[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] help needed! PSSTC



You said you were building this on bread board. 


The TL494 uses either open collector or open emitter out puts (your choice). Make sure the leads on the pull up (or the pull down resistors depending on what configuration you are using) are  very short, and not adding inductance to the circuit. Also consider the bread board tracks, in fact any piece conductor between the TL494 and the supply bypass capacitor. 


You could also try increasing the value of these resistors to reduce the on state current and hence reduce the transient during switching. The data sheet test circuit uses something like 150 ohms but you can easily get away with 2k2 if you are driving the input to a modern IC.

I would recommend moving to veroboard and soldering everything. This gives you confidence in your connections and more control over parasitic effects. Go on you can do it.

David


________________________________
 From: Steve Ward <steve.ward@xxxxxxxxx>
To: Tesla Coil Mailing List <tesla@xxxxxxxxxx> 
Sent: Sunday, 15 January 2012 12:12 PM
Subject: Re: [TCML] help needed! PSSTC
 
The TL494 should be pretty clean output, but it might all go bad if
the power supply decoupling is poorly placed or non-existant.  If you
cant find the cause of the problem, then a capacitor (probably
something like 470pF) placed at the TL494 outputs to slow down the
transitions and filter out the transition glitches would be the first
thing i try.

Steve

On Sat, Jan 14, 2012 at 2:56 PM, Jason Hill <effuzionz@xxxxxxxxx> wrote:
> This ringing does start at the 494 at the beginning of each pulse on the outputs of the 494. Then at the gate driver outputs it then gets reflected to each switching event. I was under the impression that this problem should be tackled at the 494 outputs, but as we did some research and someone had a similar problem w their sstc failing quite frequently due to similar waveforms. His solution is the one I have described earlier where he attached a R-C-D snubber across the gate-drain of the MOSFET's. He was able to rectify the problem but as I said before, having this circuitry on bread boards just to get a "proof of concept" so to speak before we started looking into getting a pcb mocked up for it, proved to be part if the problem as well. We have now gotten everything properly placed on pcb and hardwired to reduce complications when testing this circuitry again.
> Steve, my initial reaction was to place a low pass filter at the output of the 494. Was this something similar to what u were thinking? Again I am a first time builder and tuning these signals is going to be the rise or demise of this project. I hope I have clarified things. What would y suggest as a solution to cleaning up the output at the 494?
>
>
> Sent from my iPhone
>
> On Jan 13, 2012, at 11:34 PM, Steve Ward <steve.ward@xxxxxxxxx> wrote:
>
>> It sounds like you are fixing the problem at the wrong location.  If
>> there is ringing at the output of the TL494, then FIX IT THERE FIRST.
>>
>> If all the waveforms up to the MOSFET gate look clean, but the MOSFET
>> gate signal is overly distorted, or its going into oscillation at HF,
>> then you should probably look at snubbers.  But if your gate driver is
>> outputting bursts of 13MHZ at each switching event, then the problem
>> is upstream.  So either you are not explaining the problem correctly,
>> or your solution is not the right way to go about it.
>>
>> Steve
>>
>> On Fri, Jan 13, 2012 at 12:34 PM, Jason Hill <effuzionz@xxxxxxxxx> wrote:
>>> The transient does exist on the 494 w out the gate drivers hooked up and we have swapped the 494 out and we still get the same signals. The transient exists on the beginning of the pulse of the 494 and then gets reflect on the both the on and off thru the gate drivers. We have tried a few different oscopes along w probes that were all calibrated before being used.
>>> It seemed that we have found someone else who has had the same problem and that persons solution was to use a snubber circuit on the input of each MOSFET.
>>> The transient is oscillating at about 13MHz. We are running some trial and error w the cap values that will help eliminate this signal. The snubber we will try and implement first is n r-c series connection from gate to drain. Depending on the results of our first go at it we may try implementing a snubber that would also incorporate a diode but we are not to sure what diodes would be suitable for this application. Seems that we have plenty to try.
