Hi Martin,

Not been around much and I am sorry I missed your question. All 3 subs are Engel's - the Type 7 a beauty and the 2 Typhoons in various states of refit and definately not beauties...

No surprise I am slowly filling one Typhoon with my own control system ("Chase the bubble") but other projects keep jumping the queue so progress is painfully slow.

Regards
Jonathan

Hi Jonathan,
This is way to heavy for me but i would like to know more about the Typhoons and the Type7.

Martin555.

Hi Doug,

I'll post a bit more when I have more time...

I assume the display is showing us the pulse width in µS? >>> YES

I'm further assuming that the slider display on the left should show the analogue of the control stick/servo tester pot position. >>> NO - this is just a display of the value on an iPhone screen - the slider you refer to is the "microphone" slot of the iPhone case


Can you post a pic or diagram of the test setup please? >>> To follow

By what is the PWM read? >>> MICROPROCESSOR - accurate to >5us

I'm also wondering if the capture, conversion, WiFi circuits are introducing their own delay/jitter/distortion. >>> NO EFFECT OR IMPACT

I also have some expensive long tubes that float below the surface - 2 x Typhoon and Type 7

Regards
Jonathan

Then ...
Hope you have a good trip to Hungary without too many complications and
Üdvözlettel, hogy a jobbik fele Zsuzsa !
(Hope I haven't just insulted her Grandma or summat!!😮)
Maybe Zsuzsa will quarantine you in your new workshop? 😁😂🤣
Best wishes to you both.
Cheers, Doug 😎

Thanks Doug.
I am ok to quarantine in my own house. A red notice has to go on the door and the Police will visit at least four times to check.🙀
For the record, Zsuzsa says it was correct.👍🏻

A man of many talents and projects Pete👍
Have fun with the Magyar, hope you can take your Police Launch and tools with you in quarantine🤞 (And/or the Missus!😮) Ó, La La!!😉
Stay safe Shipmate. (Maradjon biztonságban Shipmate! I hope that's right😁)
Cheers, Doug 😎

Hi Doug.
The build is slow at the moment as I am concentrating on Düsseldorf. I can wait as I am away in Hungary for a month soon. So long as our Government don’t mess up again.
I will finish the Police Launch whilst there as it looks like I will have 2 weeks quarantine.

Hi Peter,
"Wish I understood half of this stuff.
Reading between the lines, would it be better if I bought some analogue servos as I am sure the dive planes will not like to shudder.😊 "
You might be right Pete🤔
But don't look for the cheapest.
Look for something with metal gears and a high torque value.
They should come close to the higher holding power of the digital servos.
As previously discussed; the holding power is important to prevent the pressure of water currents on the dive planes altering their position!

How is the sub build going anyway?
Maybe you have time to wait and see what the investigations of Jonathan and myself bring to light?
Cheers, Doug 😎

Hi Jonathan,
Interesting.
I assume the display is showing us the pulse width in µS?
Which for standard servos should vary between 1000 and 2000. 1500 being neutral/rest.
I'm further assuming that the slider display on the left should show the analogue of the control stick/servo tester pot position.
So we should actually see that move up and down when the control input is moved.
But I can't see any correlation there!
As an ex test/development engineer I have to ask😔
Can you post a pic or diagram of the test setup please? By what is the PWM read?
I'm also wondering if the capture, conversion, WiFi circuits are introducing their own delay/jitter/distortion. Early on in my career I learned the test engineers mantra; keep the test setup as simple as possible, commensurate with the parameters to be measured, and enclosed. WiFi is NOT enclosed and is subject to it's own interference sources.
Unless you are living as a hermit in a cave due to Corona🤔
But if it is only transmitting the captured values as bits it probably has little or no influence here. It's the capture and conversion accuracy which is important.

Jitter has always been a major problem in telecommunications, esp. since digital exchanges, line amplifiers and Internet nodes, relays etc.
Many of my colleagues back then made a good living from the purveying of expensive Jitter measurement systems.

As previously written; I'll do some tests the 'traditional' way.
Using a fast scope with high impedance inputs that have the least possible influence on the results.
I hope that my dual trace scope is good enough, and fast enough, to show me simultaneously the difference between what is generated in the TX / Servo Tester and what is actually fed to the servo from the RX.
If it works it could be quite enlightening. I can scarcely hope to compete with the manufacturers though🤔 But I'll do my best.

Off to the 'bench' to sweep away my micro RC gear tests to investigate this phenomenon.

I have a vested interest as I have a large (and expensive) Akula 2 sub kit also using 40MHz RF.
So I will also test the included RX and dive plane servos and depth control systems most exhaustively! I also have a dynamic diving 1/72 type 1A U-Boot that I was hoping to convert to static diving. If I remember correctly this whole investigation started with an enquiry related to a submarine project.
C U in a week or month or so ....
(Please send a search party, with Red Cross recuperative water and food pack if nothing heard after four weeks😉)
Cheers, Doug 😎

Wish I understood half of this stuff.
Reading between the lines, would it be better if I bought some analogue servos as I am sure the dive planes will not like to shudder.😊

A sneek preview to one element of my lastest project Servo+.

PWM Display - The PWM is read then displayed on a WiFi enabled device - in the video this is myiPhone and a servo tester - look no wires!

Jonathan

Thanks Jonathan😊
No bragging from me!
Comes from my years spent as a service engineer and then test engineer (eventually dept. Head). Had to 'get inside' many electronic devices, not only my main interest of RF equipment.
I simply looked at the 'facts' reported and the specs (AFAP) in the same way.
The 3uS dead band stuck out like a sore thumb!
Couple that with the intended usage with an older crystal controlled 40Mhz technology and signal stability was the logical conclusion. If the control signal jitters - so will the actuator. QED😁
I strongly suspect that the stability of the signals generated by modern 2.4Gig radios is significantly better. Would be very surprising if it wasn't. Like the generation improvements I saw in my career in professional/MIL comms.
Must fire up the digital scope (like you👍) and investigate.
I have older (OK VERY old!🤔) 35 and 40MHz sets that I can compare with modern Spektrum and TGY 2.4Gig sets. A DX6 and an i6.
I'll take a closer look at what comes out of the RXs, and also out my my ancient Robbe servo tester, and my own home brew version.
Sounds like there's another Blog in there somewhere.🙄
If you do similar would be veeery interestink to compare notes😉
BTW I couldn't do what you are up to with your App! Not unless i could program it all in Basic😁😂🤣
KUTGW👍
Cheers, Doug 😎

Well done Doug this (Dead Band) was your inital deduction - not something I have come across much in the past but the "old dog has learnt a new trick"

I reprogrammed the Hitec servo changing the dead band from 2us to 5us.

Whilst the servo was not completely happy there was a dramtic improvement when driven from the servo tester.

I can only deduce the output PWM output of my servo tester is rough! I'll capture it at some point to see what scale of "ruff" I am talking about...

Regards
Jonathan

"Analogue servos tend to draw less current, and have a wider deadband, so don’t show an issue on the tester."
I think this is where I came in!😁
Right on Gents👍
😎

Jonathan,

That’s consistent with my experience. My old analogue servo tester was jittering on it’s output. Turns out that the heavy current drawn by the digital servo caused the supply voltage to the servo tester to dip, which caused the tester output to jitter, which then caused the servo to jitter, which then caused more current to be drawn ..... perfect feedback loop!

Analogue servos tend to draw less current, and have a wider deadband, so don’t show an issue on the tester.

Graham93

Gut feel at the moment is that this is caused by the Dead Band Width being very narrow/small on jittery digital servo's as per Doug's initial suggestions.

Using an old fashioned servo tester caused the jitter in my video, I have driven this same servo smoothly using a microprocessor generated (accurate & stable) PWM signal

The servo I am using is programmable so I hope to enlarge/widen the Dead Band width from it's default setting of 2us and see what happens.

Servo Dead Band width = as the PWM signal often changes each frame by a fraction even when no change to the RC handset controls/sticks to stop the servo moving there is a tolerance filter to stop the servo always moving - it's the amount the signal is allowed to change without affecting the output/servo arm. It's an error tolerance

Regards
Jonathan

I knew I had come across "jittery" digital servo's before! This servo works just fine but not in this video...the plot thickens

I just duplicated the issue I believe - more digging required.

