Ventana Servo Amplifier Design Notes
P. McGill
27Aug15
edited 20Jul17
edited 7Jun22

* Wago DAC modules output +/- 10 VDC with load “> 5 kΩ”. This means each output good for 	5 kΩ / 10 V = 2 mA.

* Servo valve actuators are two 1 kΩ coils in series, so each valve needs
	2 kΩ / 10 V = 5 mA.

* Can use LT1010 unity-gain buffer. Each LT1010 needs max 9 mA supply at +/- 12 V, plus 5 mA output for servo for a total of at least 14 mA per channel. 10 channels would need at least 140 mA at +/- 12 V. This is a lot of current and only provides a unity-gain voltage buffer.

* Best driver is a current source. We’d want +/- 10 V in to produce +/- 5 mA out. This could be done with a Improved Howland Current Pump (see page 2 of https://www.apexanalog.com/resources/appnotes/an13u.pdf). Could use Ri = 100 kΩ, 0.1%, Rf = 5 kΩ, 0.1%. Rs would be (Vin / Io)*(Rf / Ri) = (10 V / 5 mA) * (5 kΩ / 100 kΩ) = 100 Ω.

* Can use LT1498 op-amp. This 10 MHz part consumes only about 2 mA per amplifier. At full drive on all 10 channels, current would be 10 x (2 mA + 5 mA) = 70 mA.

* Supplying op-amps with +/- 15 V provides additional compliance, so circuit could still meet spec into 3 kΩ, i.e. 50% higher than 2 kΩ servo valve resistance. This allows servo valves to heat up without affecting the accuracy of the drive current.

* Testing of first board shows that it works great with better than 1% accuracy, however a disconnected input signal wire produces full drive at output; this would be disastrous on a 40 HP ROV. Problem solved by adding a 10 kΩ resistor across each channel input so that a disconnected input wire forces the input to zero volts.

* Design files are in ProjectLibrary/901415\ Ventana\ Ctrl\ Sys\ Upgrade/Drawings/Orcad\ asBuilt/11021201\ Ventana\ Servo\ Amp

* Email from Mike Burczynski on 3Jun22 on the history of Ventana's servo valves:

Hi Paul -

Looks like you are correct, the -078 coils are in parallel wired internally with only 2 wires showing. I did not open the valve pack as the Carson sailed for the Natalie ash spreading but I checked other stuff ....

Just to refresh all our memories, here is a historical summary of the thruster servos:
- 2013 - I learned our MOOG 240's (& 211's for the camera) were becoming obsolete and parts to service ours were becoming scarce or unavailable, coils were ran in series (2 wires shorted)  with 4 external wires 
- 2014 - we purchased replacement CEI servos recommended by our service technician at HSD who actually was working out of his own shop in Oregon, coils were ran in series (2 wires shorted) with 4 external wires.
- The CEI servos were plagued with issues, main ones being non responsive and causing large system pressure losses.
- We traced this unreachable CEI company address to a car garage in the lower foothills of the Sierras, somewhere near Strawberry if I recall correctly, we decided to lick our wounds and find another solution.
- 2014 and into 2015 I started working with MOOG representatives on finding replacement servos for both lines (240 & 211).
- There were no drop in replacements available due to our size constraints of our valve pack size, hydraulic port holes and proximity of the servos.
- The -078 was identified as the closest candidate BUT several custom modifications had to be made: 1) making end caps smaller to meet the physical dimensions of the valve pack cover 2) relocating the null adjuster from side to top due to proximity of servos 3) rotating nozzle flapper motor assembly 90 degrees to accommodate proximity of servos
- 2016 cut PO for -078 servos with Royal Hydraulics

There are probably a couple other mods I'll find / remember as I'm sifting through a hundred emails. BUT I have verified the two spare -078 that I have are 2 wire and in PARALLEL as you guys found out. I don't know why, I'm still sifting through all my files / emails but the PDF file I've attached confirms that they are wired internally for parallel. 

The -030 and -032 that replaced the 211's was a different process with MOOG. We did not have to make any custom modifications we just had to wait for MOOG to release them from their Defense & Aerospace division and make available in their Industrial division, coils are ran in series (2 wires shorted) with 4 external wires.

I'll keep digging and let you know what I find, we've got another COVID lock down in DMO so I'm outta here.

MikeB