Now I have finished both mono blocks with new Schottky diode bridge PCBs. For mains AC I use my re-generator which has regulated 230 VAC and stays stable at 230.6 VAC. My rails at both mono blocks are -+24.78 VDC -+0.01V. I think that is quite impressive. I have 1V more than when using the 35A diode bridges and more than 2V more from when I used the SiC diode bridge PCB's. All was connected to the regulated AC so measurements are "comparable". To get +-24.8V DC from 2x18 (18.6) VAC in a CRCLC configuration with rail fuses is quite good. I think the Schottky diodes could be used without any heatsinks at all. Now with the very nice symmetry I expect the 150/250 etc. harmonics to be eliminated. Sound is also very good and speakers silent. Toroid's also very silent.
Can you humour me and do a measurement with everything connected but with the Amplifier turned off ?
Tony.
Tony.
I can do that when I get to the REW measurements.
As I remember when amp is turned off and 2i2 running and REW active some 1 kHz (test tone) is still visible at REW but with low amplitude of course.
As I remember when amp is turned off and 2i2 running and REW active some 1 kHz (test tone) is still visible at REW but with low amplitude of course.
Yes that is normal for the 1Khz bleed through. What I want to see is if there are any 50 Hz harmonics present 🙂
Tony.
Tony.
Now I made the first measurements using 2i2 with same setup as previous but with attention to if 2i2 could pickup some noise. And there is some strange things happening. The two images attached......the only difference is that 2i2 is moved 20 cm. The image with lowest peaks below 1 kHz is with 2i2 placed just on top of transformer and the other image 2i2 is placed 20 cm away from transformer. Do you think 2i2 is affected by the transformer magnetic fields?
If 2i2 should be placed far away from transformer I must use a much longer speaker cable which also can pickup some magnetic fields?
Probably the image with less noise is closets to what is coming from the amp?
But I need to improve the measurement setup.
What i noticed was that during the first 1/2 hour from when PC was powered on a lot of interruptions of the 2i2 probably caused by updating the PC, virus scan etc. Is there a way to ensure only REW has access to 2i2?
It does not help that the PC is old and slow.....
If 2i2 should be placed far away from transformer I must use a much longer speaker cable which also can pickup some magnetic fields?
Probably the image with less noise is closets to what is coming from the amp?
But I need to improve the measurement setup.
What i noticed was that during the first 1/2 hour from when PC was powered on a lot of interruptions of the 2i2 probably caused by updating the PC, virus scan etc. Is there a way to ensure only REW has access to 2i2?
It does not help that the PC is old and slow.....
Attachments
There was a request for an image if amp was turned off but 2i2 and REW still active. It is attached. If my hand just comes close to 2i2 I see a large 50 Hz peak and if I touch the input gain knob it becomes larger.....so it is very sensitive. I had to keep away not to disturb the measurement when amp is off.
Attachments
It can be affected by transformer fields of cause, if it is placed close to the transformer. But the largest problem source is shown in your setup photo.
Look how large an antenna is:
Attachments
Last edited:
Try to place your input and output signal cables to they were close to each other and don't make a large area.
Attachments
Last edited:
Are you able to repeat the plot, but:
(a) no amp input signal (REW generator set to min); amp input cable connected; amp output sensing cable and cable to 2i2 unaltered; temporarily disconnect the OUT connection within the interface module such that no signal gets back to the 2i2, but the 2i2 still has a gnd connection back to the amp. This allows a 0V loop from the 2i2 out to the amp and back in.
(b) same as (a) but with nominal amp input signal level. This allows the amp internals to have normal signal and 0V circuit currents flowing.
Is the laptop connected to its charger during these tests?
I always sit my EMU USB interface in an aluminium tray (which can have an inverted tray as a top cover), and the signal input plug sleeve is electrically alligator clipped to the tray. This acts as a shield of sorts. The aluminium trays are what we refer to as BBQ trays, or for food serving.
(a) no amp input signal (REW generator set to min); amp input cable connected; amp output sensing cable and cable to 2i2 unaltered; temporarily disconnect the OUT connection within the interface module such that no signal gets back to the 2i2, but the 2i2 still has a gnd connection back to the amp. This allows a 0V loop from the 2i2 out to the amp and back in.