May give me a chance to give my latest project (see image) an outing but sadly it's not ready yet having hit more brick walls than I care to remember....

I will update when I have got to the bottom of this

Jonathan

"Perhaps I will have to buy a different make of servo.😐"
Maybe a servo called 'Corona' ❄️❄️😮 wasn't such a good choice Pete!
It's got a Bug!
😎

I had a similar issue with the "mini servo" I had on the spotlight of the crash tender it was jerky and finally just buzzed, got extremely hot (melting) and then stopped. I have a feeling it was because it was a cheap china import. I paid more for a replacement which seems OK at the moment but time will tell.

A bit of a delay as work has just exhausted me lately.
I have tried most things suggested as you can see in the attached photos. I have now determined that the servos are the issue. I have bought a tester and the problem persists with no Tx or Rx at all. I have connected a different type of servo to the tester and no problem at all. The digital mini servos are as rough as badgers.
Jonathan has kindly offered his help so I hope to get to the bottom of it soon. Perhaps I will have to buy a different make of servo.😐

I’m seeing a similar problem with a couple of digital servos. I haven’t gotten to the bottom of it yet but I believe it is a power problem, rather than anything to do with radio or crystals or analogue vs digital servos.

I have an old, home built servo tester, that has worked reliably for 40 years with older servos with no jitter. When I connect one of the new servos I have just bought it jitters. I think the current the new servo is drawing is causing problems with glitches on the power line upsetting the servo tester electronics. That is based on putting a digital multimeter on the power line - I can see dips in the supply voltage. Time to get the oscilloscope out and take a proper look.

Graham93

Thanks Doug and Jonathan and all contributors.
I will try all of the suggestions one by one. I think the mixers is first if I can find them.
The Rx battery is new but I must admit, the Tx battery does not hold a charge for that long. The charged voltage reads 11.4v but soon goes down to 10.9v. Strange because I thought this battery should be around 9,6v.
I will report back when I have tried all of them. I wonder if a factory reset will set all of the mixes to zero.
Thanks for the explanation on digital and analogue. Very interesting but at times I lost the plot.
Peter.

Morning Peter,

Great too see so many suggestions to find a solution.

I had a little voice in my head yesterday that I did not listen to!

I believe this kit is second hand...it's possible the batteries have not been well treated/ left in a discharged state or just old and not holding a full charge...

You mentioned you had a fully charged TX battery - it might be worth a few cycles of slow charge / discharge

The last time I saw multiple servos jittering was when the TX battery was low.

Is the TX battery healthy? Does the radio have a TX battery voltage display on the TX face plate - if so what does is state.

Unless there is a hardware issue I still think there is a high chance this is power related - I would look at both Tx & RX battery

-- A couple of other quickies - does it have an extendable antenna - if so make sure fully extended
-- Does it have a plug in module in the back of the radio - if so remove and reseat
-- Try using channel 3
-- Can you post some pics of the radio transmitter front face and back as well as the receiver
-- If I have the correct instructions for the radio you have - try a "factory reset"

Factory Reset
1.With the transmitter switched off, turn on your transmitter. while pressing both of the two Edit keys (the two keys on the. ...
2.Press the Up or Down arrow key until you get into the Reset. (REST) menu. ...
3.IF YOU ARE SURE YOU WANT TO RESET and clear out. ...
4.Press the Up or Down arrow keys to get to another setup. ...
5.Switch power back on

Just my tuppence 😊

Regards
Jonathan

Mornin' Ed,
And to you and Jenny👍
Dinner was delicious; roast turkey with a pot pourri of vegetables straight out of the oven on a 'help yourself' platter!😋😋😋
Cheers, Doug 😎

Thanks Doug,

At least now I know the difference!

I hope dinner went well with you and Gisela!

BTW, Please give my regards to Gisela!

Cheers, Ed

"Good question! Doug, Where are you?"
Ne paniquez pas Ed😉 Je suis là! Just got back from dinner with Gisela😋😋😋
Oh Lordy Lordy!
To paraphrase Ollie & Stan-
'That's another fine mess we've gotten ourselves into!'
OK. One thing at a time:-
"how do you know that the servo you have is Digital or Analogue ?"
Martin, You look at the specs before you buy it!😉 Or check the label on the case.
Most manufacturers also print DIGITAL on the label.
If you find it in a boot sale or are given it check the label and Google the type number on it.
Either way it's almost irrelevant as for as 'will it work with my RC set' is concerned.
The difference is what happens inside the servo itself.
Basically the digital servo sends control pulses to the servo motor at about 6 times the rate that an analogue servo decoder does. Thus making it more precise in operation, faster starting/accelerating, more torque and better holding power. But the signal it gets from the RX is the same; PWM - Pulse Width modulation. See attached pics, timing diagrams showing the pulse trains, for comparison.
I seem to remember we discussed this way back in Peter's Gato build when talking about servos for the dive planes!!
One practical way to find out is to try to move the servo by hand when in the 'null' position.
An analogue servo will allow you to move it, a digital servo will not! You will notice a distinct difference in the resistance - Holding power!

"(doesn't mention anywhere on the Hitec site that the RX is digital or analogue capable)"
It doesn't need to JB! All the servo wants is the PWM pulse train.
'Digital' has been used as a marketing 'selling point' ever since the first proportional sets appeared. Has nothing to do with 'digital' transmission as we know it today; Internet, dumbphones etc.
Maybe what you are thinking about are the serial bus RXs. A whole different kettle of fish!
'Serial bus' has (AFAIK) little to do with analogue or 'digital' servos as such but is an extra coder/interface in the RX providing addressed outputs to extend the basic channel/function numbers/capabilities of the RX. S-Bus aficionados please correct me if my interpretation is wrong. Haven't had any reason to use it up to now.

Dual conversion. The following will be well known to G6SWJ and any other Hams or RF professionals on board - so they can skip this paragraph!😉
This is a feature of the RX and has nothing much to do with the TX, or it's crystals.
The 'conversion' basically means converting the high frequency received from the TX, e.g. 40MHz here, down to a lower frequency that can be handled by a demodulator to strip off the useful info (i.e. the PWM pulse train) that the servo/ESc needs.
This is done by mixing another frequency with the incoming frequency to produce a lower frequency that can be handled by the demodulator and thus converted into something that the servo or ESC can understand and use.
Along the way potential sources of interference and thus corruption of the useful info are reduced as far as possible.
Simple single conversion radios use one conversion mixer at a relatively low frequency, so called intermediate frequency (IF); e.g broadcast radios (trannies!) IF of 455 or 470KHz for AM band (Radio Luxembourg 'Under the bedclothes club'?) and 10.7MHz for the FM band.
After the mixer comes a low pass filter which strips off the original RF frequency leaving just the lower frequency that the demodulator can handle.
With me so far? WAKE UP AT THE BACK THERE!
Why dual conversion?
A single conversion at a low frequency is good for adjacent channel signal rejection (suppression), i.e. interference from the next door channel in the 10KHz channel spacing of RC systems (27, 35, 40 Megs etc, 2.4Gig has different rules!).
But it's not so good for other spurious signal suppression 🤔 e.g. so called Image Rejection.
That's where the Dual Conversion comes in. 😉 Dual simply means that two mixers are used; First and Second IF.
A first IF at a much higher frequency improves rejection of spurious (unwanted) signals further away from the wanted signal than just the neighbouring channels. Thus improving the selectivity of the RX and reducing interference effects.😊
E.g. the image frequencies appear at input frequency (~40MHz) + and - twice the IF frequency (i.e. the frequency of the local oscillator which is mixed with the RF received from the TX. Thus a higher 1st IF moves the 'images' outside the range of the 1st IF filter😊

BTW Peter, I agree with you. All the Hitec 'manuals' 🙄 I have seen (versions 4 to 6 and Sport versions, for all frequency bands) are 'the stuff that hits the fan'🤔
Loads of set up info for various flying objects but not much else of any use☹️
As with most (99%?) of TX/RX systems made today your Hitec is aimed at the flyboy market.
So the first thing to do Out Of the Box is to switch off (or reduce to 0% if OFF is not an option) all the mixer functions.
I still think that your problem is basically the very narrow dead band of the servo.

