Replace the Output Coil / Filter
Dear All,
first sorry for the question if it has been alreaday answered before in this long thread ...
Did someone replace the output filter coils in the ucd180 and does
someone have experience with it ?
Would it be worthwhile to do so ?
I think especially integrating the filter-response of an 12dB filter
into 2 ucd180's for direct speaker filtering ...
Let us say - about 2800 Hz with 12dB for Bass and 6 or 12 dB
Highpass for Highs.
Thank you.
😎
Dear All,
first sorry for the question if it has been alreaday answered before in this long thread ...
Did someone replace the output filter coils in the ucd180 and does
someone have experience with it ?
Would it be worthwhile to do so ?
I think especially integrating the filter-response of an 12dB filter
into 2 ucd180's for direct speaker filtering ...
Let us say - about 2800 Hz with 12dB for Bass and 6 or 12 dB
Highpass for Highs.
Thank you.
😎
Re: Replace the Output Coil / Filter
Making a low-pass filter by changing the output filter of the UcD is impossible. There is a feedback loop that determines the response of the UcD, you would have to modify that feedback loop. I think it is on the small PCB mounted on the bigger PcB. You can't access the components on that PcB and if you could, you would completely modify and/or mess up the behavior of the UcD. So you would need some kind of active filter in front of the module itself. A 1st order high or low pass should not be that difficult with the opamps that are already on the module but a 2-order maybe more difficult.
Best regards
Gertjan
drummer44 said:Dear All,
first sorry for the question if it has been alreaday answered before in this long thread ...
Did someone replace the output filter coils in the ucd180 and does
someone have experience with it ?
Would it be worthwhile to do so ?
I think especially integrating the filter-response of an 12dB filter
into 2 ucd180's for direct speaker filtering ...
Let us say - about 2800 Hz with 12dB for Bass and 6 or 12 dB
Highpass for Highs.
Thank you.
😎
Making a low-pass filter by changing the output filter of the UcD is impossible. There is a feedback loop that determines the response of the UcD, you would have to modify that feedback loop. I think it is on the small PCB mounted on the bigger PcB. You can't access the components on that PcB and if you could, you would completely modify and/or mess up the behavior of the UcD. So you would need some kind of active filter in front of the module itself. A 1st order high or low pass should not be that difficult with the opamps that are already on the module but a 2-order maybe more difficult.
Best regards
Gertjan
Re: Re: Replace the Output Coil / Filter
Maybe it should be emphasized that the feedback is taken after the filter, this being one of the things that makes the UCD special.
PSU and stuff
OK, so I break the rule, I post before I search 🙂 But after reading each and every reply on this thread I don't feel capable of repeating that with the "Power supply for UCD and ZAPPulse" one.
People have talked about cap sizing, diodes etc but I feel that the transformer sizing subject hasn't been covered too much. I have ordered the main filtering caps (4 x 10000uF/63V Vishay BC, http://www.tme.pl/arts2/en/a09/ce_2200_385psr__2.html), diodes (BY329), bypass caps for rectifier bridge(http://www.wima.com/mks4.htm) for the PSU and now I have to order the transformer(s) and I'm confused by a few things. Hypex sells 160VA transformers and they advertize them as suitable for powering 2 UCD180's. This denies all I know about transformer sizing. I was planning to order a pair of 250VA ones (value computed from 95% efficiency of UCD's and 0.8 power factor), one per channel, but maybe I could save some money and spend somewhere else? Not to mention that the 250VA low-profile (as long as the UCD's are 35 mm tall why not a slim case?) ones are 15 cm wide and they somewhat restrict me on case width.
Another thing, in the "UcD400 Q & A" thread, Bruno says:
Bypassing the bridge rectifier is a MUST. On one of my first designs, emission was wholly and completely dominated by the 50Hz rectifier!
and I can swear that somewhere he said that bypassing is optional with soft recovery diodes.
And while we're at it, are the caps I ordered suitable for this task (bypassing bridge diodes)?
And third. The selection criterion for the diodes was stock availability, and these being the ones with the highest current rating fast soft recoveries my supplier had... Max supported current is 6A, peak repetitive current is 14A. Will these survive inrush?
And last 🙂 Are these types of caps (polyester) suitable for bypassing the rectifier?
Best Regards,
Calin
OK, so I break the rule, I post before I search 🙂 But after reading each and every reply on this thread I don't feel capable of repeating that with the "Power supply for UCD and ZAPPulse" one.