>>>
>>> Sent from my iPhone
>>>
>>> On Jan 13, 2012, at 11:28 AM, "McCauley, Daniel H" <daniel.h.mccauley@xxxxxxxx> wrote:
>>>
>>>> Did you measure the gate waveforms as the MOSFETs directly using an oscilloscope probe?
>>>> What do those waveforms look like?
>>>> Is that voltage transient present on the output of the TL494 even when not connected to the TC4422s?
>>>>
>>>> I've built many coils using the TL494 and have not seen this problem before.  Did you try swapping TL494 ICs?
>>>> Are you sure the transient you are seeing is "real" and not just due to poor probes or probe inductance?
>>>>
>>>> Dan
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Hello there!
>>>>
>>>> My name is jason and i am a first time
>>>> coil builder. my partner chris who has built a small coil before had used a kit from eastern voltage research to construct an audio modulated coil for a project, and we decided to take it to the next level for a project we are both now working on. we are using a design that is also from eastern voltage research but the kit we were interested in was no longer available because he was having trouble getting the PCB's for this particular kit. so we got ahold of the owner and had him send out the parts for his plasmasonic 2 kit and decided to try building it ourselves.
>>>>
>>>> We built a 12v power supply w an LM7812 and using this to power the low level electronics. this powers a TL494 which seems to be what most are using for this type of circuitry. out of the TL494 we are getting about a 13MHz transient off of the outputs of the TL494. this seems to be ok but as the signal then goes through the gate drivers, which are TC4422's, this transient gets reflected at the pulse and when it turns off. this is where our problem starts getting worse! obviously w these transient signals, the A/B h-bridge we are using to power our coil is firing the mosfet's at the wrong times.
>>>>
>>>> Both of us have experience w electronics and are both attending classes at UMO for a bachelors in electrical engineering technology, just so you have an idea of what skills we may have attained. we are attempting to integrate snubber circuitry to the gate driver outputs to reduce the noise for a smoother signal, but have not yet used them in any of the labs and/or projects we have taken on. i have attached the schematics that eastern voltage research provided and some pictures of our waveforms (1 of the gate driver outputs, and the rest are of the signal at the gate of the
>>>> mosfets) that we hope will spark some interest among the experienced coil builders of this forum. i signed up for the mailing list on this site and have enjoyed reading, and have learned quite a bit from the discussions here. we would appreciate any help that you all would have for us rookies.
>>>> It seems that through our experiences that this snubber circuitry should be placed in the h-bridge from gate to drain on all 4 of the mosfet's. i will be trying a number of different things, but we would really appreciate any advice!
>>>>
>>>> (FYI- we are building this prototype mostly on breadboard. the H-Bridge is constructed on PCB that we have soddered.)
>>>>
>>>> Thank you for your time and i look forward to any response.
>>>> Jason Hill & Chris Geele
>>>> If there are any other pictures of the waveforms or the actual circuitry/breadbord/pcb that are needed i would be happy to post them
>>>>
>>>> I wasn't able to post attachments of schematics or waveforms and the schematics are not available to the public as they are property of eastern voltage research....if you would like to see the schematics or the pictures i have i would be happy to send them to you. i was not aware that i could not use attachments on this mail listing.
>>>>
>>>> _______________________________________________
>>>> Tesla mailing list
>>>> Tesla@xxxxxxxxxx
>>>> http://www.pupman.com/mailman/listinfo/tesla
>>>> _______________________________________________
>>>> Tesla mailing list
>>>> Tesla@xxxxxxxxxx
>>>> http://www.pupman.com/mailman/listinfo/tesla
>>> _______________________________________________
>>> Tesla mailing list
>>> Tesla@xxxxxxxxxx
>>> http://www.pupman.com/mailman/listinfo/tesla
>> _______________________________________________
>> Tesla mailing list
>> Tesla@xxxxxxxxxx
>> http://www.pupman.com/mailman/listinfo/tesla
> _______________________________________________
> Tesla mailing list
> Tesla@xxxxxxxxxx
> http://www.pupman.com/mailman/listinfo/tesla
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla