I have started building an experimental build of an F5 variant, not sure whether to call it an enhanced F5 or a light version of F5T. The board is based on F5T V2, but I am only putting in 1 pair of output device + 1 MUR diode, also running a rail voltage of 26V (20V-0-20V Trafo).
The thing is I cant get the voltage across the 1R source resistor over 25mV. Could this be due to the low rail voltage? After reading through the sea of threads, I gather that I could do one or all of the following depending on my luck:
1. Increase R5 & R6 to 2.2k
2. Check whether the JFETs are good, if not change them.
Any other things I could try?
The thing is I cant get the voltage across the 1R source resistor over 25mV. Could this be due to the low rail voltage? After reading through the sea of threads, I gather that I could do one or all of the following depending on my luck:
1. Increase R5 & R6 to 2.2k
2. Check whether the JFETs are good, if not change them.
Any other things I could try?
You need to make sure of the polarity of each diode sharing per MOSFET pair. N vs P orientation.
To set the proper bias, you need to recalculate also the equivalent resistance using a single diode as compare with the parallel equivalent as mentioned in F5T article.
Cheers.
@abadsantos, thanks for the advice. Calculations would be a bit out of my capabilities, but I am thinking it would be pretty much the same, coz I completely omit all parts meant for the second pair of devices, including R19, R20, R14, D2, Q4 (and of course the other half as well).
@MASantos, the main reason for not having a second pair of devices is because I dont have the setup and enough devices to match, and further more this is a practice build for me to get used to the circuit, coz my actual intention is to build a V3, and I already have the diyaudio store boards already, just waiting for me to feel comfortable starting it.
Anyways, I added a 1k resistor in series of R5 and R6, and managed to bias the amp up to about 400mV across the 1R resistor R17 & 18. I got a some new questions:
1. In the article Papa Pass mentioned that the bias should be set to 0.4V across the 1R resistors if the diodes are installed. Would it be possible to set the bias higher than that?
2. My heatsinks are now warm, the temperature probe on my multimeter reads 45 degrees, so it leaves some room to set the biase higher, if the bias should be kept at 0.4V across the 1R resistor, to get more class A power, I would need to have more devices in parallel, is that correct?
Appreciate your advices, and sharing the photo I just took of the practice build. Still missing the soft-start and speaker protection boards, and some final tidying of wires to do, but nevertheless, thanks to Mr. Pass for his generosity of allowing us to build the amp and for all the advices and guidance given, so here it is.....
https://dl.dropboxusercontent.com/u/98416047/IMG_20130623_144427.jpg
Best regards
Liu
@MASantos, the main reason for not having a second pair of devices is because I dont have the setup and enough devices to match, and further more this is a practice build for me to get used to the circuit, coz my actual intention is to build a V3, and I already have the diyaudio store boards already, just waiting for me to feel comfortable starting it.
Anyways, I added a 1k resistor in series of R5 and R6, and managed to bias the amp up to about 400mV across the 1R resistor R17 & 18. I got a some new questions:
1. In the article Papa Pass mentioned that the bias should be set to 0.4V across the 1R resistors if the diodes are installed. Would it be possible to set the bias higher than that?
2. My heatsinks are now warm, the temperature probe on my multimeter reads 45 degrees, so it leaves some room to set the biase higher, if the bias should be kept at 0.4V across the 1R resistor, to get more class A power, I would need to have more devices in parallel, is that correct?
Appreciate your advices, and sharing the photo I just took of the practice build. Still missing the soft-start and speaker protection boards, and some final tidying of wires to do, but nevertheless, thanks to Mr. Pass for his generosity of allowing us to build the amp and for all the advices and guidance given, so here it is.....
https://dl.dropboxusercontent.com/u/98416047/IMG_20130623_144427.jpg
Best regards
Liu
In the absence of datasheet that will confirm the equivalent resistance of your diode, I suspect you are still running low as compare with real MUR. Compared to your heatsink, as mine is more larger and massive, I was forced to retreat the vias to 0.35V across the 1R resistor (the highest I can get is 0.37V because of too much heat) and now summer is coming, I even lowered it 0.30V.
As Papa mentioned, 0.4V belongs only to FAB
As Papa mentioned, 0.4V belongs only to FAB
Thanks for the hint, I found the datasheet with the Forward Current vs Forward Voltage curve. The diode starts conducting around 0.45V, so I guess I could manage a little higher than 0.4V (Please correct me if this is wrong, so I dont burn the devices in the attempt of going on higher bias).
Heat is not a problem for me, coz I am only running at around 27V rails. As of now the amp is running at 0.4V bias (phew! I was not paying enough attention to the article, if I was using the real MUR diodes I might have been in trouble already!) and the heatsinks are only 45 degrees, and I live in Malaysia, where the weather doesnt help cooling the sinks down.
Heat is not a problem for me, coz I am only running at around 27V rails. As of now the amp is running at 0.4V bias (phew! I was not paying enough attention to the article, if I was using the real MUR diodes I might have been in trouble already!) and the heatsinks are only 45 degrees, and I live in Malaysia, where the weather doesnt help cooling the sinks down.
Clui is an enlightened Member.
Yes, understand what the diode brings to the amplifier.
I recommend a much lower Rs value than 1r0.
Maybe try a second (or third) 1r0 in parallel, or simply change to 0r3.
400mV across 0r3 is about 1.3A of bias, about the same as a standard F5.
This allows ~2.6A of ClassA output. The diodes will be turned on for some of that ClassA output, on every cycle that exceeds ~1.5Apk
I am truly enlightened by all your good advice and insights! Would have been stuck with only 10W of class A power otherwise, never thought of changing the source resistors.....and I would make sure I get external sinks for the diodes in my V3 build so that I wont have to worry about the varying temperature on the main sinks affecting the turn on point of the diodes....Thanks a million!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.