A minor change to that BOM that I am aware of:
There are problems with the size of C1...C4, see post #661 and Patricks reply in #664.
I think I will replace the zener with the tl431, for more flexibility. My plan is to trim the regulators to 18 V with the lab psu. Do I need to use a dummy load? In thos case it should be around 5 ohm 100 w. Correct?
Thanks,
Davide
Thanks,
Davide
I would use a dummy load, as Vgs is load dependent.
Value depends on bias, but is just Ohm's Law.
I suppose you know how to calculate power.
Make you you have some reserve in resistor power rating.
I normally use at least factor of 2.
Patrick
Value depends on bias, but is just Ohm's Law.
I suppose you know how to calculate power.
Make you you have some reserve in resistor power rating.
I normally use at least factor of 2.
Patrick
Excelent shots john
also very good build - how does It sound how much Is your bias
samoloko sorry for my late reply.
It’s not ready to listen to yet, I’m awaiting the regulator heatsinks, I’ve biased for 1.5A as a starting point, I may dial it up during colder months 🙂
I'm a bit confused by the load resistor requirement for testing the CRegC power supply. In Fitzfish's build guide (See Post #122 here), he mentions needing a resistor of a few hundred ohms to test the CRegC. I guess I am a bit confused why I can't simply swap in the CRegC in place of my current CRC supply, and connect my big 8ohm load resistor across the + and - Power Supply terminals for testing?
Hi Greg,
Working late (as usual...) and happened to see your question. I've had so little time that you won't see me here much. My apologies.
The CRegC power supplies are mosfet followers. Without some loading their outputs can float up to the input voltage. Adding at least some minimal load as suggested ensures that the output voltage of the CRegC boards are correct when testing them or if you use these as you supplies to test/bias the amplifier.
Dave
Working late (as usual...) and happened to see your question. I've had so little time that you won't see me here much. My apologies.
The CRegC power supplies are mosfet followers. Without some loading their outputs can float up to the input voltage. Adding at least some minimal load as suggested ensures that the output voltage of the CRegC boards are correct when testing them or if you use these as you supplies to test/bias the amplifier.
Dave
Thanks Dave. So in lieu of my big 200W 8ohm resistor, I should use something a bit smaller such as a 10W 300ohm resistor? For testing purposes, can I simply tie this loading resistor across the + and - output terminals of the CRegC power supply (basically swapping it out for the F5X amp board)?
Thanks,
Greg
Thanks,
Greg
Greg,
There is no harm using a 200W 8 ohm resistor.
What Dave suggested is a quick way to test that the regulator circuit functions.
But if you want to check the actual voltage and heat and everything, you need to load it under operating conditions.
And that means you need to draw the correct current.
Say if you are using +/-16V rails 4A, you need a 8 ohm 200W load (resistor).
If you wish to use +/-18V 3A, you need a 12R 160W load.
And yes, just put it across the +Vout and -Vout.
BUT you need to connect the Gnd of the positive and the negative regulators.
They are totoally separate on the regulator board, and are only connected at the F5X star ground.
Patrick
There is no harm using a 200W 8 ohm resistor.
What Dave suggested is a quick way to test that the regulator circuit functions.
But if you want to check the actual voltage and heat and everything, you need to load it under operating conditions.
And that means you need to draw the correct current.
Say if you are using +/-16V rails 4A, you need a 8 ohm 200W load (resistor).
If you wish to use +/-18V 3A, you need a 12R 160W load.
And yes, just put it across the +Vout and -Vout.
BUT you need to connect the Gnd of the positive and the negative regulators.
They are totoally separate on the regulator board, and are only connected at the F5X star ground.
Patrick
Since we didn´t had a dummy load, we did a kind of backward integration that worked quite well.
1) Startup, bias adjust and test the power amp board with a lab power supply.
2) Put the CRegC board with cap boards in between lap power supply and power amp. Now you could fully test the CRegC Board having the amp as real load (which behaves rather like a current source then as a resistor)
3) Add the rectifiers, test (both polarities)
4) Replace the lap power supply by the transformer, test
5) Stepwise integrate the relays of the pretested protection board
Markus
1) Startup, bias adjust and test the power amp board with a lab power supply.
2) Put the CRegC board with cap boards in between lap power supply and power amp. Now you could fully test the CRegC Board having the amp as real load (which behaves rather like a current source then as a resistor)
3) Add the rectifiers, test (both polarities)
4) Replace the lap power supply by the transformer, test
5) Stepwise integrate the relays of the pretested protection board
Markus
I am quite sure the answer to the question is yes, but I prefer to ask:
The spacing of the holes between the diodes for the old and new sinks are the same ?
Thanks,
D.
The spacing of the holes between the diodes for the old and new sinks are the same ?
Thanks,
D.
David,
The shorter measurement between the diodes is the same and still fits the rectifier pcb. However the longer measurement has changed, to save space I’m assuming.
The diodes pins now bend sooner, exactly at the mark they go from thick to thin.
😀
The shorter measurement between the diodes is the same and still fits the rectifier pcb. However the longer measurement has changed, to save space I’m assuming.
The diodes pins now bend sooner, exactly at the mark they go from thick to thin.
😀
Thank John,
I guess it's better to have the new heatsinks to solder the diodes.
I am putting together an order for the last few things I am missing.
Can somebody indicate the part number of the male moles connectors (for the cables) that I will need between the protection board and the rest (regulators etc.)?
Thanks,
Davide
I guess it's better to have the new heatsinks to solder the diodes.
I am putting together an order for the last few things I am missing.
Can somebody indicate the part number of the male moles connectors (for the cables) that I will need between the protection board and the rest (regulators etc.)?
Thanks,
Davide
...
Can somebody indicate the part number of the male moles connectors (for the cables) that I will need between the protection board and the rest (regulators etc.)?
...
Digikey: WM18981-ND
Source: BOM in .pdf-file in post #483
Thanks, I mean the other side, for the cable.
D
If you scroll down the Digikey website for this connection header, you'll find a section called "Mating Products". There's only one mating prodcut, namely:
WM18953-ND - CONN RCPT HOUSING 2MM 4POS SGL
and on the page for this connector housing you'll find appropriate crimp contacts in the "Assciated Products" section.
Last edited:
David,
The shorter measurement between the diodes is the same and still fits the rectifier pcb. However the longer measurement has changed, to save space I’m assuming.
The diodes pins now bend sooner, exactly at the mark they go from thick to thin.
😀
John,
I'm looking at the rectifier PCB and the new rectifier diode hole layout, and I'm not seeing how this still works. Do you have a photo of your rectifier setup with the new heatsinks?
Disregard my post above. I see how they fit now, and I'm not sure why I wasn't seeing it before. 🙂
- Home
- Amplifiers
- Pass Labs
- F5X -- the EUVL Approach - The Build Thread