People have talked about cap sizing, diodes etc but I feel that the transformer sizing subject hasn't been covered too much. I have ordered the main filtering caps (4 x 10000uF/63V Vishay BC, http://www.tme.pl/arts2/en/a09/ce_2200_385psr__2.html), diodes (BY329), bypass caps for rectifier bridge(http://www.wima.com/mks4.htm) for the PSU and now I have to order the transformer(s) and I'm confused by a few things. Hypex sells 160VA transformers and they advertize them as suitable for powering 2 UCD180's. This denies all I know about transformer sizing. I was planning to order a pair of 250VA ones (value computed from 95% efficiency of UCD's and 0.8 power factor), one per channel, but maybe I could save some money and spend somewhere else? Not to mention that the 250VA low-profile (as long as the UCD's are 35 mm tall why not a slim case?) ones are 15 cm wide and they somewhat restrict me on case width.
Another thing, in the "UcD400 Q & A" thread, Bruno says:
Bypassing the bridge rectifier is a MUST. On one of my first designs, emission was wholly and completely dominated by the 50Hz rectifier!
and I can swear that somewhere he said that bypassing is optional with soft recovery diodes.
And while we're at it, are the caps I ordered suitable for this task (bypassing bridge diodes)?
And third. The selection criterion for the diodes was stock availability, and these being the ones with the highest current rating fast soft recoveries my supplier had... Max supported current is 6A, peak repetitive current is 14A. Will these survive inrush?
And last 🙂 Are these types of caps (polyester) suitable for bypassing the rectifier?
Best Regards,
Calin
You are calculating for 100% continuous max. power output from the UcD's? This never happens in music, not even 50%?
Could this be it?
I was thinking 800VA for 2 UcD400's, yet JP says 500 is enough for stereo, and Bruno even said I can party all night into 4 ohms loads with 500VA, except when I'm living in south america he said... (something about all parties there being done with heavy heavy clipping 😀 )
drummer44 said:Dear All,
first sorry for the question if it has been alreaday answered before in this long thread ...
Did someone replace the output filter coils in the ucd180 and does
someone have experience with it ?
Would it be worthwhile to do so ?
I think especially integrating the filter-response of an 12dB filter
into 2 ucd180's for direct speaker filtering ...
Let us say - about 2800 Hz with 12dB for Bass and 6 or 12 dB
Highpass for Highs.
Thank you.
😎
The filter response controls the self-oscillation of the UcD, you can´t change it.
Yes, this must be it. Still, most manufacturers use this rule when sizing transformers, invoking the reason of momentary peak power in transients (percussion for instance). It seems to be generally accepted that the rule of thumb is max. absorbed power times the power factor. If you'll ask me that's BS, and this would only hold true if listening to sine waves.
Still, doesn't the core saturate when required to deliver more than the transformer can handle, resulting in distorsion?
It's true that my (future) speakers will run out of excursion at about 20W and I only live in a flat. But I like to know that there is a reserve.
All in all, what would be a sane transformer for 2 UCD180's?
Still, doesn't the core saturate when required to deliver more than the transformer can handle, resulting in distorsion?
It's true that my (future) speakers will run out of excursion at about 20W and I only live in a flat. But I like to know that there is a reserve.
All in all, what would be a sane transformer for 2 UCD180's?
DC Protection circuits
Can the Quad 405 protection circuit used on UcD180?
See:
http://www.geocities.com/ResearchTriangle/Lab/6722/quad405cirb.html
It makes a shortcircuit to the ground in the output of an amp in case of DC. It consists of 2N4992 and SC141B devices, 10uF cap and 15k resistor. If a MOSFET breaks in UcD it propably blows also the remaining MOSFET or fuses which are between PSU and UcD module. Is this any good, or are there better solutions to protect the speakers (other than relays)?? Maybe a big bulb on the amp connected in series with the speakers. When there's a light you know it's time to change the output transistors, or at least switch off that amp for a while...
Can the Quad 405 protection circuit used on UcD180?
See:
http://www.geocities.com/ResearchTriangle/Lab/6722/quad405cirb.html
It makes a shortcircuit to the ground in the output of an amp in case of DC. It consists of 2N4992 and SC141B devices, 10uF cap and 15k resistor. If a MOSFET breaks in UcD it propably blows also the remaining MOSFET or fuses which are between PSU and UcD module. Is this any good, or are there better solutions to protect the speakers (other than relays)?? Maybe a big bulb on the amp connected in series with the speakers. When there's a light you know it's time to change the output transistors, or at least switch off that amp for a while...
use a good/fast relay+ fast/protection in the output of the amp, nothing can beat that. For me it works absolutely 100% fail/foolproof.
Used the circuit for 25 years and 100's of amps , never had a blown speaker, only amps. It is a no fuss solution.
Used the circuit for 25 years and 100's of amps , never had a blown speaker, only amps. It is a no fuss solution.
Is this good DC-protection circuit http://sound.westhost.com/project33.htm#tests ?
Some other good circuits to use with UCD?
What type of relay can you recommend?
I feel little bit affraid all the time without DC-protection. Seas W22 or Millennium are too expensive to me to blow..
Some other good circuits to use with UCD?
What type of relay can you recommend?
I feel little bit affraid all the time without DC-protection. Seas W22 or Millennium are too expensive to me to blow..
Use the UCD180's with a 225VA transformer, 2 channel...no problem at all. Have 2 UCD 400 amps, 1 with 300VA and the other with 500VA, absolutely not the slightliest difference in sound. Bass is the same. Don't know why there is somuch need for this huge transformers. Also 1 transformer is enough for a stereo device. If you have a proper earthing in your case it'll act as dual mono blocs. Separation/crosstalk has to be absolutely minimal to have a stereo amp. with depth in the sound. High frequency crosstalk sounds awfull. Try to get -100DB at 20kc crosstalk.
DC Protection
Isn't the UcD module short-circuit protected? What really happens then with the Quad type shorting DC Protection circuit when it kicks on? Will it trigger the short-circuit protection on? I'm a little bit afraid, too. The ESP #33 might work. You could also build it so that it short circuits the PSU's after the fuses, or just switches off them, but then what happens when the DC protection circuit looses the DC, will it switch on the power immediately again? The bulb protection circuit is in fact used with some PA speakers to protect the tweeters. It also gives some amplitude compression for free.
Btw, wasn't the Hypex PSU intended to have some DC protection circuit built in? Can't see it there.
Isn't the UcD module short-circuit protected? What really happens then with the Quad type shorting DC Protection circuit when it kicks on? Will it trigger the short-circuit protection on? I'm a little bit afraid, too. The ESP #33 might work. You could also build it so that it short circuits the PSU's after the fuses, or just switches off them, but then what happens when the DC protection circuit looses the DC, will it switch on the power immediately again? The bulb protection circuit is in fact used with some PA speakers to protect the tweeters. It also gives some amplitude compression for free.
Btw, wasn't the Hypex PSU intended to have some DC protection circuit built in? Can't see it there.
Nope, DC protection is in the hypex power supplies. There is over current protection in the modules.
The offset detection circuitry is too slow. It requires too much voltage and than you have welded contacts at an amp. breaking up. The speed is also determened by the threshold voltage.
Use a good differential circuit to detect the offset. You should detect offsets at 200mv.
This Quad protection is very unelegant and a very raw way of protecting. Would not trust it. The UCD modules have to be protected in a sophisticated manner=relay.
I'll probably do so. After all if we consider the peak to average ratio of (normal) music, that makes sense. And second, I won't ever need those 180 Watts, and there is one other reason for using one single transformer of power 2xP instead of two of power P (one per channel): when one channel needs some extra power that the other doesn't, it can take it from the common transformer. When you have one per channel you're more limited. And after all, if Bruno says it's OK, and a dozen of serious manufacturers do that, who am I do disagree? 😉Bgt said:
Why? I've been told most drivers can last several tens of volts for some milliseconds.Bgt said:
You mean a comparator circuit? Also why this low offset as it can't hurt speakers?
Has anyone tried the Velleman K4700 kit, is it OK?
Also which relays perform well when used in power lines?
That ESP#33 takes power directly from transformer secondaries so it's supposed to work fast (when the power is applied), but shouldn't the comparator circuit in amp output be latchable and keep the power disconnected until power is down? A schematic anywhere?
Output protection
This needs to be run by Bruno but a simple latching circuit that shorts out the output should work fine as long as it is strong enough to dump the power supply without welding its contacts. This would be the case if an output device failed. In other less severe cases causing offset, shorting the output will imeaditly shut the amp down as it can no longer oscillate, no feedback! The latching circuit could be powered from a separate source, like the new front end power supply every one is going to use, right? This would necessitate killing the power, correcting the problem then repowering.
This seems the right way to go to me rather than have a relay in the signal path or power path.
Roger
This needs to be run by Bruno but a simple latching circuit that shorts out the output should work fine as long as it is strong enough to dump the power supply without welding its contacts. This would be the case if an output device failed. In other less severe cases causing offset, shorting the output will imeaditly shut the amp down as it can no longer oscillate, no feedback! The latching circuit could be powered from a separate source, like the new front end power supply every one is going to use, right? This would necessitate killing the power, correcting the problem then repowering.
This seems the right way to go to me rather than have a relay in the signal path or power path.
Roger
for the velleman kit look here: http://www.diyaudio.com/forums/showthread.php?threadid=3438
looks ok
looks ok
Bgt said:
I have unerstood that DC-protection relays are better to connect to powersupply V- and V+ lines over loudspeaker lines (no relay contacts in singnal path).
I think Velleman K4700 can not be used in ps line, only in speaker line?
It also connects relays back on after delay. It should not do that without manual reset (main swich off etc.)?
http://www.velleman.be/Downloads/0/illustrated/Illustrated_assembly_manual_K4700.pdf
The Velleman kit does indeed swich off your speakers from the output. I don't see any disadvantage in putting a relay in the speaker lines. A good relay has a very low ohmic resistance, so I don't see a problem in that. It is a bit more practical than switching power lines.