(b) same as (a) but with nominal amp input signal level. This allows the amp internals to have normal signal and 0V circuit currents flowing.
Is the laptop connected to its charger during these tests?
I always sit my EMU USB interface in an aluminium tray (which can have an inverted tray as a top cover), and the signal input plug sleeve is electrically alligator clipped to the tray. This acts as a shield of sorts. The aluminium trays are what we refer to as BBQ trays, or for food serving.
Last edited:
I see a good picture and no visible problem on this spectrum. I mean I see quite a good setup.
The coaxial cable you use now - is good enough (have a good screen), I think. The only problem with it - it is rigid so it is a bit hard to manage it right.
Last edited:
It is ok in general for the sensitive measurement system to be sensitive. Possibly, partly or mostly it is because your measurement setup doesn't have a connection with a ground. I'm not sure from your photo if your mains cable is two-wire or three-wire?
My mains cables are 3-wire but the last one to the mains outlet is only 2-wire because there is no PE in mains outlet. But when I build something I always build like I had 3-wire mains outlet.
I tried to disconnect laptop PSU during measurement but it made no difference.
When I moved the 2i2 further away I made the antenna area larger and therefor one of the images shows more noise......is my conclusion based on above "antenna comments".
I have more flexible screened wires which are for internal connections. So I could build more flexible RCA cables so the loop can the closed. I will "play" a bit with measurement setup now I know what the problem is and see what I can achieve.
It seems the "antenna" picked up the field from amp as it looks ok when amp was off. First thing I will try is to move setup away from amp and I will try to close the loop as much as possible. My Supra speaker cables has a twisted configuration which should pickup less noise than just parallel configuration so I will try using a Supra speaker cable from amp to box to get away from amp fields. The cable is 1.5m. I will give it a try.
I tried to disconnect laptop PSU during measurement but it made no difference.
When I moved the 2i2 further away I made the antenna area larger and therefor one of the images shows more noise......is my conclusion based on above "antenna comments".
I have more flexible screened wires which are for internal connections. So I could build more flexible RCA cables so the loop can the closed. I will "play" a bit with measurement setup now I know what the problem is and see what I can achieve.
It seems the "antenna" picked up the field from amp as it looks ok when amp was off. First thing I will try is to move setup away from amp and I will try to close the loop as much as possible. My Supra speaker cables has a twisted configuration which should pickup less noise than just parallel configuration so I will try using a Supra speaker cable from amp to box to get away from amp fields. The cable is 1.5m. I will give it a try.
Also I can try to screen the USB cable as suggested above. I have copper tape than may be used for that? I can experiment....
This is where if you use a balanced to se converter and an se to balanced converter so that you are running fully balanced both out of and into the 2i2 it should eliminate at least some of that noise pickup. that way you can make the non balanced connections short, and you can have a decent length on the balanced ones so you can get the focusrite away from the DUT as well.
As also mentioned by someone earlier, some sort of plate with a connection to the earth of your TRS that everything sits on should also help. I was surprised how much difference it made to measurements of circuits on my breadboard with a 3mm aluminium plate under it. Steel should be even better.
Tony.
As also mentioned by someone earlier, some sort of plate with a connection to the earth of your TRS that everything sits on should also help. I was surprised how much difference it made to measurements of circuits on my breadboard with a 3mm aluminium plate under it. Steel should be even better.
Tony.
When you write TRS is it then Tip, Ring, Sleeve of the Jack converter or does TRS mean something else in this context?
Yes TRS tip ring sleeve. So you connect a wire from the sleeve to the plate. If you don't have fully balanced, just connect tip and ring to the hot and earth of your DUT, and it will be better than if ring is shorted to sleeve.
Tony.
Tony.
I don't think there is a strong need to change cables. I think there is only a need to manage them a bit.
If there is an in-built possibility to use balanced inputs somewere - then it is will help too (of cause).
But in general I think TS has good enough spectrums (and so measurement setup) now. At least, good enough for his monoblock amps.
He just need to play a bit with it to get an experience.
If there is an in-built possibility to use balanced inputs somewere - then it is will help too (of cause).
But in general I think TS has good enough spectrums (and so measurement setup) now. At least, good enough for his monoblock amps.
He just need to play a bit with it to get an experience.
Last edited: