Thanks, Bill. I will use a smaller resistor in my next build, which is now underway. Did you change any other components because of the higher voltage in the relay circuit? D4 or R23?
Peace,
Tom E
Peace,
Tom E
No other parts were changed. At 50 volts rating D4 is ok. R23 can be tuned for desired LED brightness but I did not change the value.
Following Bill's recommendation I built mine with 100R at R14 for 25V secondaries - I get 23.7v for the relay circuit.
Cheers
Geoff
I see your point. I measured once the circuit was stable. Just looking at the simple bits I can see that reducing R14 will reduce the rise time for the voltage across C14. This could lead to a brief period (at switch on and before Q1 switches the relay) when the voltage far exceeds 24v.
Bill, did your Spice simulations look at this question?
Cheers
Geoff
Omron specifies the maximum coil voltage at 170% of nominal or 40.8 volts for the 24 volt relay. This is noted to not be a continuous rating so it can be taken to be a start up transient limit.
Relays are not precision components and Omron lists a 10% variation in coil resistance (and therefore coil current) as part of normal production. The coil resistance also rises 0.4% for each degree Centigrade of temperature rise. Given this information, I tried to keep the coil voltage within 10% of rating, 21.6V - 26.4V. The Spice simulation shows a 26.2 volt peak across the relay initially. It would have been 30 volts if the circuit applied the coil voltage instantaneously but it does not. It takes about 0.2 seconds for C16 to charge up to the point that the transistors Q1 and Q2 have enough base current to turn the relay on. During the charge time the supply voltage drops from 30 volts to 26.2 volts and that is the highest voltage the relay coil sees.
The Spice results are attached, green trace is relay coil voltage and red trace is the 24V supply rail. The vertical scale is 3 volts/division and the red trace starts out at 30 volts at the left side of the plot.
Relays are not precision components and Omron lists a 10% variation in coil resistance (and therefore coil current) as part of normal production. The coil resistance also rises 0.4% for each degree Centigrade of temperature rise. Given this information, I tried to keep the coil voltage within 10% of rating, 21.6V - 26.4V. The Spice simulation shows a 26.2 volt peak across the relay initially. It would have been 30 volts if the circuit applied the coil voltage instantaneously but it does not. It takes about 0.2 seconds for C16 to charge up to the point that the transistors Q1 and Q2 have enough base current to turn the relay on. During the charge time the supply voltage drops from 30 volts to 26.2 volts and that is the highest voltage the relay coil sees.
The Spice results are attached, green trace is relay coil voltage and red trace is the 24V supply rail. The vertical scale is 3 volts/division and the red trace starts out at 30 volts at the left side of the plot.
Attachments
Bill_P -
Your posts are always very interesting...
I'm wondering if the Spice sim was done against a 68 ohm R14 or what value???
Thanks,
Your posts are always very interesting...
I'm wondering if the Spice sim was done against a 68 ohm R14 or what value???
Thanks,
Sorry, yes it was 68 Ohms with 22VAC transformer, amplifier quiescent load current as it would be during power up.
I have almost finished my amp.
I hace if inside a case well fitted, It yet needs to be fixed right. The enclosure with allocate another two channels. They are a chipamp.com kit, which sounds very nice. I will be able to make fast direct comparisons between one standar topology with "standar" components (although very good ones: panny FM, dale resistors, wimas...) and Mauro's topology with high end components.
I have my tantalum resistors already fitted and soldered in. I am trying to listen carefully, but I don't know if I am able to discern such a difference. I am not any high end audiophile tester, so I would need thousands of hours listening a system to be able to difference it with such a precision. Only two resistors in the signal path, and both of them were previously very good ones, no less than PRPs.
At first, I would say that I don't hear any difference, maybe a bit more of clarity. But I am not sure at all, sorry.
Maybe is a change so so small that not everybody is able to discern it. Or maybe it doesn't produce any change at all to put other two high end resistors.
Thank you very much Uriah for giving me the chance to try them and being part these experiments.
I will take some photos as soon as I get back my camera.
Regards,
Regi
I hace if inside a case well fitted, It yet needs to be fixed right. The enclosure with allocate another two channels. They are a chipamp.com kit, which sounds very nice. I will be able to make fast direct comparisons between one standar topology with "standar" components (although very good ones: panny FM, dale resistors, wimas...) and Mauro's topology with high end components.
I have my tantalum resistors already fitted and soldered in. I am trying to listen carefully, but I don't know if I am able to discern such a difference. I am not any high end audiophile tester, so I would need thousands of hours listening a system to be able to difference it with such a precision. Only two resistors in the signal path, and both of them were previously very good ones, no less than PRPs.
At first, I would say that I don't hear any difference, maybe a bit more of clarity. But I am not sure at all, sorry.
Maybe is a change so so small that not everybody is able to discern it. Or maybe it doesn't produce any change at all to put other two high end resistors.
Thank you very much Uriah for giving me the chance to try them and being part these experiments.
I will take some photos as soon as I get back my camera.
Regards,
Regi
Regi,
Do you mean you hear no difference between a standard LM3886 and the MyREFC or do you mean you hear no difference in MyREFC between PRP and Tantalum?
Thanks for being a part of this experiment Regi. You have been extremely helpful to so many people.
Uriah
Do you mean you hear no difference between a standard LM3886 and the MyREFC or do you mean you hear no difference in MyREFC between PRP and Tantalum?
Thanks for being a part of this experiment Regi. You have been extremely helpful to so many people.
Uriah
I mean in the comparison between the PRP and the Tantalums. I haven't mounted yet in the case the other standard gainclone. I will report soon too.
Regards,
Regi
Regards,
Regi
Glad to hear a report about the tantalums even if they make no difference in your system. $4 each is a lot for a tantalum resistor. I would not like to wish I had a bunch of tantalums in my system.
Uriah
Uriah
I wish I would have a whole system entirely made of tantalums. All stages, all signal paths. And having besides another system made of something more standard, like Vishay Dales.
Hearing to both, we could be able to perceive these differences. Maybe. Or maybe not. But the tantalum's one would worth its weight in gold 😀
Hearing to both, we could be able to perceive these differences. Maybe. Or maybe not. But the tantalum's one would worth its weight in gold 😀
For anyone needing instructions on how to build their LDR attenuator
Here are some instructions for completing the Mount and Balance board with LDRs and 5VDC supply.
Uriah
Lightspeed Instructions
Here are some instructions for completing the Mount and Balance board with LDRs and 5VDC supply.
Uriah
Lightspeed Instructions
Last edited:
By the way I put them for sale again $30 for the LDRs/Board/Components. Freaking great deal. Its on both sites so if you want to buy MyREFC boards as well the shipping combines and you save a bit.
I wont clog with more sales pitches.
I wont clog with more sales pitches.
very nice detailed instructions, with your permission I will add them to the documentation archive and upload.
Regards,
Regi
Feel Free to do so. A DIY friend put them together for me and he will be working on the instructions for the Lighter Note as well. I certainly appreciate his good work.
Uriah
Uriah
- Status
- Not open for further replies.