Hi Jonathan.
I have been using 1 and 2. I did try one in 5 which was operated by a switch.
Peter.👍

Hi Peter,

I guess your head must be spinning!

What channels are you plugging the servo's into on the RX?

Regards
Jonathan

I have the full instructions that came with the set. There is no trouble shooting section unfortunately.

Thanks JB but the one thing that might discount this is that the servos supplied with the Tx and Rx as a matched set also judder.
Peter.

Good question!
Doug, Where are you?

Cheers, Ed

Hi Guys,
This may sound like a silly question, but for the benefit of the new guys to this hobby and myself, how do you know that the servo you have is Digital or Analogue ?

Martin555.

Another thought Peter, have you tried an analogue servo on the RX (doesn't mention anywhere on the Hitec site that the RX is digital or analogue capable). If you have an old one to try, just turn it on briefly to see if that jitters (if not, check if it operates smoothly).Does the digital one operate smoothly (just jittering in the center)? Don't leave it on too long as if it's a digital only RX it can burn the analogue servo out. Also are the Xtals Dual Conversion and not single. Can be a number of problems, just trial by elimination I think. Hopefully this manual PDF comes through (not sure if you have this model or a manual)
JB

I am having trouble finding a pair of Hitec 40MHz dual conversion crystals. Lots of people are selling the Rx crystal but no Tx.
Does anybody know if another make that is compatible and is available.
Thanks.
Peter.👍

Having done all of the tests so far except for the ferrite rings and the capacitor, it seems to me, considering I have attached all of the set components as supplied, that something is faulty. The question is which one. Tx. Rx or crystals. Is this a fair assumption or am I missing something. I would have thought that connecting all of the parts as they came in the set would work. The saga continues.
Thanks for all of the help. I will report again when I have tried the Ferrite and capacitor.

Hi Martin.
I have not yet connected an ESC. The Rx is being run by an independent Rx fully charged battery.😊

Peter - the ferrite rings are as this video.



Re dual rates - these are the switches on the radio TX - worth switching them to the alternative positon and see if it makes any difference - I think this is a red herring but worth a try



Also try just one servo on the receiver and see if this is stable

Hi Peter,
Just as an experiment connect a separate battery to your receiver instead of using the BEC from your ESC.

Martin555.

Hi
I also have a Hitec Optic 6 and it is a bit of a pain to set up in a boat.
Looks like Hitec are basically built for aircraft

I have ordered some ferrite clip on sleeves as I could not find the coils you mention.
Peter.😊

Sorry Jonathan. I have got a little lost here. I am only just getting to know my RC systems. I have several.
My receiver is Dual Conversion. Can you tell me what dual rate is and where I should look for the switches you mention. Are these the switches that my old Futaba Tx had to halve the servo movement?
Thanks.
Peter.

Just tried the servos that came with the Tx and Rx as a set. They are jittering as well. This must mean either the setup of the Tx is causing the issue or one of the parts, Tx , Rx or crystals are faulty?
Peter.😐

Hi JB.
Thanks for the help.
I don`t know if this is relevant but the system is a Hitec Optic 6 6ch Digital Proportional RC. The receiver was with the set so I have assumed that it is digital. As I know nothing about radios and receivers yet, I might be reading this wrong regarding the digital bit. I have checked the web for information on the Rx but finding it difficult to get the correct info. Is it possible that the crystals are the problem? The receiver is a Hitec HFD-08RO Dual Conversion Ultra Narrow Band.
The whole set looks new. I will try the same setup with the supplied servos next.
Peter.

Peter, a few thoughts on the jitters, -if your RX is analogue and digital capable, have you set the receiver for digital servos?, - it may be set for analogue servos. Make sure you turn on the TX first then RX. Check for matched set of crystals, (Hitec sets need Hitec crystals to work well. Has RX or TX been dropped? Try a new set of matched Hitec crystals. Low RX battery will cause jitters.
JB

Peter,

Just a random thought - which channel's are you plugging into on the RX? - I wonder if this could be an issue with "Dual Rates" depending what channels you are using - try switching both "Dual Rate" switches and see if any difference.

