Devo7e Power Supply Mods

It’s amazing how the Deviation firmware has evolved throughout these years. Since I first tried it out last year, I fell in love with it. Today I fly all my sub 250 class heli’s (and sub 500mm fixed wings) exclusively with Deviated Devo7e. What a wonderful job the Deviation team has done! Thank you so much for your great effort – I really enjoyed it.

Now I’d like to share some experience gained when fiddling with the Devo7e, specifically in the power distribution area. I’ve learned a lot from previous posts including the one I participated. Have done some experiment and measurement and finally came up with my solution for the D7e.

If you are thinking about powering the unit with a 2s Lipo, and want to take advantage of the higher voltage (potentially translates into longer running hours) and don’t want to risk over driving the on-board ldo/modules and don’t want to lose the capability of monitoring the battery pack, try these simple steps below.


1. Lift the positive side of the big diode which feeds into the built-in RF module
2. Add a Pololu 5V stepup/down module (Part number S7V7F5, item #2119, recommended by Fernandez – thank you.) to feed into the diode
3. Add a Pololu 3.3V stepup/down module (Part number S7V8F3, item #2122) to feed into the 3-in-1 module
4. Remove the tiny 3.3v LDO
5. Instead, wire in a 1117 based LDO module with a ~1ohm resistor at the output

Refer to the picture for these numbered areas.

If you are interested in the background of these changes, they can be found here.

Enjoy flying!

Hi YJ, thanks for feedback.

Myself never really understood the whole power circuitry of 7e. Main issue, I have fluctuation in pack voltage readout in my screen at some protocols at full power such as Frsky. In addition I have some small jitter few ms at the control sticks.
I have replaced the small ldo by 3v3 pololu, as stock became hot, but still problems not solved and did several tests but can't really put the finger on it.
A some point stopped with it as couldn't get it right...

Some question;
-Why the original module need 5V, as the Cypress chip is 3v3 and how is in stock 5v managed in 7e as 4 nimh is 4.8v?
-How the save to prom at power down works?
-Can you confirm in your Tx there is no slight jitter at sticks (midpoints) (see betaflight) and Voltage displayed is stable Frsky full power?
-Can you confirm your Tx still runs and can be switched off at low volt, such as 1s?

I still hope we can just bypass all stock circuitry. Use only one step/up down 3v3 pololu, use the enable disable pin of the regulator, for on off Power. (possibly with capacitor, for delay switch off)
Connect, the ADC of ARM battery volt sense line via a small fet, switch on of with the tx.
Put 2x 18650 protected cell parallel and charge with 5v over USB or external plug, use small 1s Usb lion charge module.

@Fernandez Please see my comments below.

-Why the original module need 5V, as the Cypress chip is 3v3 and how is in stock 5v managed in 7e as 4 nimh is 4.8v?
<

---

The module appears to be fed by the diode which is connected to the "input power circuit" which is around 5v if powered by nimh's. The module has an on-board 3.3v ldo so I was told. By feeding the module with 5V, it mimics the use case of 4 nimh and more importantly, reduces the stress and unnecessary and undesired power dissipation on the cypress module's internal ldo.

-How the save to prom at power down works?
<

---

I assume there's power down nmi which triggers the "save to prom" which is why the d7e uses soft power switch.

-Can you confirm in your Tx there is no slight jitter at sticks (midpoints) (see betaflight) and Voltage displayed is stable Frsky full power?
<

---

I don't have any Frsky equipment. But I'd be more than will to experiment if there's test procedure that does not involve an actual Frsky receiver.

-Can you confirm your Tx still runs and can be switched off at low volt, such as 1s?
<

---

I can try but doubt it because the main board (cpu, memory) is still powered by the added 1117 3.3v ldo and a "shadow" 3.3v ldo which I'm not able to locate. Both receive power directly from the battery pack (through the input power circuit). When the battery voltage drops to below what the ldo can handle, the mb will cease operation.

I still hope we can just bypass all stock circuitry. Use only one step/up down 3v3 pololu, use the enable disable pin of the regulator, for on off Power. (possibly with capacitor, for delay switch off)
Connect, the ADC of ARM battery volt sense line via a small fet, switch on of with the tx.
Put 2x 18650 protected cell parallel and charge with 5v over USB or external plug, use small 1s Usb lion charge module.
<

---

That would be nice. A couple of things need to be done in addition to the one you listed above: find and remove the "shadow" 3.3v ldo; Remove the internal 3.3v ldo in the cypress module or use a 4-in-1 instead (assuming it only requires 3.3v); depending on the person's skill set, it's doable I guess.

Personally I'd more like to see a nice looking transmitter such as the Flysky i10 can be "Deviated"

nice work guys , jfip the 4in1 needs vin > ~3.6v iirc as it uses it's own 3.3v vreg, also as you remove the stock rf board to use the 4in1 the existing ldo's only power the mb and screen, which is well within their capabilities though it doesn't allow use of 2s li-* cells. personally on my u7e's i've moved over to using 1.5v output li-po aa's, they are very light and have all the good characteristics of li-* cells without the added issues of voltage problems for the 7e, these are the ones i use >> li aa cells @ aliexpress they are better than others as they only hold 1.5v until 90% depleted and then drop to 1.1v, that way you can still use the low voltage warning on the tx by setting it at ~4.6v. a lot of the other li-* aa's hold 1.5v until their lvc which is no use on a tx as there is no warning when they are about to stop outputting power.

YJ wrote: @Fernandez Please see my comments below.


-How the save to prom at power down works?
<

---

I assume there's power down nmi which triggers the "save to prom" which is why the d7e uses soft power switch.



still on my to do list to make my devo u7e charge via 5V, over USB or the socket, nd install 2x18650.

I would like to know how the system save the settings at switch off, I ussume there is pin to CPU so at switch off the power switch, it start to save to prom, and keep the MOSFET powered for some second.

Anyoone who knows how it works, so we can try to adapt and use the enable disable circuitry of the polulo and get rid of all psu circuitry of DEVO7e.

(I still have this fluctuation of my T batt voltage when using f.i. Frsky)

yeah the mcu has built in power management, from the datasheet

The device has an integrated power-on reset (POR)/power-down reset (PDR) circuitry. It is always active, and ensures proper operation starting from/down to 2 V. The device remains in reset mode when VDD is below a specified threshold, VPOR/PDR, without the need for an external reset circuit. The device features an embedded programmable voltage detector (PVD) that monitors the VDD/VDDA power supply and compares it to the VPVD threshold. An interrupt can be
generated when VDD/VDDA drops below the VPVD threshold and/or when VDD/VDDA is higher than the VPVD threshold. The interrupt service routine can then generate a warning message and/or put the MCU into a safe state. The PVD is enabled by software.

the pins in question are

Hello to everyone, unfortunately putting a battery at 3s I burned the power of my new 7e. I was wondering if you can bypass all using two 5v and 3.3v controllers to get it back to work. Thank you
Update, I connected a 3.3v regulator and replaced a fet. Now the radio goes on but does not see the cyrf module

Probably need 5v.


Edit 2: now all work, but if I turn off, I need to remove the battery for turn on again...