Hi Owen. The capacitance is in fact less than specified in the circuit - Pana FC 1000/63, raw dc is about 53v. Not sure if the bypasses might have anything to do with it but they are as specified.
My guess is that the short circuit protection in the regulators kicks in while trying to charge the output caps and the particular make of regulators may be to blame. The ones i use say OnSemi and were most probably bought in Joburg, so there is a decent chance they are fake.
Nikola, certainly keen to find out what you did. Already ordered higher spec regulators...
My guess is that the short circuit protection in the regulators kicks in while trying to charge the output caps and the particular make of regulators may be to blame. The ones i use say OnSemi and were most probably bought in Joburg, so there is a decent chance they are fake.
Nikola, certainly keen to find out what you did. Already ordered higher spec regulators...
Based on the above I would hazard a guess that you either have fakes, or possibly just poor specs on the on-semi parts.
I always use the Fairchild 317/337 parts, but I'm not sure what's available where you are.
Both yourself and Nikola are from the same area, so maybe it was a bad batch, or fakes...
Anyone else seen this issue?
Pick up either the Fairchild parts, or the LT parts as suggested and see it that fixes the problem.
Cheers,
Owen
That certainly makes excellent sense but i am inclined to think i brought mine from South Africa... probably even more likely to be fakes.
Just took off the Lundahl transformers and am back to single phase cap coupled. There is simply no comparison, never thought the amorphous Lundahls were so bad.
Normally, Lundahls transformer require long run-in time, more than 300 hours.
Hi Owen,
- What is the input DC voltage? - 50V
- What is the output DC voltage? - 43V
- What is the input capacitance? - 1000 uF/100V Nichicon FW bought from digikey
UFW2A102MHD Nichicon | 493-3185-ND | DigiKey
- What is the output capacitance? - same as the input ones
- What is the bypass capacitance? - I suppose it is the adj. terminal capacitance - 10uF.
- What 317/337 brand / type are you using? - pictures attached.
I tried to cope with the problem with reversed schottky diodes at the regulator outputs. Did not solve the latch-up completely.
Tried to use tracking (see sch.), but decided I'd better not, because of cross noise/hum issues.
Solved the problem with thermistors at secondary side of the transformer windings. Before rectifier diodes.
It's funny that the only guys having problems are from Bulgaria, those evil bastards
The only thing that I can think of is that the LMs are fakes and that caused the problems, everything else was bought from digikey, I am not sure where the LMs were sourced from... Anyway, it's been working for almost a year without a problem and sounds wonderful, that is why I've never mentioned it, it was clear that it had to be taken care of in one way or the other.
- What is the input DC voltage? - 50V
- What is the output DC voltage? - 43V
- What is the input capacitance? - 1000 uF/100V Nichicon FW bought from digikey
UFW2A102MHD Nichicon | 493-3185-ND | DigiKey
- What is the output capacitance? - same as the input ones
- What is the bypass capacitance? - I suppose it is the adj. terminal capacitance - 10uF.
- What 317/337 brand / type are you using? - pictures attached.
I tried to cope with the problem with reversed schottky diodes at the regulator outputs. Did not solve the latch-up completely.
Tried to use tracking (see sch.), but decided I'd better not, because of cross noise/hum issues.
Solved the problem with thermistors at secondary side of the transformer windings. Before rectifier diodes.
It's funny that the only guys having problems are from Bulgaria, those evil bastards
The only thing that I can think of is that the LMs are fakes and that caused the problems, everything else was bought from digikey, I am not sure where the LMs were sourced from... Anyway, it's been working for almost a year without a problem and sounds wonderful, that is why I've never mentioned it, it was clear that it had to be taken care of in one way or the other.Attachments
Last edited:
This particular pair has been running for even longer. To be honest my hearing is probably not so sensitive as i hear very insignificant change in transformer sound over time.
My LMs look nothing like yours, as mentioned previously they are marked OnSemi. Last night hooked a variac and reduced the incoming mains voltage by about 10% - all latch-up problems seem to have gone away.
Somewhere in the back of my mind i have some strange recollection that not all 317s work the same as far as the protection goes. Some may limit the output current to the max allowed, while others shut down and only a hard reset will get them going again. Not sure if this has any basis in reality but i think it's not the first time i encounter this problem.
