Merry Chrismas

@routercnc9517

Жыл бұрын

Merry Christmas to you too!

Hi RouterCNC. Thanks very much for this video series. It will enable me to take steps to avoid some mistakes I would otherwise probably make. Kudos for showing them in detail.

@routercnc9517

Жыл бұрын

You are welcome!

Hi folks, I had trouble with a wall wart type 5 volt psu. The voltage was low. I used an open frame psu instead and that fixed the problem.

Hi folks, on my breadboard set up I included an aluminium shield over the top of the Texas Instruments/Clough42 daughter board.

@routercnc9517

Жыл бұрын

It can’t hurt but I suspect the noise gets in through the cables in the first instance as this will be the easiest route.

Another approach to the modifying the timing gear case would be to make it deeper by fitting a sufficiently wide strip around the edge where it meets the lathe, possibly by welding. or pop rivits.Saves a lot of case cutting. However I think what is shown looks better! I was able to fit my stepper motor under the main drive motor on a 290V.

@routercnc9517

Жыл бұрын

That could work but would look quite bulky. Also I don’t like the look of pop rivets. I did consider welding but then you’ve got the distortion and the paint matching to consider (I’ve only got a bit of Warco touch up paint). I’m happy with the way it turned out in the end. Someone mentioned rotating the motor 90 deg and driving through bevel gears which I think for this application would be OK as the backlash would come out if you start well past the start point.

You are right to walk away when you are tired, I have to discipline myself in these situations to do the same!

@routercnc9517

Жыл бұрын

Usually when you go back to it fresh you go straight to the problem because you have been unconsciously mulling it over. But it is tempting to stay and give it one more try . . .!

I'm definitely interested in watching you sort all of this out. personally I appreciate you showing the realities of modifying and re-engineering. It truly helps me realize that I am not alone when having technological setbacks.

@routercnc9517

Жыл бұрын

Watch out for the next episode as I think I’ve cracked it!

Machines: this is the way. Much easier than dealing with people. You will conquer accordingly.

@routercnc9517

Жыл бұрын

Thanks!

I had a lot of trouble to get my ELS run properly. The problem was the VFD drive of the spindle generates a lot of noise. I had to re route the encoder wires. Using a 5V power bank instead of a power supply from the wall socket made it to rund safe.

@routercnc9517

Жыл бұрын

OK good to know. I think my VFD is also putting out a lot of electrical noise. I spent more time with it (after the upload) and might have solved it . . . watch out for the next video.

Wow, what a headache. Admire your perseverance. Fwiw I had to systematically eliminate things one at a time when I had signal integrity issues with my ELS set-up. It can be frustrating, but you'll get there eventually.

@routercnc9517

Жыл бұрын

Thanks! I think I’ve cracked it. Removing encoder screen to earth connection appears to have sorted it. I’ll show in the next video.

@HM-Projects

Жыл бұрын

@@routercnc9517 ground loop ? that's the worst

@routercnc9517

Жыл бұрын

@@HM-Projects Seems likely. It appears when the VFD is on and the motor is running. If I scope the enclosure chassis (which is earthed) there is a lot of electrical noise, and the encoder screen was connected to that. I remember James suggesting to ground the encoder screen if at all possible (it's a bit fiddly to make the connection in the GX16) but in my case it is better without. I'll run it for a bit longer and make sure it has solved it but so far so good!

@HM-Projects

Жыл бұрын

@@routercnc9517 good if it's solved, encoders that output differential signal is another option. You can convert it to a single ended input near the MCU.

@routercnc9517

Жыл бұрын

@@HM-Projects Is that because all encoder lines should have about the same noise (since the screen surrounds them all), so the only difference is the signal? Sounds interesting. I guess I would connect one of the diff signals to 0V near to the MCU since this is after the screening. I'll bear it in mind if I see issues when running it long term. Thanks.

Look at getting some RF Chokes, they come in a form that can clip around cables and help absorb interference :)

@routercnc9517

Жыл бұрын

Could work but I'm pretty sure this is transmitted through the cable, rather than through the air. My understanding is the chokes are suited to making the wire less like an antenna? Anyway if I get stuck I do have some I can try.

i think that meanwell power supply is fine you just didnt connect the scope ground also it was a 50hz wave so that is just mains interference

@routercnc9517

5 ай бұрын

Yes I tested it recently and it works fine. When you are in the middle of debugging it’s easy to get confused!

@routercnc9517

5 ай бұрын

Sorry if you got multiple replies. I lost network just as it was sending and it hung the system

Have you reached out to Clough42? He seems like a helpful guy. You may need to add capacitors to smoth out the interference

@routercnc9517

Жыл бұрын

I’ve managed to solve it. Noise through the ground cable from the VFD into the encoder screen. Removed the encoder screen connection to ground and all sorted.

Watching with interest as I'm about a month behind you in my conversion! I had similar issues with a cheap ebay rpm readout that I tried to add to the lathe a few years ago. It was caused by VFD noise. Turning by hand work fine, the moment I turned the VFD on it went haywire. Doesn't appear to be your issue though. Have you grounded the screened cable from the encoder?

@routercnc9517

Жыл бұрын

Yes I did. And spoiler alert after uploading I went into the workshop and found the issue. It was VFD noise coming BACK through the earth cable to the enclosure. By removing the encoder screen to earth it magically all worked. I have footage to include in the next episode. Happy again!

