This is common with transformers - because normally you overrate transformer 2x needed DC-current. (i even go up too 3x ratad current when using class-a / CCS circuit..) so since we never use full rated current, there is less internal loss in transformer windings - and hence higher secondary voltage than specified. I have only bought one transformer that did undervolt the secondaries... and that was cheap a chinese r-core..
that seems about right.Ryssse, my mains is 241v and I get 26.4 with no resistors, same chokes, traffo and diodes. Those talema transformers are potted toroids.
Vout = 241/230*22*(1+0.15) = 26.51Vac for a 230:22Vac with 15% regulation running on 241vac.
14% regulation gives ~26.3Vac
Ordered 2 new transformers (2x25v)as I don´t wanna start "customazing" the CCS.
Decisive action, I like that
And the first thing you do is power ON via a Mains Bulb Tester and measure the input and output voltages without any load connected.Ordered 2 new transformers (2x25v)as I don´t wanna start "customazing" the CCS.
From there you can calculate/predict the lowest and highest no load voltages from this new transformer.
Finally got around to making a video of my Paradise starting up. Before I fitted the Calvin boards all the LEDs came on together. But now the set under the heatsink light up on power on...the two side sets take their time.
Anyway, the video...
https://youtu.be/1MfZBAg_-kE
The PSU caps have no effect on the start-up, I just trying difference capacities atm from my bit box.
Each board has +/-27v from the PSU.
Anyway, the video...
https://youtu.be/1MfZBAg_-kE
The PSU caps have no effect on the start-up, I just trying difference capacities atm from my bit box.
Each board has +/-27v from the PSU.
Last edited:
To me it seems that the PSU goes in a constant-current mode, while the circuit tries to use more current than the CCS part of the PSU is happy to deliver.
Then, after a while, parts of the circuit heat up, capacitors get filled, voltages do stabilize, current slow down, and the demand of current start to diminish.
From then on, the excess CCS current is being shunted to ground by the voltage regulator (shunt regulator) and the voltage regulator starts delivering its intended voltage.
If this takes a long time, it would indicate that the current delivered by the CCS is on the low side (to say the least), or OTOH, it may be that some part of the circuit takes long (maybe to long) to stabilize and is using the PSU CCS as some fail-safe boot-up protector.
In the first case, the CCS is delivering not enough current, the fix is simple, lower the current sense resistor. But there is a danger, the second case could be true.
In the second case, the CCS delivers enough, one needs to examine where the current goes and why, when this is know fixes can be made. Because the voltage regulator, after a while, stabilizes it seems to me that this (the second case) is the problem that needs to be solved.
The way to go about this, program an lab-PSU to deliver the same maximum current as the CCS and also the same max voltage as the shunt PSU. Connect parts of the load to this PSU and make a graph of time against current and voltage for this part. By this find out where the problem is.
Then, after a while, parts of the circuit heat up, capacitors get filled, voltages do stabilize, current slow down, and the demand of current start to diminish.
From then on, the excess CCS current is being shunted to ground by the voltage regulator (shunt regulator) and the voltage regulator starts delivering its intended voltage.
If this takes a long time, it would indicate that the current delivered by the CCS is on the low side (to say the least), or OTOH, it may be that some part of the circuit takes long (maybe to long) to stabilize and is using the PSU CCS as some fail-safe boot-up protector.
In the first case, the CCS is delivering not enough current, the fix is simple, lower the current sense resistor. But there is a danger, the second case could be true.
In the second case, the CCS delivers enough, one needs to examine where the current goes and why, when this is know fixes can be made. Because the voltage regulator, after a while, stabilizes it seems to me that this (the second case) is the problem that needs to be solved.
The way to go about this, program an lab-PSU to deliver the same maximum current as the CCS and also the same max voltage as the shunt PSU. Connect parts of the load to this PSU and make a graph of time against current and voltage for this part. By this find out where the problem is.
Last edited:
On your second case, the slow turn on of the LEDs only occurs when the Calvin board is fitted and if I remember correctly I had to tweak the pot on the Calvin board to get the side LEDs on at all - however this maybe a red herring and I need to double check this by turning the pot to each end stop and powering up the board. I'm just wary of popping some component on the boards.
The boards, when running sound fabulous...
Sent from my SM-G930F using Tapatalk
The boards, when running sound fabulous...
Sent from my SM-G930F using Tapatalk
And without the Calvin boards the LEDs light up with no delay?
Sent from my SM-G930F using Tapatalk