A low ESR capacitor, coupled with a low DCR mains transformer may easily exceed 1.5A of charging current during start-up. Why the variac fixes this i am not quite sure, nor am i sure if the high input voltage has any effect on the protection circuit.
Yup. I guess what the variac achieves is a reduction in the rate of change of the output voltage thanks to its added inductance and has nothing to do with the actual voltage.
Nikola's thermistors achieve the same. Of course it's a better idea to use voltage regulators with a more tolerant short circuit protection.
Part of the issue is no doubt the 300VA mains transformers i use. Resistance of the secondaries is about 0.1R
Nikola's thermistors achieve the same. Of course it's a better idea to use voltage regulators with a more tolerant short circuit protection.
Part of the issue is no doubt the 300VA mains transformers i use. Resistance of the secondaries is about 0.1R
Last edited:
You could definitely be on to something with that... I use dual 50VA toroids for my setup which have a much higher secondary DCR than a 300VA unit would.
Also, if Nikola solved his issue with thermistors, then that's a pretty strong indicator that it is tied to secondary DCR.
The rest of both setups sounds perfectly reasonable, and I can't think of anything other than the above that could be causing it, unless of course the parts are fake.
I attached pictures from a known good batch of Fairchild parts which should match the pictures taken by Nikola. There are a handful of suspicious differences, but if he bought his 10 years ago and they were diffused in different places, then it could just be changes in production. I have attached pictures for reference.
Anyhow... Perhaps Nikola can confirm what size toroid he is using so we can send out a general warning to everyone to avoid high VA toroids, or adjust the circuit accordingly.
Cheers,
Owen
Attachments
Remember at startup the cap keeps the output briefly at zero, so max dropout can also be exceeded. It may well be a combination of exceeding both current and voltage limit of the regulator which in some makes triggers the protection.
Seems like my heatsink is a bit too small and in order to keep temperature below 60C had the voltage output reduced to 41v. It may be a fluke but at this output voltage start up seems trouble free.
better than failing open =) the options at this voltage are not many and varied, but its not an uncommon use, i've seen people running them like this in tube amps in the hundreds of volts, you just need to keep their limits in mind.
its probably a combination of all these things, the 300VA tranny certainly wont be helping lol. I killed a set of the LT when I moved to this place before I realized my mains was significantly higher than 230v at night here and funnily enough it was with a single low DCR 150VA toroid. once I set a higher Vout of +/-48v all was merry again and thats the only time i've had any problems, I subsequently changed to 2 x 50VA units to lower the heat and havent had a problem since.
its probably a combination of all these things, the 300VA tranny certainly wont be helping lol. I killed a set of the LT when I moved to this place before I realized my mains was significantly higher than 230v at night here and funnily enough it was with a single low DCR 150VA toroid. once I set a higher Vout of +/-48v all was merry again and thats the only time i've had any problems, I subsequently changed to 2 x 50VA units to lower the heat and havent had a problem since.
it seems there must be a lack of communication going on, or some breakdown in communication. also I have not heard back from bigpandahk from the email I sent on late Friday night/Saturday morning early ~12.30AM AEST with final postage quotes despite chasing me up last week; so Chan, please get back to me Asap so I can get these all out the door, I have the payments from the USA and EU already, so just waiting for that to clear to my bank and the payment from Asia. Everything is packed and ready to go.
Chan, please confirm you have received my email or not so I can resend, perhaps to another account, I sent it as a reply to your paypal account email. Perhaps you are on holiday somewhere for CNY?
I presumed that the EU guys were all in the loop, please nobody else email me for updates, from either late tomorrow afternoon, or Wednesday morning, it will nolonger be in my hands. Emphrygian will be taking care of it from then on for the EU. Misterogers for USA/CAN and BigPandahk for Asia.
Chan, please confirm you have received my email or not so I can resend, perhaps to another account, I sent it as a reply to your paypal account email. Perhaps you are on holiday somewhere for CNY?
I presumed that the EU guys were all in the loop, please nobody else email me for updates, from either late tomorrow afternoon, or Wednesday morning, it will nolonger be in my hands. Emphrygian will be taking care of it from then on for the EU. Misterogers for USA/CAN and BigPandahk for Asia.
Last edited:
Hi
I didn't receive any email from you and I've check the junk mailbox. Please PM the details to me again. I will pay the shipping with full insurance as soon as I receive the amount.
Regards
Chan