Not related to this video per say, (although I love this series). I’m currently installing a VFD on my Chester cub 620. I’m wanting to do a reversible conversion,which I have figured all out. I want to utilise the vfd speed pot up in the original control panel but don’t want to drill into it. I’m wondering if I could send over some pics and dimensions for a new plate to be made as it has some nice engraving as part of it.

@routercnc9517

Жыл бұрын

I’m only “allowed” in the workshop for a limited time each week and have urgent projects to finish before March so your best bet is to join mycncuk forum and put the drawings in the RFQ section. There are lots of other hobbyists on there who can also help.

It acted like you only read the Z encoder (index) pulse and not the Quadrature pulses. Put the scope on A or B outputs. James only wires to one side of each, and does not use Z at all although it is hooked up. My encoder is spot on to a laser tachometer measured on the chuck. Check the encoder output, interface from the Launch XL 1A, 1B to the encoder A and B. The encoder BarA and BarB are not hooked up, check the 5vdc and ground as well. You are running, just missing the pulse train.

@routercnc9517

Жыл бұрын

It's a while since I made the video but I think I did check all outputs and 2 of them did not seem right. However, I since removed the screen connection from the encoder cable where it enters the GX16 plug into the enclosure (so only connected to the encoder) it is now running OK. A laser rpm meter reads the same as the display, so looking good!

I'm having a similar problem, but I'm afraid the issue is with my board. I'm not getting any RPM signal to the screen. When testing the voltage on the board encoder pins (encoder disconnected), is there supposed to be 5v coming from the A, B, and Z pins? I want to verify the voltage of those pins before I fry another encoder.

@routercnc9517

Жыл бұрын

The encoder pins on the microprocessor should have an output of 5V, ground, and 3 signal returns from the encoder marked A,B,Z (sometimes Z is Labelled I). I did not check mine but would expect 5V output from the 5V pin with respect to ground pin. The ABZ pins are input signals to the board so you would not see anything with the multimeter there. To check the encoder power it with 5V and spin it with a drill. Check with a multimeter at the ABZ outputs as it should read something although it will be the rms of the signal but should start and stop with the rotation. If you have a scope then you can check the pulse train directly. Hope this helps.

@kpertuit1

Жыл бұрын

@@routercnc9517 Thank you for the help! I think I have a faulty board that fried my encoder. I'm definitely getting the same voltage in the 5V and A, B, I pins. You confirmed my suspicion that it's a bad board. I didn't want to fry another encoder, so it looks like ill be ordering another board as well. Thanks again for the reply.

Just got my system running after frying two processor cards :( Have you used your scope to check for noise on the mains? Also how/where did you bond the cable shields to ground? The shield should only be bonded on one end, I noticed your (old) encoder had the shield connected at the encoder- did you also bond the shield to the cabinet ground? That can cause issues :(

@routercnc9517

Жыл бұрын

Sorry to hear about your processors. I managed to solve the issue by removing the encoder screen connection to the GX16 plug. So as you say it is only connected to the encoder at the encoder end. It is now running without issues. I haven’t scoped the mains but might have a look. Just pleased to get it working at last. Thanks.

@howardeisenhauer1391

Жыл бұрын

@@routercnc9517 Great News 😀 I'm probably going to have to go for #4- my reverse stopped working and I've traced it back to the processor card 😕 Those switching transistors on the CPU really will not take much abuse, I know James likes to have the power isolated from ground but leaving it floating can let things get out of whack from a static point of view. I suspect I'll end up running a 10K or so resister to ground from the card as an equalizer.

@routercnc9517

Жыл бұрын

OK good luck with it!

@routercnc9517

Жыл бұрын

I should add I’m not set up to probe the mains side. I don’t have differential high voltage probes $$$, or an isolation transformer, and my scope does not run on batteries. Also if I added an isolation transformer to the enclosure (just for testing) the system would be isolated from mains ground and I might accidentally fix the issue (only while under test). They are meant for checking the device itself. I would have to be careful with conclusions plus they are also $$$!

@cadewey6181

Жыл бұрын

@@routercnc9517 Sure. In Aircraft where you find miles of wires and connectors, carrying pulse, RF, ethernet, AC and DC with hundreds of Amps, the hardest job is keeping the Audio channels clear. Only one end of shielded audio wires are grounded. Usually with aircraft built with thousands of rivets, the local ground has a potential different than the source ground. Any wire that carries a current is no longer a shield. I am glad you solved it. I also put the switching power supply far away from the Launch board.

Great job! Why not welding mild steel tho ?

@routercnc9517

Жыл бұрын

I was very tempted to do that and fabricate it from steel. But the case is quite thin, which is not ideal for welding, plus with the big hole I wanted to add back some reinforcement. I thought about lots of options but went with this one in the end. I'm not entirely happy with it as it looks ugly! But it will function for now.

@albertovique23

Жыл бұрын

@@routercnc9517 nice Job anyway.

Now I'm not feeling confident about my ELS upgrade

@routercnc9517

Жыл бұрын

I did manage to fix it by removing the encoder screen connection at the GX16 connector end on the enclosure. VFD noise was coming in through the earth cable into the enclosure. I’ll put the footage in the next video.