DIY Purifi Amp builds

[headshake]

I am curious to see what ghent comes up with.

I poked around some case makers in the USA looking for ones that have a plain faceplate. This is the current contender: https://www.hawkusa.com/manufacture...GB-0MWgYeur2N_vC58hjJ_tqjOkDJGyRoCRlsQAvD_BwE

https://www.budind.com/view/Aluminum+Enclosure/ValuLine+Series

1.75" X 17" X 8.00" - $71USD.
A faceplate with cutouts for that box is what I am working on now. Ponoko.com can do the custom plate.

I wonder what the total cost will be and if it will be less than what Ghent is offering.

 

[Deleted member 65]

If in EU I assembled a pair of nCore monoblocks 2013 using these cases. Very nice but expensive.

https://19zoll.com/en/products/aluminum-cases/

 

[JimB]

Adding the mounting holes seems easy enough, but I am not sure about the cabling to the back panel. I am unclear as to how those connections should go.

Is it likely that Ghent will have a Purifi case ready in the next month or so? I gave my old amp to my son and really need something soon.
...

I'm short on time now, but a few comments. I suggested to Ghent that they offer a case variant that would permit using the FE board set as-is. This requires them to do new case bottom holes and a new back panel (ideally a new front panel, as well). This would be the easiest to assemble, and the highest performance (if you can use the Purifi FE output sockets). I also suggested a variant with binding posts on the back (a different back panel). I don't know what they will offer, or when.

Meanwhile, using an existing Ghent NCORE-SE(200W), you'll have to do your own mounting holes in the bottom of the case. Check. What else? Two paths.
1. Get a back panel with only the power socket and then punch/drill it yourself to accommodate the Purifi FE board I/O connectors and Standby switch. This would not require any electrical work.
2. Use the standard Ghent back panel and their I/O connectors. Here, you must wire from the FE board to the case back panel, AND mount the Purifi board set back enough to provide clearance for the location of the Ghent XLR sockets. I no longer have my NCORE-SE(200W) amp to check fit (it went to a family member on Dec. 25!), but, I don't think you can get the Purifi FE board 'under' the panel mounted XLR sockets, even with all the interfering parts depopulated. Then you have to decide how you will wire from the FE to the back panel. That will be dead easy for the speaker outputs as there are free holes. You can leave those PCB speaker sockets in place, or not. The XLRs, however, are another matter. You can:
- a. tack leads on the underside of the FE board and route them to the back panel;
- b. use "pins" to plug into the FE XLRs (essentially, a plug without the body) and then wire to the back panel; or
- c. remove the FE board XLRs and wire directly from the PCB holes to the back panel connectors. You'll need some skill and technique to avoid damaging the PCB in doing this.

FWIW, my preferred implementation, and one I suggested to Ghent, was a back panel that exposed the Purifi FE XLRs, but with well positioned, panel-mounted binding posts, ideally without holes for the FE speaker sockets, which I would then remove. I see nothing to be gained by replacing the XLRs.

I've also suggested to Purifi that they offer a version of the FE board with the I/O sockets and bypass switch depopulated. No commitment to that yet, either.

That is all for now.

 

[BobPM]

Great information. I have built a few crossovers over the years, but have stayed away from modifying PCB. I think I will stay put on this until I can either get a more complete kit, or obtain buffer boards with easier connection options.

 

[JimB]

... I think I will stay put on this until I can either get a more complete kit, or obtain buffer boards with easier connection options.

If you can handle making holes in the back panel, see if Ghent can provide an NCORE-SE(200W) with only the power socket and the vent slots on the back panel (that is, without holes for the XLRs or binding posts).

 

[JimB]

...
2. Use the standard Ghent back panel and their I/O connectors. Here, you must wire from the FE board to the case back panel, AND mount the Purifi board set back enough to provide clearance for the location of the Ghent XLR sockets. ... The XLRs, however, are another matter. You can:
- a. ...
- b. use "pins" to plug into the FE XLRs (essentially, a plug without the body) and then wire to the back panel; or
- c. ...
...

