way. I have stuck the body of the resistor in the opposite hole and have the bended lead in the hole were the body normally is. In this way it is much further away from C3 and C12 and has more room to "breath".

I also discovered that my back up SMPS was actually not switching off the switchable voltages (21, 30, 8 and 5 V) due to a defect transistor Q5 . This device had a shorted CE. The receiver however worked okay and looked swiched OFF were actually the voltages were still ON. (clock in display). I did not measure the power consumption but this might be much higher then when the switchable voltages are really OFF.

Transistor Q5 (2SA1273/2SA928A in my devices) seem to be dimensioned to narrow. Vce max. for these devices is 30 V. In the OFF stage however the voltage accross collector-emitter is continously more then 30 V. The SMPS is still switching to keep the 12 V and 5VS line "live". So Q5 will continuously see more then 30V across CE.

I have replaced this transistor for a BD229 which i had laying around here. This device can withstand higher CE voltages, although Ic max is a little bit lower (1.5A versus 2A) But I don't expect currents that high in the 30V line.

Hopefully I don't need my spare soon, but you never know. The devices get pretty hot and this is indead not very good for the lifetime of the elco's. I might have to replace some more soon. But for now I am a happy AD-3000 viewer again.

[Llew]

Surely with T5 off, no current is being drawn, so the collector will be 0V referenced to chassis via the load through the 30V line. This gives you the >30V reading as that supply is off load.

My home-made ESR meter can be seen here (post #31) -

http://www.satellites.co.uk/satellit...ave-one-2.html

Llew

[patdiamant]

Echostar used to be affiliated with Hughes Tracker Systems, later HTS, so there is a chance a decent stab with Google will bring up the other PSUs.

Hi Channel Hopper
Houston Tracker Systems, HTS, Echosphere, Echostar, DISH Network, are in the same family. (Not Hughes ...)
HTS receivers had often failures in their SMPS. I had repaired mine many times.</p></p>

[vlammie]

Hi Llew

My SMPS broke down again. This time C10 shows a hard shortage. It blow the fuse.

Lucky i have a spare SMPS now. Replaced it but then discoverd that the motor was not rotating in 1 direction. I got a "motor error". After investigation i discovered that a trace on the PCB runs close the the metal mounting stud inside the receiver. This trace is running from the connecter to one of the relais. This trace is then shorted to ground and the motor is not moving anymore. I have taking of a part of the stud and this solved the issue.
Furthermore I think the east relay is drawn in the "activated" position instead of in the resting position. During the initialisation fase of the receiver (ON) both relais are activated for a short time, bringing both M1 and M2 to 36 Vdc. This results on no movement of the motor. Normal only the east or west relais will be activated, switching eithere M1 or M2 to 36 Vdc and leaving the opposite connected to ground.

[Llew]

You're absolutely right, vlammie. I just drew the diagram 'as is', without regard to the position of the contacts. I'll correct it in due course.

Thanks.

[Sixoence]

Hi - Have an AD3000 and stumbled across this thread. It's currently OK. I have a spare PC PSU fan that seems could help the cooling from earlier postings. Can anyone advise on how this should be installed to that I have a chance of lengthening the receivers life. My electronics knowledge is basic although have a soldering iron, which I've used with varying degrees of success.

[Llew]

You can connect it to the 12V supply (one wire to connector CN2, 6 wires from the left on the mainboard), and the other to any chassis point).

I used this method with a CPU cooler fan (4cm) mounted upright to the middle mainboard screw with a small 90 degree angle piece fashioned out of a strip of metal.

This gave a good airflow by sucking in air from the RH vents , across the CPU and SMPS and out through the LH vents.

A larger fan would probably have to be mounted under a vent in the top lid (preferably the one above the SMPS).

Llew