Re a capacitor as a glitch buster - my understanding this is trying to fix an inadequate power supply problem - would think it's better to investigate the power supply but worth a try..

Do you have a standalone BEC you could use to power the RX/Servos - I would be suspect of that RX receiver battery...

Regards
Jonathan

I have charged the Rx battery but no difference. I have just tried two totally different servos. Although less, they both still have the jitters.
I have just seen on YouTube a solution using a 2200uf 16v capacitor acting as a glitch buster. I have ordered some capacitors to see if this works. Does anybody have any experience using this type of capacitor connected to a spare output. The videos show that it seems to work but you must connect a cover to the battery terminals when not in use.
I will read about trimming and mixes this afternoon to see if I can work out how to turn them off..
Peter.😊

I am sure there will be a solution. My personal experience of Hitec kit is it is of good quality.

To confirm can you test with other non Corona Digital servos and see if you experience jitter.

Coincidentally I have been working on a failsafe/pwm conditioner/end point adjuster in the last few days - the main use would be for 40mhz subs but does have a wider application. It's in it's early stages - works fine on the bench but needs to be soak tested....the only issue may be space/payload of the device!

I will add in a configurable "dead band" as per Doug's observation/comments.

Regards
Jonathan

Thank you Jonathan.
One statement you have made has rung alarm bells. The Tx battery is freshly charged but the Rx battery has not. I will charge it ready for the next test.
I do not have a servo tester. I have, however, done a test on two servos not in position. I disconnected one servo and there was no change to the remaining one. They are all digital servos running on 40MHz on PPM.😀
Peter.

I have done a few tests.
The Tx in the next room has no effect. No easy cure then.
I then set up two different servos of the same type away from the frame to see if my structure was causing an issue. The result was the same.
I have looked into the settings of the TX and as I have never done any adjustments or setups like this before, I have a lot to learn first. The Hitec manual is not the easiest to follow.
I am looking at trim but this seems to be a huge project. I will look into mixes but I have to find them first. The setup is for PPM so that is a good start.
Peter.

Hi Peter,

A few observations:

1. Power supply - is the power source they are connected fully charged / up to the job - those servos can draw a fair current - 2 of them connected could easily be drawing over half an amp which may lead to a voltage drop. If you disconnect one servo is the one left connected more stable?
2. Do you have a servo tester? - If so put one of the servo's on that and see if it judders

Regards
Jonathan

Found this article which might help with background info

What is Dead bandwidth?

This is largely discussed for its advantages and disadvantages. Dead bandwidth is a way to say how accurate the servo holds its position. Suppose a servo has a dead bandwidth of 5us (microsecond) and a command is sent to center a servo (which is 1.5ms). Now what happens if the pulse changes from 1.5ms to 1.5ms 3us? Since the dead bandwidth of a servo is 5us, nothing happens until the pulse is greater than 1.5ms 5us.

This is a technique used by servo manufacturers to avoid servo dancing at its center position by telling it to stay in position until the difference between new command and old command is greater than dead bandwidth.
What are Analog and Digital Servos?

Servos are either analog or digital. Difference is how the motor is controlled by the controller circuit. Motor in an analog servo receives a signal from the amplifier 30 to 50 times a second (30Hz-50Hz) known as position refresh speed of servo, allows amplifier to update motor position. In a digital servo, the amplifier sends signals more often at a rate of 300-400 times per second.
There are many advantages of a digital servo.

Faster response: Digital Servo uses a faster microcontroller which is almost 10 times faster than the one in Analog servos

Greater torque: Delivers full torque from start of movement with increased holding power

Programmability: Although most digital servos are preprogrammed, user gets an option to program the servo for center and end points, direction of rotation, speed, dead bandwidth adjustment etc.

Digital servos are slightly on the expensive side compared to analog servos and also power hungry as they have to send more pulses, run a much faster processor along with few additional components.

Technically analog servos are very different than digital servos. From the user’s perspective, the functionality remains the same. Digital servos are controlled in the same way just like any other analog servos. The hard work is all done by the circuitry. If you are concerned about accuracy and control, go for digital servos. If you are concerned about cost and battery power, choose analog servos.