I am trying to save space and have attached the amp board to a heatsink, so no more flying leads except short wires to attach the BD139 to the NMfet.
Since the FH9 is not a class A heater, I am thinking of attaching the All Cees psu to the heat sink also. It would be mounted with adequate standoff from the heat sink panel and in close proximity to the amp board.
Any problems doing this?
Thanks,
MM
Since the FH9 is not a class A heater, I am thinking of attaching the All Cees psu to the heat sink also. It would be mounted with adequate standoff from the heat sink panel and in close proximity to the amp board.
Any problems doing this?
Thanks,
MM
Hi FH9HVX builders,
Here is one situation I need some clarity on. I have one channel that operates flawless on a power up test. This channel has 8mV offset and with 32mv on 0.22 ohm resistors, I get a nice 150mA.
Since I am rebuilding my layout, I have made a mistake on the other channel. I thought I had the 2 IRFP transistors and the BD139 soldered in place to the pcb, well I did not , I had them fastened to the heatsink only and powered up the channel. As soon as I started the power up with the Variac the voltages went crazy, mucho mV. I shut it down and realized my mistake.
So I soldered the transistors in place and tried power up again. No success, I can hardly increase the voltage without high offset and high bias voltage. So I think I damaged the transistors and possible other components.
Questions:
1. Can I power up the damaged amp board without the 2 main IFRP and BD139 transistors (since I cannot power up with them anyways) and compare the voltage values taken at various points between the damaged amp board and my other fully functional amp board in an effort to trouble shoot damage done ?
2. Since I have some problems already with 1 channel, I would like to take this opportunity to change the IRFP transistors to the more robust FQA type (I have) as I have no more spare IRFP's. Do you recommend just swapping the transistors on the bad amp board and seeing what happens on power up, or, doing a bit of trouble shooting as in Question 1 on the bad amp board before powering up with the new FQA type. I can completely desolder components and test if necessary.
Sorry for the long and maybe somewhat confusing post.
MM
Here is one situation I need some clarity on. I have one channel that operates flawless on a power up test. This channel has 8mV offset and with 32mv on 0.22 ohm resistors, I get a nice 150mA.
Since I am rebuilding my layout, I have made a mistake on the other channel. I thought I had the 2 IRFP transistors and the BD139 soldered in place to the pcb, well I did not , I had them fastened to the heatsink only and powered up the channel. As soon as I started the power up with the Variac the voltages went crazy, mucho mV. I shut it down and realized my mistake.
So I soldered the transistors in place and tried power up again. No success, I can hardly increase the voltage without high offset and high bias voltage. So I think I damaged the transistors and possible other components.
Questions:
1. Can I power up the damaged amp board without the 2 main IFRP and BD139 transistors (since I cannot power up with them anyways) and compare the voltage values taken at various points between the damaged amp board and my other fully functional amp board in an effort to trouble shoot damage done ?
2. Since I have some problems already with 1 channel, I would like to take this opportunity to change the IRFP transistors to the more robust FQA type (I have) as I have no more spare IRFP's. Do you recommend just swapping the transistors on the bad amp board and seeing what happens on power up, or, doing a bit of trouble shooting as in Question 1 on the bad amp board before powering up with the new FQA type. I can completely desolder components and test if necessary.
Sorry for the long and maybe somewhat confusing post.
MM
Before putting transistors in the "bad" channel - copare measurements from the "good" channel for resistor, diode and other transistors values. If that all checks out with no power applied, I would feel fairly safe putting in new transistors (if it was me - I would switch the good channel IRFs for the FQA - and use those IRFs to "test" the bad channel, much easier to replace IRFs right now if smoke escapes).
Most important in this whole process - start up the "bad" channel by itself, with a DBT using a 40W, then 60w, I even use a 150w bulb as a close to not current limiting final DBT test. Passing all these - you should be good to go full wall voltage and set you bias. Then test the entire amp together.
Most important in this whole process - start up the "bad" channel by itself, with a DBT using a 40W, then 60w, I even use a 150w bulb as a close to not current limiting final DBT test. Passing all these - you should be good to go full wall voltage and set you bias. Then test the entire amp together.
You can run the amp safely without the output MOSFETs and BD139 temp comp sensor. It’s basically using it as a preamp and I have actually done that before. Works quite well for being able to drive some high output voltages. Remove the MOSFETs and check your DC setpoints on the transistors in the input LTP and VAS stages and compare to the good channel.
When you power up an amp and do not have the BD139 temp comp sensor connected and securely in place, if there is current going through the MOSFETs, they can fry in an instant from thermal runaway. Same can happen if the MOSFETs are not clamped to a heatsink properly. My guess is that everything up to the MOSFETs and BD139 is still good.
Hope you find the issue. This amp should be quite easy to start. It’s important to be ready to shut off power on first start up if you measure more that a few volts DC offset or more than 500mA bias current. Use two DMMs connected to monitor bias current (across one of the emitter resistors) annd l another to monitor offset when starting up. The moment things look wonky, turn off power.
When you power up an amp and do not have the BD139 temp comp sensor connected and securely in place, if there is current going through the MOSFETs, they can fry in an instant from thermal runaway. Same can happen if the MOSFETs are not clamped to a heatsink properly. My guess is that everything up to the MOSFETs and BD139 is still good.
Hope you find the issue. This amp should be quite easy to start. It’s important to be ready to shut off power on first start up if you measure more that a few volts DC offset or more than 500mA bias current. Use two DMMs connected to monitor bias current (across one of the emitter resistors) annd l another to monitor offset when starting up. The moment things look wonky, turn off power.
Thank you bullittstang and X for the information and advice. Since this was a working stereo amp (with hum) before I decided to change the hook up design and layout it gives me confidence that it can be repaired. The overall plan is to remove the IRFP's and BD139's from both channels and replace with new FQA's and BD139's. While I have the transistors removed I will compare voltage at points on the good channel and compare to the damaged channel. Once all is checked out and repaired if necessary, I will solder in the new FQA's and BD139's and power up each channel separately, keeping my finger at the Variac shutoff in case anything goes wrong. I will set the bias and check the offset on each channel to ensure they are ready for installation in the chassis.
If I am forgetting anything or there is a glaring error in my plan, please let me know.
MM
If I am forgetting anything or there is a glaring error in my plan, please let me know.
MM
I did some voltage testing on the good and the bad pcb and the readings are fairly close to each other, so I think I can continue and construct with the new transistors. The image below shows the voltage from the good pcb near the top of component and the voltage from the bad pcb near the bottom of component. All measurements are DCV and referenced to ground. Comments and thumbs up to continue appreciated.
Thanks X, I thought the measurements were good also. One quick questions:
1. I am fresh out of BD139 with the plastic body. I have BD139G in my inventory, they are the ones with metal backs. These should work I assume, mounted on insulators between the MFets ?
Ready to mount transistors and test again.
MM
1. I am fresh out of BD139 with the plastic body. I have BD139G in my inventory, they are the ones with metal backs. These should work I assume, mounted on insulators between the MFets ?
Ready to mount transistors and test again.
MM
I used BD139G on my build - so you are good, using the insulator as you mentioned.
I would also say - you got lucky that you only lost the O/Ps and the BD139 - the amplifier gods were smiling on you that day.
I would also say - you got lucky that you only lost the O/Ps and the BD139 - the amplifier gods were smiling on you that day.
Yes, I believe you are correct. I praise the amplifier gods and it is nice to get a break every now and then. I shall continue on my quest 😉
Looking through the premium BOM it shows these Molex contacts https://www.mouser.co.uk/ProductDetail/538-39-00-0039 does anyone have a good crimping tool suggestion that doesn't cost $300?
