DLH Amplifier: The trilogy with PLH and JLH amps

Member
Joined 2010
Paid Member
I will ask the question's:

1. Is the DLH High Power Amplifier to act like a teaser?

2.Marc, have you built / tested this DLH HPA, and if so, any details?

3.Finally, if tested and all is good, is Prasi willing to generate a pcb, and share like before? or,

4. Is it possible that the extra OS's be added P2P to an existing DLH pcb.

Regards,

Myles
 
I have not built it yet.

I can assure that it will work without problems, mainly if it is operated in bridge mode, where offset displacements due to the action of temperature will no longer be a problem. The PSRR in this mode of operation will be impressive.

I tried to replicate the input circuit of the original DLH (that is, with BJTs), but when I added several pairs of mosfets to the output, it severely worsened the quality in relation to the original DLH. In that configuration, I did not find ways to improve the quality to acceptable values.

That is why I found acceptable quality values ​​also using mosfets in the input circuit. This has a price to pay: lower excursion of the signal towards negative values ​​of rail (between 6 to 12 volts of reduction of amplitude of a Vee of 24 volts).

When I can, I upload simulations.

Best regards

Horacio
 
DLH High Power THD at 1 W, 8 ohms and 1 KHz.jpg

DLH High Power AC Analysis at 8 ohms.jpg

DLH High Power PSRR at 8 ohms (single output).jpg

All these simulations are not in bridge mode.

It is expected that in bridge mode some of these parameters are still much better, which is no small thing.
 
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
Just use 120mm silent fan CPU heatnpipe coolers. 120w per mosfet can be handled no problem. On Alpha BB the mosfet body temps don’t exceed 38C at 10% fan speed. This is dissipating 200w per channel and it’s very effective (and inexpensive compared to a passive Rad).

672472d1522616478-aksa-lender-mos-hybrid-aleph-alpha-amplifier-alpha-bb-test-mono-fast-closeup-jpg


Also, instead of matching IRFPs just use a big IXYS TO264 designed to dissipate 600w each and can handle up to 80amps.
 
I would like to implement the idea of ​​xrk971: that is, replace the IRFP pairs with a single pair of IXYS giants.

I really liked the speed with which my friend prasi has presented that beautiful PCB.

Who will be the first to build this beautiful big beast ?.

Best regards

Horacio
 

Attachments

  • DLH High Power OLG vs CLG at 8 ohms.jpg
    DLH High Power OLG vs CLG at 8 ohms.jpg
    625.7 KB · Views: 442
attachment.php


attachment.php


All the simulations were carried out for the same position of the 2K5 trimpot, that is, 66% of its total displacement (considering a trimpot of linear variation). It could happen that at other adjustment positions, according to the output load, the profile of the distortion takes on both an appearance and a different value.

Best regards
 

Attachments

  • DLH High Power THD vs Frecuency (at 2 ohms and 1 W).jpg
    DLH High Power THD vs Frecuency (at 2 ohms and 1 W).jpg
    111.6 KB · Views: 752
  • DLH High Power THD vs Output Power (at 2 ohms and 1 KHz).jpg
    DLH High Power THD vs Output Power (at 2 ohms and 1 KHz).jpg
    245.2 KB · Views: 762
  • Like
Reactions: 1 user
Horacio,

Prasi has done a beautiful pcb, I agree, he is quick too and considers all the electrical issues very well....... we are lucky at DIYaudio, aren't we?

The problem with multiple pairs is matching. Difficult to match four pairs; you would need at least 50 to select from, and you would have to measure them quickly to avoid thermal warmup, which hugely changes Vgs at the selected quiescent. A big pair of monsters is always better and there are some very good ones, like the IXYS 1040W devices. Here are 18 IXYS devices, Pmos and Nmos, with 90-140A and 960W to 1060W. All are suitable for your DLH, although you might want to reconsider the gate drive requirements carefully. These are designed for tractive/propulsive/smps application, I believe; but they move pretty quickly if you can move the charge into and out of the gate swiftly. These devices have outrageous transconductance, which is pretty good in linear operation as it reduces the compressive distortions.