I didn't mean to scare you off, @BobPM. In my second scenario, using a standard Ghent case, my suggestion about XLR pins is just to use that component of an XLR plug, wiring pin 1 to 1, 2 to 2 and 3 to 3 of the plug to the back socket. You would be making a short, special cable.

For the speaker outputs, just wire from here to the binding posts (black to black and red to red).

 

[BobPM]

Not so much scared off as cautious after reading this thread of a Purifi amp build where a drill press was used to put in the holes for the connectors. http://www.htguide.com/forum/showth...amplifiers&s=2b3cf6bd511321a5b6bc657be9e4cdf5

The process appeared to require a degree of precision I don't have the equipment to meet. The 2nd option may work although I am not sure how adding additional connectors in the connection chain will impact sound.

 

[BobPM]

Actually, some of the closeups on the above build thread show how the connectors were added to the PCB and it doesn't look much more difficult that some of the work I've done building crossovers.

 

[JimB]

Actually, some of the closeups on the above build thread show how the connectors were added to the PCB and it doesn't look much more difficult that some of the work I've done building crossovers.

I don't know about drilling holes. I see no need for that. In that thread, they show an example of wiring to the speaker outputs as I described, above. This involves soldering the wires to the PCB in the existing holes - no added connectors.

I DO see milling to make nice back panel holes (NOT needed if you use the existing Ghent back panel):

Adding a kind of custom internal XLR cable should not add significant degradation, if done cleanly. It adds one extra gold-to-gold contact point. You could implement this by buying a short, pre-made male-to-male XLR cable, cut it in half to make two cable ends, remove the outer shell parts and wire the cut ends to the back panel connectors. However, this should really only require an inch or two of wiring in the Ghent case.

In that other thread, it appears they tack soldered the XLR leads under the FE PCB:

 

[JimB]

From the NEW FE manual (v1.10), which will be on the Purifi website, shortly, when removing the speaker output connectors to remote them (such as to binding posts on the back panel), an 0603 size 0 ohm jumper should be fitted at each connector (4 total):

This implies a feedback connection, through the connector, which I cannot see on a populated board (I have not removed any of these, yet). So there might be an easier method for this with a wire jumper.

Update Jan. 8, 2020: Indeed, as we have since seen, the second connection of the DELTRON socket IS the feedback connection point.

Update Feb. 25, 2020: Purifi confirms that one can also just use a wire jumper between the feedback hole and any of the three output holes, rather than adding 0R SMD resistors. Two examples, in green:

Also, when using remote sensing, off the board, one can install a 1 ohm resistor in that location as a backup against the amp going open-loop if ever the external sense connection was broken. There should be enough room in the feedback hole to accommodate a sense wire and a resistor lead. Alternatively, one can add such backup resistors as 0603 parts on the SMD pads for R8&9_A1&2.

 

[JimB]

Regarding input stage gain, the new manual also gives us what we need, and is essentially what @pos described for the NC500 evaluation board. But in contrast to that board, the FE02A has a low pass input filter, as noted by @March Audio. It seems awfully high, though (16MHz without source impedance?). Any gain rolls off above 48KHz.

The following has been updated per Purifi, Rev. 1.11, Jan. 4, 2020.

R17 is a 0603 part, and Purifi uses and recommends the Panasonic ERA-3A series. So, for less buffer "pregain" (7 dB, say, instead of 14.6, to give a total gain of ~20dB), one would need to remove R17 and replace it with 5.36K (using available 0.1% tolerance).

A picture of the confirmed location:

Not trivial to replace.

 

[JimB]

Finally, more details are given on the voltage regulators, as well.

I'm liking this FE02A board, more and more. The only thing I really don't like is the "standby" switch. It makes sense for some bench testing, but I just don't see a point to it for us on an everyday amp. It's something to go wrong and requires a panel hole 'breaking' into that of the adjacent output socket. And removing it without damaging the PCB will be a pain for most people. I think that function is best handled by the SMPS1200 through pin 1 of J3 here.

From anyone who has already implemented this, can you confirm the interaction of S1, shunt R3 and J3p1?

