Sidebar Sidebar
  • You can now help support WorldwideDX when you shop on Amazon at no additional cost to you! Simply follow this Shop on Amazon link first and a portion of any purchase is sent to WorldwideDX to help with site costs.
  • Click here to find out how to win free radios from Retevis!

Galaxy 98VHP Alignment Question For The Experts


The MFG would like to keep the power setting no more than 200w to insure the MOSFET life. You can get more power by increasing the bias but that also makes them heat up quicker and SNAP. Same for AM mode...set the output as per the service manual. You don't want your radio "swinging" over 200w if you want it to last long. When I got my radio from Bells, the bias was set too high. It was doing 250-275 on modulation peaks on SSB. The radio would get extremely hot for low period key downs...and eventually went SNAP crackle and POP. Setting that bias correctly is the most important thing for a happy camper :)
 
  • Like
Reactions: Shadetree Mechanic
Interesting. I recently bought a new dx 959 blue version. First thing I did was pop the cover and did some measurements. Dead key was 2 watts. Driver and final bias were both set at 4.20v exactly. It had the 520 mosfets. For shits and giggles, I set the bias to what the older finals called for on the CBS tricks page. With the lower voltages, I couldn't even get 4 watts with the pot turned up all the way. Evidently, those mosfets need 4.20 volts to be biased properly and for full output.
 
  • Like
Reactions: Shadetree Mechanic
I just looked at the 98vhp alignment page on CB tricks. It says the driver and finals are irf520s. Its calling for 3.75v volts on the driver, and 3.50v on the finals. 4.0v on the amp board. I wish I had a brand new untouched to see what the factory settings are. My 2970n2 is basically the same setup.
 
They don't give you a way to break the power connection to the amplifier board and insert a current meter.

We adopted the practice of just putting a 20-Amp meter in line with the radio's positive power lead, keying it on sideband with the mike gain at zero, and setting the two bias trimpots for minimum reading. This would be zero bias on the amplifier transistors.

Each trimpot now gets advanced to show 100 mA increase in the reading. Idle current of 25 mA per transistor may sound low, but it improved reliability without reducing power very much.

Setting gate bias on MOSFETs with a voltmeter on the bias voltage is risky business.

The actual current draw of the transistors is what you should be measuring, even when they make it tough to do.

73
 
  • Like
Reactions: Unit 75, loosecannon, Shadetree Mechanic and 1 other person
nomadradio said:
Setting gate bias on MOSFETs with a voltmeter on the bias voltage is risky business.

The actual current draw of the transistors is what you should be measuring, even when they make it tough to do.
Click to expand...
Interesting. To tell you the truth....I just measured gate voltage. Actually measuring the current makes more sense as all MOSFETS are not created equal. Thank you for that tip.
-75
 
I like to run my IRF520s at ~50mA per device, but it depends upon the application. In the 4400 Turbo amp (has half of the 98VHP amp) ~50mA gave me the best linearity and the cleanest signal out of the amplifier with the stock 520s. They weren't matched at the factory and I gained a couple db lower IMD3 after I put 4 matched ones in it. I'm super picky about my emission quality.

I never bias Mosfets by gate voltage, it's a useless measurement unless you've already previously tested the turn-on characteristics of that particular FET or batch of FETs. The knock-off copies of the IRF520s vary even more wildly than the legitimate ones, and even they can be all over the place. As for your original question for why the FSM said to limit the output to 200W, I've only ever seen those amp sections do about 150-180W while maintaining decent linearity with the stock FETs. This is using a 4:1 AM envelope or a two-tone SSB test signal from a clean bench source to eliminate the radio as a factor. Remember also, if the radio is putting out a dirty signal, it's not going to be any better leaving the amp.
73s
 
So on a dual final mosfet radio for example I can just put my dmm to the amps setting and adjust bias for lowest draw on each mosfet? This is the better way to do it? Thanks
 
Mk121 said:
So on a dual final mosfet radio for example I can just put my dmm to the amps setting and adjust bias for lowest draw on each mosfet? This is the better way to do it? Thanks
Click to expand...

Not exactly. The older bipolar radios had what we called a mirror board in front of the finals. This board connected the regulator's output (TP9) to the driver (TP8) and final stages (TP7). To set the biasing on those radios, you simply pulled the mirror board out (it was a pressure fit) and connected the ammeter across TP9 + TP8 to measure the driver current and across TP9 + TP7 to measure the finals. This would still be the approach to setting the bias on the newer Mosfet radios, except that every EPT3600##x board that I've seen since the switch to Mosfets has ditched the mirror board and hardwired those 3 points together. You'd need to desolder the wire that jumps these points together and temporarily connect them through the ammeter, then replace the wire after you had the bias pots in the right place. You also cannot have any RF going through the TX chain while setting the bias via the current method either. In a radio with SSB modes this is no problem because you can use either sideband mode with 0 mic gain since there is no carrier. In an AM only radio you would need to remove the capacitor that connects the pre-driver to the gate of the driver Mosfet to stop any RF from entering the driver + final stages. Also, you don't bias for, "lowest draw," you would bias for linearity vs heat output as it's usually a trade-off above a certain point. In a radio with SSB modes, this means Class-AB for the driver and finals. With an AM only radio some things can be done to allow for a lower bias current on the finals which can reduce the heat output in the Mosfets and the regulator slightly.

Other RCI radios that use the 6900##x boards, like the 98VHP, it looks like they still have the mirror boards.

As for better or not? It's a much more accurate way to set the class of operation for each stage, if that's what you're after.

73s
 
  • Like
Reactions: Shadetree Mechanic, Unit 75, sp5it and 1 other person
Thanks for the detailed response blasphemy000. So let's say I want to bias my dual final radio for class ab operation. According to what I read that means the transistor will be conducting between 180 and 360 degrees of a cycle. How do you go about doing that? How to see it on the oscilloscope? Thanks
 
If the bias current is greater than 0mA it's technically Class-AB, around 50mA is a typical setting, although sometimes a bit more is required.
You can't really see the conduction angle on the scope due to LC part of the circuit providing the rest of the wave while the Mosfet is off. You can do a two-tone test in SSB mode to get a view of linearity after setting the bias. This can be viewed on the scope. Using your scope in XY mode, regulator voltage on the X axis and a sample of the RF output on the Y axis can create a trapezoid test for audio vs RF envelope linearity in AM mode. A spectrum analyzer is the best way to view how clean your output is, but not everybody has access to one.
73s
 
  • Like
Reactions: Mk121, loosecannon, Shadetree Mechanic and 1 other person

dxChat
Help Users
  • No one is chatting at the moment.


      Top Bottom
      Back