Ciao,

Hugh

IXFX100N65X2 IXYS MOSFET N-CH 650V 100A PLUS247 137 - Immediate
AU$20.19000 1 Tube? HiPerFET™ Active N-Channel MOSFET (Metal Oxide) 650V 100A (Tc) 10V 5.5V @ 4mA 180nC @ 10V ±30V 11300pF @ 25V- 1040W (Tc) 30 mOhm @ 50A, 10V -55°C ~ 150°C (TJ) Through Hole PLUS247™-3 TO-247-3

IXFX140N30P IXYS MOSFET N-CH 300V 140A PLUS 247 174 - Immediate
AU$25.31000 1 Tube? HiPerFET™, PolarP2™ Active N-Channel MOSFET (Metal Oxide) 300V 140A (Tc) 10V 5V @ 8mA 185nC @ 10V ±20V 14800pF @ 25V
1040W (Tc) 24 mOhm @ 70A, 10V -55°C ~ 150°C (TJ) Through Hole PLUS247™-3 TO-247-3

IXTX90N25L2 IXYS MOSFET N-CH 250V 90A PLUS247 125 - Immediate
AU$39.65000 1 Tube? Linear L2™ Active N-Channel MOSFET (Metal Oxide) 250V 90A (Tc) 10V 4.5V @ 3mA 640nC @ 10V ±20V 23000pF @ 25V
960W (Tc) 33 mOhm @ 45A, 10V -55°C ~ 150°C (TJ) Through Hole PLUS247™-3 TO-247-3

IXFK140N25T IXYS MOSFET N-CH 250V 140A TO264 150 - Immediate
AU$18.04000 1 Tube? GigaMOS™ Active N-Channel MOSFET (Metal Oxide) 250V 140A (Tc) 10V 5V @ 4mA
255nC @ 10V ±20V 19000pF @ 25V
960W (Tc) 17 mOhm @ 60A, 10V -55°C ~ 150°C (TJ) Through Hole TO-264AA (IXFK) TO-264-3, TO-264AA

IXTK90N25L2 IXYS MOSFET N-CH 250V 90A TO-264 26 - Immediate
AU$37.41000 1 Tube? Linear L2™ Active N-Channel MOSFET (Metal Oxide) 250V 90A (Tc) 10V 4.5V @ 3mA 640nC @ 10V ±20V 23000pF @ 25V 960W (Tc) 33 mOhm @ 45A, 10V -55°C ~ 150°C (TJ) Through Hole TO-264 (IXTK) TO-264-3, TO-264AA

IXFK100N65X2 IXYS MOSFET N-CH 650V 100A TO-264 75 - Immediate
AU$20.52000 1 Tube? HiPerFET™ Active N-Channel MOSFET (Metal Oxide) 650V 100A (Tc) 10V 5.5V @ 4mA 180nC @ 10V ±30V 11300pF @ 25V 1040W (Tc) 30 mOhm @ 50A, 10V -55°C ~ 150°C (TJ) Through Hole TO-264 TO-264-3, TO-264AA

IXTX102N65X2 IXYS MOSFET N-CH 650V 102A X2 PLUS247 73 - Immediate AU$21.06000 1 Tube? Active N-Channel MOSFET (Metal Oxide) 650V 102A (Tc) 10V 5V @ 250µA 152nC @ 10V ±30V 10900pF @ 25V 1040W (Tc) 30 mOhm @ 51A, 10V -55°C ~ 150°C (TJ) Through Hole PLUS247™-3 TO-247-3


IXTK110N20L2 IXYS MOSFET N-CH 200V 110A TO-264 61 - Immediate
AU$39.97000 1 Tube? Linear L2™ Active N-Channel MOSFET (Metal Oxide) 200V 110A (Tc) 10V 4.5V @ 3mA
500nC @ 10V ±20V 23000pF @ 25V 960W (Tc) 24 mOhm @ 55A, 10V -55°C ~ 150°C (TJ) Through Hole TO-264 (IXTK) TO-264-3, TO-264AA

IXFX120N30T IXYS MOSFET N-CH 300V 120A PLUS247 30 - Immediate
AU$17.81000 1 Tube? GigaMOS™ Active N-Channel MOSFET (Metal Oxide) 300V 120A (Tc) 10V 5V @ 4mA 265nC @ 10V ±20V 20000pF @ 25V 960W (Tc) 24 mOhm @ 60A, 10V -55°C ~ 150°C (TJ) Through Hole PLUS247™-3 TO-247-3

