Switch Upgrades

erskyTx runs on many radios and upgrade boards
ersky9x was a port of er9x for use on the sky9x board.
User avatar
MikeB
9x Developer
Posts: 17990
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Switch Upgrades

Post by MikeB »

I'm checking on my 9X with SKY board, my 9XR-PRO has a 6-pos switch installed!
I have:
Switch up: 0000
Switch centre: 07F0
Switch down: 03C8
So, first you may have your switch upside down!
When the switch is centre, it should not connect anything. In this position R83 and R90 (both 2K2) should bias the input to around half the supply. When you switch down, then a 1K should be switched in parallel with the lower 2K2 resistor (R83 I think).
I suggest checking R83 and R90 are the correct values, then see what the reading of A9 is with the switch disconnected.

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!

Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

Thanks for the suggestions. I have two 9XR Pros. The other works fine, so I'll compare the two. Back in an hour or two.
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

The numbers you give are very similar to those for my #1 transmitter (the one that works).

Yours Mine (#1) (Mine #2)
0000 0000 0000
07f0 0801 01f8
03c8 03cc 019e

On the #2 Tx (the one with problems):

R83 and R90 are both marked 222 with third 2 underlined. I read this as 2.2K as it should be.

With switch disconnected, A9 reads 01f8. Same as with switch in middle position.

Update: I checked all the wiring for continuity, value of 1K resistor across switch, operation of switch, etc. No sign of problems.

Fortunately I'm not in any hurry to use the transmitter.
User avatar
MikeB
9x Developer
Posts: 17990
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Switch Upgrades

Post by MikeB »

Check R83 and R90 with a meter if possible, and also check they are both soldered in well.

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

Both R83 and R90 are properly soldered.

When measured with a meter, R83 gives 570 and R90 gives 800. Are these numbers representative or are they affected by other circuitry? Do I need to replace the resistors?


Sent from my iPad using Tapatalk HD

User avatar
MikeB
9x Developer
Posts: 17990
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Switch Upgrades

Post by MikeB »

I'm getting 1K2 when I measure mine (on my earlier prototype). That's for both resistors.
I've just done a few calculations and I reckon you R83 is probably wrong. If it was around 750 ohms it would explain both the analog value you are getting and the resistance readings you are seeing. I get around 400 ohms between 3.3V and ground, so R83 and R90 will appear to be in parallel, with 400 ohms in series with one of them.

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Switch Upgrades

Post by Daedalus66 »

Will pick up some 2.2k surface mount resistors and replace R83. Thank you for the help.

Edit
I just checked the resistors on my #1 transmitter (the one on which the 3-position switch works correctly):
R83 731
R90 506

By the way, all measurements were made with switch in middle position (off). Other positions give different results, presumably due to 1k resistor.
User avatar
MikeB
9x Developer
Posts: 17990
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Switch Upgrades

Post by MikeB »

What did you do about the missing R22?

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

I used a piece of fine copper wire as a jumper for R22.

Just got back from the electronics store. Only a few values for SM resistors and none close to 2.2K. I did get 2.2k 1/4 watt but they are pretty big and clumsy. Will look around for old equipment to canibalize.
User avatar
MikeB
9x Developer
Posts: 17990
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Switch Upgrades

Post by MikeB »

Can you measure the voltage across the resistors when the Tx is on? (Carefully, don't short anything with the meter!).

The mid point should be 1.65 (ish) volts.

You could do a test using the big resistors. Wire them in series across 3.3V to ground. You should be able to find those on the 16-way, expansion bus connector. Then temporarily remove you R22 connection and wire the junction of the resistors to the CPU side of R22. You should then get the mid point voltage on the CPU pin and A9 should read around 07FF.

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

Will have a go tomorrow. Thank you.
Nigel
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

Am I doing this right? When I measure the voltage across R83 and R90 with the power on I get:

R83 R90
0.09. 3.16. Switch up
0.10. 3.15. Switch middle
0.0. 3.25. Switch down
User avatar
MikeB
9x Developer
Posts: 17990
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Switch Upgrades

Post by MikeB »

They don't seem very good values, but are consistent with one of those resistors being incorrect.
The mid-point voltage is also lower than that indicated by your previous A2D value, although may be different now.

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

MikeB wrote:Can you measure the voltage across the resistors when the Tx is on? (Carefully, don't short anything with the meter!).

The mid point should be 1.65 (ish) volts.

You could do a test using the big resistors. Wire them in series across 3.3V to ground. You should be able to find those on the 16-way, expansion bus connector. Then temporarily remove you R22 connection and wire the junction of the resistors to the CPU side of R22. You should then get the mid point voltage on the CPU pin and A9 should read around 07FF.

Mike.
I measured across R83 and R90, with the results shown above, but I wasn't sure where you were referring to by the midpoint.

Then I wired two 2.2k resistors in series from ground to 3.3v on the bus and of course got 1.65v at the midpoint. However, when I connected the midpoint to the side of R22 closest to the CPU (with jumper removed) the voltage dropped to 0.10v.

A9 read 01FF regardless of stick position.