Also what's the recommended alternative to MPSA92-AP which are obsolete
Also what's the recommended alternative to MPSA92-AP which are obsolete
There are basic crimpers for $20 that do the job. I have so many it’s hard to say which one. I’ll have to go to the lab and get part number but look for “Molex crimping tool” online. But not actual Molex brand.
This video is a good explanation but I actually use the gull wing crimp for the wire insulation as well. Just larger. This allows the crimp tabs to reflex and bite into the insulation to hold it from slipping versus pressure alone.
Here is closeup of my crimp job on 18ga silicone wire and Molex MiniFit Jr pins
The one I have doesn’t seem to be sold anymore. I know many folks use the “Engineer” brand crimper which is 2 steps - you crimp each one at a time wire then insulation where these are stepped and do both at once.
Look for “open barrel non insulated connectors crimper” for example this should work.
https://a.co/d/0UYEGOQ
Or this
https://a.co/d/eyyDzTp
Regular MPSA-92 will work. Thousands in stock. 300v PNP circa 40 Hfe BJT with 500mW dissipation.
https://www.mouser.com/ProductDetail/Diotec-Semiconductor/MPSA92?qs=OlC7AqGiEDm8RQC/IPFdxg==
This video is a good explanation but I actually use the gull wing crimp for the wire insulation as well. Just larger. This allows the crimp tabs to reflex and bite into the insulation to hold it from slipping versus pressure alone.
Here is closeup of my crimp job on 18ga silicone wire and Molex MiniFit Jr pins
The one I have doesn’t seem to be sold anymore. I know many folks use the “Engineer” brand crimper which is 2 steps - you crimp each one at a time wire then insulation where these are stepped and do both at once.
Look for “open barrel non insulated connectors crimper” for example this should work.
https://a.co/d/0UYEGOQ
Or this
https://a.co/d/eyyDzTp
Regular MPSA-92 will work. Thousands in stock. 300v PNP circa 40 Hfe BJT with 500mW dissipation.
https://www.mouser.com/ProductDetail/Diotec-Semiconductor/MPSA92?qs=OlC7AqGiEDm8RQC/IPFdxg==
Last edited:
Hello Everyone,
After changing from the flying leads set up and changing the IRFP's to the FQP Fairchild's, I have successfully powered the system and set the preliminary bias and voltage under no load conditions. I took the bias up to 180mA and let it cook for awhile to see if their was any bad effects. Heatsinks and transistor's warmed up nicely. DC offset was about 10mv on each channel.
I then hooked up the inputs and the binding posts. I used my chromebook / Amazon music on battery as my source and powered up into my test speakers. Well, I was greeted with some wonderful sound. I did not play very long, just long enough to get an idea that it works OK. I am happy. I have attached a pic of the test assembly.
I have to wait on my chassis to be finished (probably the new year), so I will put this aside and work on my Xmas amp, my Quasi amp, My AL pre, my 2 Yarra pre's. Lot's of fun ahead.
Thanks to all that have helped me along the way. Once the chassis arrives I will tackle the grounding.
Regards,
MM
After changing from the flying leads set up and changing the IRFP's to the FQP Fairchild's, I have successfully powered the system and set the preliminary bias and voltage under no load conditions. I took the bias up to 180mA and let it cook for awhile to see if their was any bad effects. Heatsinks and transistor's warmed up nicely. DC offset was about 10mv on each channel.
I then hooked up the inputs and the binding posts. I used my chromebook / Amazon music on battery as my source and powered up into my test speakers. Well, I was greeted with some wonderful sound. I did not play very long, just long enough to get an idea that it works OK. I am happy. I have attached a pic of the test assembly.
I have to wait on my chassis to be finished (probably the new year), so I will put this aside and work on my Xmas amp, my Quasi amp, My AL pre, my 2 Yarra pre's. Lot's of fun ahead.
Thanks to all that have helped me along the way. Once the chassis arrives I will tackle the grounding.
Regards,
MM
Going to build mine over Christmas, is there a build guide? If not are there any resistors etc that need to be mounted raised off the board?