With a single transformer psu I have just one DC+ and one DC- output. So what I need to do is to connect both V+ rails of the two channels to the same DC+ on the psu (and the same goes for the two V- on the DC-). Is it correct?
There is also the option of a single transformer with four separate outputs. For example, Toroidy provides these on special order without much additional cost and delivery time. Then you can have dual mono and a simple layout, see post 3250.
Thanks! How much filter capacitance per rail (71v, 240w into 8 ohm and 430w into 4 ohm) would you suggest?
Self made. PCB's are old ugly pertinax brown plates. Painted with black mat spray.
Yesterday I had some spare time so I started testing boards without power transistors.
The first board went smooth. No problems.
The second 😊 nope. Power supply was +-43V, 4led’s on (D10and D11 no on both boards), bias was ok, 500mV and no problem raising to 1,5V on both boards.
Adjusting CCS1 to 5V no problem.
CCS2 – TP-3 and TP-4, NO, here the problems started. 200mV.
Checked TP-105 to positive: 20V, and not 1,4V like on the other board. The voltage after the Q101 was around 20V and not 1,4V les than 43V power. The negative rail was OK.
To make it short, Q101 out, measure with tester, is OK, turn around and again in the board. This time voltage at TP-105 was ok, 1,4V.
But the problem is still there. CCS2 – TP-3 and TP-4 was around 300mV, 40V on output terminal, and after some measuring voltages at PD+ and ND-, both around 40V I decided that is too late for continuing the agony.
Clearly there is some mistake or damaged part in IPS. Will find some time these days to power the board with 64V and check the right measurements like indicated on schematic. Maybe Q10 or Q11.
Any ideas what can be wrong to shorten the searching case?
The first board went smooth. No problems.
The second 😊 nope. Power supply was +-43V, 4led’s on (D10and D11 no on both boards), bias was ok, 500mV and no problem raising to 1,5V on both boards.
Adjusting CCS1 to 5V no problem.
CCS2 – TP-3 and TP-4, NO, here the problems started. 200mV.
Checked TP-105 to positive: 20V, and not 1,4V like on the other board. The voltage after the Q101 was around 20V and not 1,4V les than 43V power. The negative rail was OK.
To make it short, Q101 out, measure with tester, is OK, turn around and again in the board. This time voltage at TP-105 was ok, 1,4V.
But the problem is still there. CCS2 – TP-3 and TP-4 was around 300mV, 40V on output terminal, and after some measuring voltages at PD+ and ND-, both around 40V I decided that is too late for continuing the agony.
Clearly there is some mistake or damaged part in IPS. Will find some time these days to power the board with 64V and check the right measurements like indicated on schematic. Maybe Q10 or Q11.
Any ideas what can be wrong to shorten the searching case?
Hi David,
Have you done all of the checks in the build guide?
Have you checked the transistors are all installed the correct way
Have you checked the on board heatsinks do not have a short to any of the transistors.
Just a few things to check
- Dan
Have you done all of the checks in the build guide?
Have you checked the transistors are all installed the correct way
Have you checked the on board heatsinks do not have a short to any of the transistors.
Just a few things to check
- Dan
Wait so - instead of it being:There are also different resistor values to accommodate for the different temperature coefficients between the EF3-3 and the EF3-4 due to the drivers being mounted on the main heatsink (EF3-4) or the separate driver heatsink (EF-3).
The EF3-3 values from the BOM would be used if EF3-4 for amplifier boards are mounted perpendicular to the main heatsink.
Its all been tested and varied many times.
You're saying these values should in fact be differentiated by where Q103 is placed, and not by what board one is using? i.e. your comment suggests that if someone were perpendicular mounting an EF3-4, and Q103 is placed on an output BJT, that the EF3-3 values should be selected here? This is not made clear in the build guide.
Not a criticism, was just re reading due to noticing curiously that the Bias on my EF3-3 amp (with Q103 on output BJT) always starts out low - say, around 25mV, and works its way up to target 40-44mV at saturation, and my EF3-4 amp (with Q103 on main heatsink between drivers) always shoots up quickly to around 55-60mV and works its way back down target 40-44mV.
I really want to try and build the wolverine using SMPSs, but I’m a little green about them, having never used them in any of my other builds.
I also want to build monoblocks, so 2x SMPS, targeting 200W@8ohm
An SMPS with 70v output at 800w potentially feels like overkill, at 90% efficiency that will give me 10.29A.
Can anyone give me some guidance on right-sizing the SMPS, or recommend one from their own build.
I also want to build monoblocks, so 2x SMPS, targeting 200W@8ohm
An SMPS with 70v output at 800w potentially feels like overkill, at 90% efficiency that will give me 10.29A.
Can anyone give me some guidance on right-sizing the SMPS, or recommend one from their own build.
Today I had some time and checked the board. Solved the problem. I used bc559 instead of 549 for q12 and q14.Hi David,
Have you done all of the checks in the build guide?
Have you checked the transistors are all installed the correct way
Have you checked the on board heatsinks do not have a short to any of the transistors.
Just a few things to check
- Dan
After change everything worked properly
So you are saying that with EF3-4 and perpendicular placement of the board one has to use EF3-3 resistor values? So R105=2k, R106=1.6k and R107=1.4k as in EF3-3. I'm just finishing EF3-3 with perpendicular board mounting to the heatsink so all the resistor values are as in the table for EF3-3 and Q103 is placed on the nearest or central power bjt. With my EF3-4, which is next on my build list, also mounted perpendicularly I have to use EF3-3 resistor values from the table above. Good to know in advance. Please correct me if I'm wrong.Wait so - instead of it being:
View attachment 1271046
You're saying these values should in fact be differentiated by where Q103 is placed, and not by what board one is using? i.e. your comment suggests that if someone were perpendicular mounting an EF3-4, and Q103 is placed on an output BJT, that the EF3-3 values should be selected here? This is not made clear in the build guide.
Not a criticism, was just re reading due to noticing curiously that the Bias on my EF3-3 amp (with Q103 on output BJT) always starts out low - say, around 25mV, and works its way up to target 40-44mV at saturation, and my EF3-4 amp (with Q103 on main heatsink between drivers) always shoots up quickly to around 55-60mV and works its way back down target 40-44mV.
I'm not saying anything, I'm waiting for the build team to clarify. Your exact situation is why I asked. Due to me going back and finding that older comment from Staurt, it is suddenly not clear (if one uses the BOM and build guide only) what bias resistor implementation to use in an "EF3-4 but mounted perpendicular to main heatsinks" arrangement.
Last edited:
I haven't used these paticular ones, but have used the 1200W variant very successfully in a stereo build, you could look at the 400W version hypex smps400a400 if wanting to go monoblock? +/-63V not +/-70V, may still get you 200W/8R thoughI really want to try and build the wolverine using SMPSs, but I’m a little green about them, having never used them in any of my other builds.
I also want to build monoblocks, so 2x SMPS, targeting 200W@8ohm
An SMPS with 70v output at 800w potentially feels like overkill, at 90% efficiency that will give me 10.29A.
Can anyone give me some guidance on right-sizing the SMPS, or recommend one from their own build.
Last edited:
@mainframe99, thank you. For some reason, I can't quote your post.
I have seen that you can order customised outputs on the Connex SMPS. So, I might go that route. They're the only guys I have noticed that can offer this. I lean toward Cobra or Hypex because they are the most mentioned, and then maybe I do need to reconsider the 70v idea.
That would open the door to 500w and 600w. Just need to figure out what the product's R, RxE and RS denominators denote, but that is easy enough to solve.
I have seen that you can order customised outputs on the Connex SMPS. So, I might go that route. They're the only guys I have noticed that can offer this. I lean toward Cobra or Hypex because they are the most mentioned, and then maybe I do need to reconsider the 70v idea.
That would open the door to 500w and 600w. Just need to figure out what the product's R, RxE and RS denominators denote, but that is easy enough to solve.
copy,quote,past@mainframe99, thank you. For some reason, I can't quote your post.
You can't quote if you reply right after someone's post. What would be the point in quoting?@mainframe99, thank you. For some reason, I can't quote your post.
I have seen that you can order customised outputs on the Connex SMPS. So, I might go that route. They're the only guys I have noticed that can offer this. I lean toward Cobra or Hypex because they are the most mentioned, and then maybe I do need to reconsider the 70v idea.
That would open the door to 500w and 600w. Just need to figure out what the product's R, RxE and RS denominators denote, but that is easy enough to solve.
Yes you can.... see.You can't quote if you reply right after someone's post.
- Home
- Amplifiers
- Solid State
- DIY Class A/B Amp The "Wolverine" build thread