EDIT: I'm going to be in Toronto soon and will try to get 2.2K surface mount resistors to replace R83 and R90. Any other values I should pick up? I have some 220 resistors.
User avatar
MikeB
9x Developer
Posts: 17990
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Switch Upgrades

Post by MikeB »

You have something wrong on the CPU side of R22. R83 and R90 are probably OK.
Check what is fitted at position C107 (should be a capacitor).
You could also try following the track from R22 to the CPU and check for and possible shorts on it. The signal also goes to R64 (should be not fitted), and to pin 7 of J20 (the expansion connector).
Getting a reading of 1FF from A9 with only 0.1 volts present is also a bit odd.

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

On quick review I don't see anything amiss. C107 looks like all the other capacitors. No obvious shorts. But it's late and I'm leaving in the morning early, so will take a closer look in a day or two. Thanks for all the help.
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

I'm back from my travels and looking at the situation again. I didn't see any signs of circuit board problems of the kind mentioned, so I thought I would go back to the two position switch setup to make sure I hadn't messed up anything along the way. Once back in stock condition, the transmitter works as expected.

So where to from here? I can do without the three position switch on this one, so am under no pressure, but I did hope to do instructions for the conversion. Right now I'm wondering if all the production transmitters have the same problem (my other Pro, on which the conversion works perfectly, is a preproduction unit from the beta test).

By the way, I now have a selection of surface mount resistors, so could try replacing some if that seems worthwhile.
User avatar
MikeB
9x Developer
Posts: 17990
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Switch Upgrades

Post by MikeB »

From your test with the two 2K2 resistors wired to 3.3V and ground, what ever is wrong seems to be on the CPU side of R22.
With R22 removed, you should get 1.65V on one side of R22 (set by R83 and R90).
You could try a 2K2 resistor from the CPU side of R22 to 3.3V and see what voltage you get. You should get an A9 reading of FFF in this case.

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

Thanks. Will have a go tomorrow.
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

As i mentioned, I'm back in two-position mode. When I measure from the CPU side of R22 to ground I get 0.01v.

Do I need to go back to the three position setup to get any useful reading, or does this one indicate something is seriously wrong?
User avatar
MikeB
9x Developer
Posts: 17990
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Switch Upgrades

Post by MikeB »

If you have a pullup, to 3.3V, from the CPU side of R22, and are then measuring 0.01V, then yes, something is seriously wrong.

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

Sorry. My misunderstanding.

With 2.2K resistor from CPU side of R22 to 3.3v, A9 reads about 024B (varies a bit each time the Tx is turned on).
User avatar
MikeB
9x Developer
Posts: 17990
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Switch Upgrades

Post by MikeB »

That suggests you have a "resistive short" of around 370 ohms to ground. Could you try putting the second 2K2 resistor in parallel with the first one (so you have 1K1 as the pull up), and see what value A9 then has?

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

With 1K resistor instead of 2.2k I get A9 reading of 043A.
User avatar
MikeB
9x Developer
Posts: 17990
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Switch Upgrades

Post by MikeB »

Very close to a "resistive short" of 370 ohms to ground still. This could, of course, be inside the processor.
A described where the track goes a few post back. You may need to inspect the track VERY closely (lots of magnification) to check for a possible 'hairline' short. It may also be worth removing C107 in case it is faulty.

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

Took out C107 and now get A9 reading of 0FFF. This is with the 1K resistor from 3.3v to CPU side of R22 location (no R22 jumper or resistor). Took very close look at other stuff - nothing there.

If this means C107 is faulty, what value is needed for replacement?
User avatar
MikeB
9x Developer
Posts: 17990
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Switch Upgrades

Post by MikeB »

C107 is 0.1uF.
Worth now removing the 1K and replacing R22 and checking you get around 07FF from the original two 2K2 resistors. The 3-position switch should work, even without C107, as a bit of noise on the signal won't affect the operation.

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Switch Upgrades

Post by Daedalus66 »

Success! Reinstalled the R22 jumper and three-position switch. The switch now works (according to DiagSwtch) and A9 shows 0FFD.

Here are the voltages across the two resistors:
Switch. R83. R90.
Up. 0.0. 3.25
Mid. 0.11. 3.08
Dn. 0.11. 3.13

Make sense?

So it was C107 that was the culprit. Is it a good idea to install a replacement? I note that you say a bit of noise probably doesn't matter, but presumably there was a reason for it in the first place.

Thank you for all the help.
User avatar
MikeB
9x Developer
Posts: 17990
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Switch Upgrades

Post by MikeB »

Good it's working!
You could try measuring the resistance of the C107 you removed, it might have been a very fine solder short and the capacitor is actually OK.

C107 is there to avoid noise (including RF noise) disturbing the signal. The values used by the 3-pos switch are a long way apart, so should be OK. Should you wish to go for a 6-pos switch then you would probably need the capacitor in place.

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!
Daedalus66
Posts: 1844
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Switch Upgrades

Post by Daedalus66 »

No sign that C107 is shorted, so maybe next time I have the Tx open I'll put it back. But no plans for a six position switch on that transmitter, anyway.

Will write up how to do the three position conversion.

Post Reply

Return to “erskyTx (was ersky9x)”