Updated Jan. 8, 2020. Per Purifi today, the User Guide has an error about removing R3 in case of implementing external standby control. R3 is required to link the signal to both channels, and should therefore be left in place. It is R4 that adds an alternate amp enable path, and which should be left unpopulated in normal use cases (as it is by default). S1 and J3p1 are, in fact, just parallel paths to ground, so either being active makes the amp active.

 

[JimB]

Corrected per new Purifi Rev. 1.11 information, Jan. 4, 2020:
People have stated a desire for the buffer of the EVAL1 to provide half the standard gain, so that it would require 4V input to reach its rated power output, rather than 2V. That requires cutting the total gain to 20dB: ~13 dB from the 1ET400A, and 7dB from the EVAL1 FE02A. That corresponds to increasing R17 from the existing value of 1.5K to 5.36K (using available 0.1% R values).

 

[Massimo]

From the NEW FE manual (v1.10), which will be on the Purifi website, shortly, when removing the speaker output connectors to remote them (such as to binding posts on the back panel), an 0603 size 0 ohm jumper should be fitted at each connector (4 total):

View attachment 44509
This implies a feedback connection, through the connector, which I cannot see on a populated board (I have not removed any of these, yet). So there might be an easier method for this with a wire jumper.

I would think the best solution is to retain the existing speaker connectors and add Speakon connectors, soldered to the free PCB holes near the corners of the Deltrons. This should not (I expect) require any changes to resistors on the board.

 

[Massimo]

From the NEW FE manual (v1.10), which will be on the Purifi website, shortly, when removing the speaker output connectors to remote them (such as to binding posts on the back panel), an 0603 size 0 ohm jumper should be fitted at each connector (4 total):

View attachment 44509
This implies a feedback connection, through the connector, which I cannot see on a populated board (I have not removed any of these, yet). So there might be an easier method for this with a wire jumper.

Can you please post a link to the new Eval1 User Guide.

 

[JimB]

I would think the best solution is to retain the existing speaker connectors and add Speakon connectors, soldered to the free PCB holes near the corners of the Deltrons. This should not (I expect) require any changes to resistors on the board.

Right. But retaining the connectors means offsetting the board back from the rear panel and then the XLRs won't fit right. Unless... you mean to have the existing connectors available through the back panel AND add speakONs? Genius! Only those who want spades, or direct wiring will be left wanting. Actually, if there is space to add speakONs on the back panel, there should be space to add binding posts.

 

[JimB]

Can you please post a link to the new Eval1 User Guide.

It's just here
https://purifi-audio.com/eigentakt/eval1/
but, it is still Rev 1.00. I got an advance copy of Rev 1.10 by email, today, as a means of answering a few specific questions I had asked them before Christmas. They said Rev 1.10 would be on their website, 'shortly'. I have shared here the aspects that addressed my questions to them. I hesitate to share the whole thing directly, in case they are still revising it before posting. I'll email them and ask if they are okay for me to share it before they post it.

Purifi informed me that there is an error in the gain information of Rev 1.10 and asked that I wait for them to correct it before sharing the whole document. I have made the correction in my earlier posts.

 

[Massimo]

Right. But retaining the connectors means offsetting the board back from the rear panel and then the XLRs won't fit right. Unless... you mean to have the existing connectors available through the back panel AND add speakONs? Genius! Only those who want spades, or direct wiring will be left wanting. Actually, if there is space to add speakONs on the back panel, there should be space to add binding posts.

I intend to CNC holes in the back panel for the existing XLR and speaker connectors. In addition, I will CNC holes above each of the speaker connector pairs to accomodate Speakon connectors.

 

[JimB]

I intend to CNC holes in the back panel for the existing XLR and speaker connectors. In addition, I will CNC holes above each of the speaker connector pairs to accomodate Speakon connectors.

Please share more as you progress.

 

[tmortimer]

I've lurked here for a while and decided to register for this thread since I've been mulling over a build with this eval kit for a few weeks.

A couple things I can't figure out:

The smps1200a400 Vaux is 17/22/25, where as the Purifi eval board needs 16.4/18/19.8 for the secondary supply. How does this end up working out? Is it because it's regulated on the Purifi side?

Second: To support a 12v trigger signal would you need an independent power supply and a circuit to invert the trigger signal which is then connected to the standby pin on the SMPS?

Thanks
Tom