IXTK102N65X2 IXYS MOSFET N-CH 650V 102A X2 TO-264 64 - Immediate
AU$21.42000 1 Tube? Active N-Channel MOSFET (Metal Oxide) 650V 102A (Tc) 10V 5V @ 250µA 152nC @ 10V ±30V 10900pF @ 25V 1040W (Tc) 30 mOhm @ 51A, 10V -55°C ~ 150°C (TJ) Through Hole TO-264 (IXTK) TO-264-3, TO-264AA

IXFH94N30P3 IXYS MOSFET N-CH 300V 94A TO-247 38 - Immediate
AU$15.16000 1 Tube? HiPerFET™, Polar3™ Active N-Channel MOSFET (Metal Oxide) 300V 94A (Tc) 10V 5V @ 4mA 102nC @ 10V±20V 5510pF @ 25V 1040W (Tc) 36 mOhm @ 47A, 10V -55°C ~ 150°C (TJ) Through Hole TO-247 TO-247-3

IXTX110N20L2 IXYS MOSFET N-CH 200V 110A PLUS247 13 - Immediate
AU$39.65000 1 Tube? Linear L2™ Active N-Channel MOSFET (Metal Oxide) 200V 110A (Tc) 10V 4.5V @ 3mA
500nC @ 10V±20V 23000pF @ 25V 960W (Tc) 24 mOhm @ 55A, 10V -55°C ~ 150°C (TJ) Through Hole PLUS247™-3 TO-247-3

IXFK140N30P IXYS MOSFET N-CH 300V 140A TO-264 0
AU$25.04000 1 Tube? HiPerFET™, PolarP2™ Active N-Channel MOSFET (Metal Oxide) 300V 140A (Tc) 10V 5V @ 8mA 185nC @ 10V±20V 14800pF @ 25V 1040W (Tc) 24 mOhm @ 70A, 10V -55°C ~ 150°C (TJ) Through Hole TO-264AA (IXFK) TO-264-3, TO-264AA

IXFX140N25T IXYS MOSFET N-CH 250V 140A PLUS247 0 Standard Lead Time 24 Weeks AU$14.59800 30 Non-Stock? Tube? GigaMOS™ Active N-Channel MOSFET (Metal Oxide) 250V 140A (Tc) 10V 5V @ 4mA 255nC @ 10V±20V 19000pF @ 25V 960W (Tc) 17 mOhm @ 60A, 10V -55°C ~ 150°C (TJ) Through Hole PLUS247™-3 TO-247-3

IXFK120N30T IXYS MOSFET N-CH 300V 120A TO-264 0
Standard Lead Time 26 Weeks AU$14.78920 25 Non-Stock
? Tube? GigaMOS™ Active N-Channel MOSFET (Metal Oxide) 300V 120A (Tc) 10V 5V @ 4mA 265nC @ 10V ±20V 20000pF @ 25V 960W (Tc) 24 mOhm @ 60A, 10V -55°C ~ 150°C (TJ) Through Hole TO-264AA (IXFK) TO-264-3, TO-264AA
 
Last edited:
Disabled Account
Joined 2015
xrk971, did you ask yourself the question, why PA amplifiers don't use overclocker's heat sink? Ok ,you are replacing the air filters time to time ,changing the fans every 3or 4 years (yearly on sea cost) . But the layer of grease, accumulated on the foils by kitchen vapor and air pollution, how do you clean it up ? At the gas station?
 
Founder of XSA-Labs
Joined 2012
Paid Member
xrk971, did you ask yourself the question, why PA amplifiers don't use overclocker's heat sink? Ok ,you are replacing the air filters time to time ,changing the fans every 3or 4 years (yearly on sea cost) . But the layer of grease, accumulated on the foils by kitchen vapor and air pollution, how do you clean it up ? At the gas station?

In fact, PA amps use tunnel heatsinks and large fans and air filters much like a rack mounted sever PC. Have you seen the inside of a Pro Class AB?

Here is Behringer EP4000:
An externally hosted image should be here but it was not working when we last tested it.


And Mackie EM4000:
mac_mseries_amp_open.jpg


They don’t use cpu heatpipes because they spread the load over many MOSFETs. But a CPU cooler would be no different than cleaning a PC case once a year assuming you leave it on all the time. Which we don’t. Class AB amps that draw 400w are o Lt turned on a few hours at a one during listening sessions to save electricity.